Artificial Intelligence-Assisted Cancer Status Detection in Radiology Reports

Cancer research is dependent on accurate and relevant information of patient's medical journey. Data in radiology reports are of extreme value but lack consistent structure for direct use in analytics. At Memorial Sloan Kettering Cancer Center (MSKCC), the radiology reports are curated using gold-standard approach of using human annotators. However, the manual process of curating large volume of retrospective data slows the pace of cancer research. Manual curation process is sensitive to volume of reports, number of data elements and nature of reports and demand appropriate skillset. In this work, we explore state of the art methods in artificial intelligence (AI) and implement end-to-end pipeline for fast and accurate annotation of radiology reports. Language models (LM) are trained using curated data by approaching curation as multiclass or multilabel classification problem. The classification tasks are to predict multiple imaging scan sites, presence of cancer and cancer status from the reports. The trained natural language processing (NLP) model classifiers achieve high weighted F1 score and accuracy. We propose and demonstrate the use of these models to assist in the manual curation process which results in higher accuracy and F1 score with lesser time and cost, thus improving efforts of cancer research.

SIGNIFICANCE

Extraction of structured data in radiology for cancer research with manual process is laborious. Using AI for extraction of data elements is achieved using NLP models' assistance is faster and more accurate.

Genomic ancestry in kidney cancer: Correlations with clinical and molecular features

INTRODUCTION

Genetic ancestry (GA) refers to population hereditary patterns that contribute to phenotypic differences seen among race/ethnicity groups, and differences among GA groups may highlight unique biological determinants that add to our understanding of health care disparities.

METHODS

A retrospective review of patients with renal cell carcinoma (RCC) was performed and correlated GA with clinicopathologic, somatic, and germline molecular data. All patients underwent next-generation sequencing of normal and tumor DNA using Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets, and contribution of African (AFR), East Asian (EAS), European (EUR), Native American, and South Asian (SAS) ancestry was inferred through supervised ADMIXTURE. Molecular data was compared across GA groups by Fisher exact test and Kruskal-Wallis test.

RESULTS

In 953 patients with RCC, the GA distribution was: EUR (78%), AFR (4.9%), EAS (2.5%), SAS (2%), Native American (0.2%), and Admixed (12.2%). GA distribution varied by tumor histology and international metastatic RCC database consortium disease risk status (intermediate-poor: EUR 58%, AFR 88%, EAS 74%, and SAS 73%). Pathogenic/likely pathogenic germline variants in cancer-predisposition genes varied (16% EUR, 23% AFR, 8% EAS, and 0% SAS), and most occurred in CHEK2 in EUR (3.1%) and FH in AFR (15.4%). In patients with clear cell RCC, somatic alteration incidence varied with significant enrichment in BAP1 alterations (EUR 17%, AFR 50%, SAS 29%; p = .01). Comparing AFR and EUR groups within The Cancer Genome Atlas, significant differences were identified in angiogenesis and inflammatory pathways.

CONCLUSION

Differences in clinical and molecular data by GA highlight population-specific variations in patients with RCC. Exploration of both genetic and nongenetic variables remains critical to optimize efforts to overcome health-related disparities.

A Phase 2 Trial of Talazoparib and Avelumab in Genomically Defined Metastatic Kidney Cancer

BACKGROUND

Although different kidney cancers represent a heterogeneous group of malignancies, multiple subtypes including Von Hippel-Lindau (VHL)-altered clear cell renal cell carcinoma (ccRCC), fumarate hydratase (FH)- and succinate dehydrogenase (SDH)-deficient renal cell carcinoma (RCC), and renal medullary carcinoma (RMC) are affected by genomic instability. Synthetic lethality with poly ADP-ribose polymerase inhibitors (PARPis) has been suggested in preclinical models of these subtypes, and paired PARPis with immune checkpoint blockade (ICB) may achieve additive and/or synergistic effects in patients with previously treated advanced kidney cancers.

OBJECTIVE

To evaluate combined PARPi + ICB in treatment-refractory metastatic kidney cancer.

DESIGN, SETTING, AND PARTICIPANTS

We conducted a single-center, investigator-initiated phase 2 trial in two genomically selected advanced kidney cancer cohorts: (1) VHL-altered RCC with at least one prior ICB agent and one vascular endothelial growth factor (VEGF) inhibitor, and (2) FH- or SDH-deficient RCC with at least one prior ICB agent or VEGF inhibitor and RMC with at least one prior line of chemotherapy.

INTERVENTION

Patients received talazoparib 1 mg daily plus avelumab 800 mg intravenously every 14 d in 28-d cycles.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

The primary endpoint was objective response rate (ORR) by Immune Response Evaluation Criteria in Solid Tumors at 4 mo, and the secondary endpoints included progression-free survival (PFS), overall survival, and safety.

RESULTS AND LIMITATIONS

Cohort 1 consisted of ten patients with VHL-altered ccRCC. All patients had previously received ICB. The ORR was 0/9 patients; one patient was not evaluable due to missed doses. In this cohort, seven patients achieved stable disease (SD) as the best response. The median PFS was 3.5 mo (95% confidence interval [CI] 1.0, 3.9 mo). Cohort 2 consisted of eight patients; four had FH-deficient RCC, one had SDH-deficient RCC, and three had RMC. In this cohort, six patients had previously received ICB. The ORR was 0/8 patients; two patients achieved SD as the best response and the median PFS was 1.2 mo (95% CI 0.4, 2.9 mo). The most common treatment-related adverse events of all grades were fatigue (61%), anemia (28%), nausea (22%), and headache (22%). There were seven grade 3-4 and no grade 5 events.

CONCLUSIONS

The first clinical study of combination PARPi and ICB therapy in advanced kidney cancer did not show clinical benefit in multiple genomically defined metastatic RCC cohorts or RMC.

PATIENT SUMMARY

We conducted a study to look at the effect of two medications, talazoparib and avelumab, in patients with metastatic kidney cancer who had disease progression on standard treatment. Talazoparib blocks the normal activity of molecules called poly ADP-ribose polymerase, which then prevents tumor cells from repairing themselves and growing, while avelumab helps the immune system recognize and kill cancer cells. We found that the combination of these agents was safe but not effective in specific types of kidney cancer.

IL12/23 Blockade with Ustekinumab as a Treatment for Immune-Related Cutaneous Adverse Events

Background: Immune-related cutaneous adverse events (ircAEs) are frequent and may reduce quality of life and consistent dosing. IL12/23 has been implicated in psoriasis, which is reminiscent of the psoriasiform/lichenoid ircAE phenotype. We report the use of ustekinumab as a therapeutic option. Methods: Patients at Memorial Sloan Kettering Cancer Center, New York, who received immune checkpoint inhibitors and were treated with ustekinumab or had the keywords "ustekinumab" or "Stelara" in their clinical notes between 1 March 2017 and 1 December 2022 were retrospectively identified via a database query. Documentation from initial and follow-up visits was manually reviewed, and response to ustekinumab was categorized into complete cutaneous response (CcR, decrease to CTCAE grade 0), partial cutaneous response (PcR, any decrease in CTCAE grade exclusive of decrease to grade 0), and no cutaneous response (NcR, no change in CTCAE grade or worsening). Labs including complete blood count (CBC), cytokine panels, and IgE were obtained in a subset of patients as standard of care. Skin biopsies were reviewed by a dermatopathologist. Results: Fourteen patients with psoriasiform (85.7%), maculopapular (7.1%), and pyoderma gangrenosum (7.1%) ircAEs were identified. Ten (71.4%) receiving ustekinumab had a positive response to treatment. Among these 10 responders, 4 (40%) demonstrated partial cutaneous response and 6 (60%) demonstrated complete cutaneous resolution. Six patients (42.9%) experienced interruptions to their checkpoint inhibitor treatment as a result of intolerable ircAEs, and following ircAE management with ustekinumab, two (33.3%) were successfully rechallenged with their checkpoint inhibitors. On histopathology, patients primarily had findings of interface or psoriasiform dermatitis. No patients reported an adverse event related to ustekinumab. Conclusions: Ustekinumab showed a benefit in a subset of patients with psoriasiform/lichenoid ircAEs. No safety signals were identified. However, further prospective randomized controlled trials are needed to confirm our findings.

Real-world clinical outcomes of patients with metastatic renal cell carcinoma receiving pembrolizumab + axitinib vs. ipilimumab + nivolumab

BACKGROUND

Immune-Oncology (IO) therapies have changed first-line (1L) treatment paradigm for metastatic renal cell carcinoma (mRCC) in last few years with robust clinical trial data. We examined clinical outcomes among clear cell mRCC (mccRCC) patients who received pembrolizumab + axitinib (pembro-axi) or ipilimumab + nivolumab (ipi-nivo) in the US community oncology setting.

METHODS

This retrospective cohort study utilized data from electronic health records and chart review within The US Oncology Network to identify adult patients with mccRCC initiating 1L pembro-axi or ipi-nivo from January 01, 2019 to December 31, 2020 and followed through March 31, 2021. Physician-recorded response (real-world overall response rate [rwORR] and real-world disease control rate [rwDCR]) was assessed descriptively. Real-world progression-free survival (rwPFS), real-world time to next treatment (rwTTNT) and time on treatment (rwToT) were estimated using Kaplan-Meier analysis. Association of 1L systemic treatment with time-to-event outcomes was examined using multivariable cox proportional hazards models.

RESULTS

Study included 331 mccRCC patients (pembro-axi:44%, ipi-nivo:56%). Median age was 65 years, 75.5% were male, and 82.5% had intermediate/poor (I/P) IMDC risk score. RwORR and rwDCR were 71.0% and 80.0% for pembro-axi and 45.2% and 58.6% for ipi-nivo. In multivariable analysis, pembro-axi was associated with longer rwToT (aHR, 0.53 [95% CI, 0.40, 0.71]), rwTTNT (aHR, 0.60 [95% CI, 0.42, 0.87]), and rwPFS (aHR, 0.70 [95% CI, 0.49, 0.99]) compared to ipi-nivo (P < 0.01).

CONCLUSIONS

Our study provides insight into newer mccRCC treatment tolerability and effectiveness in the real-world US community setting. Our real-world results were comparable to data from clinical trials, which is encouraging for mccRCC patients.

Safety and Activity of Immune Checkpoint Inhibitors in People Living With HIV and Cancer: A Real-World Report From the Cancer Therapy Using Checkpoint Inhibitors in People Living With HIV-International (CATCH-IT) Consortium

PURPOSE

Compared with people living without HIV (PWOH), people living with HIV (PWH) and cancer have traditionally been excluded from immune checkpoint inhibitor (ICI) trials. Furthermore, there is a paucity of real-world data on the use of ICIs in PWH and cancer.

METHODS

This retrospective study included PWH treated with anti-PD-1- or anti-PD-L1-based therapies for advanced cancers. Kaplan-Meier method was used to estimate overall survival (OS) and progression-free survival (PFS). Objective response rates (ORRs) were measured per RECIST 1.1 or other tumor-specific criteria, whenever feasible. Restricted mean survival time (RMST) was used to compare OS and PFS between matched PWH and PWOH with metastatic NSCLC (mNSCLC).

RESULTS

Among 390 PWH, median age was 58 years, 85% (n = 331) were males, 36% (n = 138) were Black; 70% (n = 274) received anti-PD-1/anti-PD-L1 monotherapy. Most common cancers were NSCLC (28%, n = 111), hepatocellular carcinoma ([HCC]; 11%, n = 44), and head and neck squamous cell carcinoma (HNSCC; 10%, n = 39). Seventy percent (152/216) had CD4+ T cell counts ≥200 cells/µL, and 94% (179/190) had HIV viral load <400 copies/mL. Twenty percent (79/390) had any grade immune-related adverse events (irAEs) and 7.7% (30/390) had grade ≥3 irAEs. ORRs were 69% (nonmelanoma skin cancer), 31% (NSCLC), 16% (HCC), and 11% (HNSCC). In the matched mNSCLC cohort (61 PWH v 110 PWOH), 20% (12/61) PWH and 22% (24/110) PWOH had irAEs. Adjusted 42-month RMST difference was -0.06 months (95% CI, -5.49 to 5.37; P = .98) for PFS and 2.23 months (95% CI, -4.02 to 8.48; P = .48) for OS.

CONCLUSION

Among PWH, ICIs demonstrated differential activity across cancer types with no excess toxicity. Safety and activity of ICIs were similar between matched cohorts of PWH and PWOH with mNSCLC.

The impact of immunosuppressive agents on immune checkpoint inhibitor efficacy in patients with advanced melanoma: A real-world, multicenter, retrospective study

BACKGROUND

Immune-related adverse events (irAEs) associated with immune checkpoint inhibitors (ICIs) are often managed via immunosuppressive agents (ISAs); however, their impact on ICI efficacy is not well studied. The impact of the use of ISAs on ICI efficacy in patients with advanced melanoma was therefore investigated.

METHODS

This is a real-world, multicenter, retrospective cohort study of patients with advanced melanoma who received ICIs (n = 370). Overall survival (OS) and time to treatment failure (TTF) from the time of ICI initiation were compared among patients in subgroups of interest by unadjusted and 12-week landmark sensitivity-adjusted analyses. The association of irAEs and their management with OS and TTF were evaluated using univariate and multivariable Cox proportional hazards regression models.

RESULTS

Overall, irAEs of any grade and of grade ≥3 occurred in 57% and 23% of patients, respectively. Thirty-seven percent of patients received steroids, and 3% received other ISAs. Median OS was longest among patients receiving both (not reached [NR]), shorter among those receiving only systemic steroids (SSs) (84.2 months; 95% CI, 40.2 months to NR), and shortest among those who did not experience irAEs (10.3 months; 95% CI, 6-20.1 months) (p < .001). Longer OS was significantly associated with the occurrence of irAEs and the use of SSs with or without ISAs upon multivariable-adjusted analysis (p < .001). Similar results were noted with anti-programmed death 1 (PD-1) monotherapy and combination anti-PD-1 plus anti-cytotoxic T-lymphocyte antigen 4 (CTLA-4) therapy, and with 12-week landmark sensitivity analysis (p = .01).

CONCLUSIONS

These findings in patients with melanoma who were treated with ICIs suggest that the use of SSs or ISAs for the management of irAEs is not associated with inferior disease outcomes, which supports the use of these agents when necessary.

Real-world Treatment Patterns and Clinical Outcomes for Metastatic Renal Cell Carcinoma in the Current Treatment Era

Background

Immuno-oncology (IO) agents and tyrosine kinase inhibitors (TKIs) have revolutionized the treatment paradigm for metastatic renal cell carcinoma (mRCC). Data on real-world usage and outcomes are limited.

Objective

To examine real-world treatment patterns and clinical outcomes for mRCC.

Design setting and participants

This retrospective cohort study included 1538 patients with mRCC who received first-line treatment with pembrolizumab + axitinib (P + A; n = 279, 18%), ipilimumab + nivolumab (I + N; n = 618, 40%), or TKI monotherapy (TKIm; cabozantinib, sunitinib, pazopanib, or axitinib; n = 641, 42%) between January 1, 2018 and September 30, 2020 in US Oncology Network/non-network practices.

Outcome measurements and statistical analysis

The relationship with outcomes, time on treatment (ToT), time to next treatment (TTNT), and overall survival (OS) was analyzed using multivariable Cox proportional-hazards models.

Results and limitations

The median age of the cohort was 67 yr (interquartile range 59.5-74.4), 70% were male, 79% had clear cell RCC, and 87% had an intermediate or poor International mRCC Database Consortium risk score. The median ToT was 13.6 for P + A versus 5.8 for I + N versus 3.4 mo for TKIm (p < 0.001) and the median TTNT was 16.4 for P + A versus 8.3 for I + N versus 8.4 mo for TKIm (p < 0.001) . Median OS was not reached for P + A, 27.6 mo for I + N, and 26.9 mo for TKIm (p = 0.237). On adjusted multivariable analysis, treatment with P + A was associated with better ToT (adjusted hazard ratio [aHR] 0.59, 95% confidence interval [CI] 0.47-0.72 vs I + N; 0.37, 95% CI, 0.30-0.45 vs TKIm; p < 0.0001) and better TTNT (aHR 0.61, 95% CI 0.49-0.77 vs I + N; 0.53, 95% CI 0.42-0.67 vs TKIm; p < 0.0001). Limitations include the retrospective design and the limited follow-up for characterization of survival.

Conclusions

We noted substantial uptake of IO-based therapies in the first-line community oncology setting since their approval. In addition, the study provides insights into clinical effectiveness, tolerability, and/or compliance of IO-based therapies.

Patient summary

We examined the use of immunotherapy for patients with metastatic kidney cancer. The findings suggest rapid implementation of these new treatments by oncologists working in the community setting, which is reassuring for patients with this disease.

Progression-free Survival After Second Line of Therapy for Metastatic Clear Cell Renal Cell Carcinoma in Patients Treated with First-line Immunotherapy Combinations

Immunotherapy (IO)-based combinations used to treat metastatic clear cell renal cell carcinoma (ccRCC) include dual immune checkpoint inhibition with ipilimumab and nivolumab (IO/IO) and several combinations of vascular endothelial growth factor receptor-targeting tyrosine kinase inhibitors (TKIs) with an immune checkpoint inhibitor (TKI/IO). IO/IO and TKI/IO approaches have not been compared directly, and it is unknown whether patients who do not respond to first-line IO/IO can salvage long-term survival by receiving a second-line TKI. Progression-free survival after second-line therapy (PFS-2) evaluates the ability to be salvaged by second-line therapy. We retrospectively evaluated 173 patients treated with first-line IO/IO or TKI/IO for metastatic ccRCC at Memorial Sloan Kettering Cancer Center and report PFS-2, overall survival, and response to second line of therapy (ORR2nd) for groups defined by first-line category. Although ORR2nd was significantly higher with IO/IO than with TKI/IO (47% vs 13%, p < 0.001), there was no significant difference in median PFS-2 for TKI/IO versus IO/IO (44 vs 23 mo, log-rank p = 0.1) or restricted mean survival time (RMST) for PFS-2 when adjusted for propensity score (33 vs 30 mo; difference 2.6 mo [95% confidence interval {CI}: -2.6, 7.9]; p = 0.3). There was also no significant difference in RMST for overall survival when adjusted for propensity score (38 vs 37 mo; group difference 1.0 mo [95% CI: -3.4, 5.5]; p = 0.7). These findings do not support a change in current utilization practices for IO/IO and TKI/IO treatment strategies for ccRCC. PATIENT SUMMARY: In cases of metastatic clear cell renal cell carcinoma, no significant difference was observed in progression-free survival after second line of therapy between patients receiving ipilimumab plus nivolumab and those receiving a combination of a tyrosine kinase inhibitor and an immune checkpoint inhibitor.

Impact of sarcomatoid features on treatment outcomes in metastatic clear cell renal cell carcinoma treated with first-line immunotherapy combinations

687

Background: The presence of sarcomatoid features in clear cell renal cell carcinoma (cRCC) has historically been associated with poor response to tyrosine kinase inhibitor (TKI) monotherapy and poor overall survival; however, immunotherapy (IO) combination therapies have shown more promise in treating this variant. Front-line anti-PD-1 based IO combinations used in ccRCC include ipilimumab/nivolumab (IO/IO) and several combinations of a VEGFR-targeted TKI with a PD-1 inhibitor (TKI/IO). Here, we compare progression-free survival after therapeutic 1st-line (PFS) and 2nd-line (PFS-2) in patients who received IO/IO vs TKI/IO combinations as 1st line treatment for metastatic ccRCC, and test whether the treatment effects differ based on the presence or absence of sarcomatoid dedifferentiation. Methods: A retrospective analysis was performed on patients with ccRCC initiating 1st line combination IO at Memorial Sloan Kettering Cancer Center between 1/1/2014 and 12/30/2020. Patient cohorts were defined by 1st line treatment type: IO/IO or TKI/IO. PFS and PFS-2 were estimated using the Kaplan-Meier method. Restricted mean survival time (RMST) was calculated for PFS and PFS-2 in each 1st line treatment group and modelled using a generalized linear model adjusted for IMDC risk. To test for heterogeneity of treatment effect among subgroups, sarcomatoid features (presence/absence) is included in the models and an interaction test is performed. Results: Ninety patients (28 sarcomatoid) received 1st line IO/IO and 83 (17 sarcomatoid) received 1st line TKI/IO. Median PFS time is 6.8 months (95% CI: 4.5, 12.2) for IO/IO and 21 months (95% CI: 15, 25) for TKI/IO, p=0.009. After adjusting for IMDC risk, and after 48 months of follow-up, RMST for PFS was 10 months for IO/IO and 24 months for TKI/IO (p=0.02) and RMST for PFS-2 was 20 months for IO/IO and 23 months for TKI/IO (p=0.24). In the RMST model, the interaction between treatment group and presence or absence of sarcomatoid features is not significant for PFS (0.95) or PFS-2 (0.29). Conclusions: For ccRCC patients treated with 1st line IO/IO or TKI/IO, adjusted RMST for PFS was significantly longer for the TKI/IO group, but there was no statistically significant difference in adjusted RMST for PFS-2. Anti-PD-1-based therapy is an effective approach to treating ccRCC with sarcomatoid features. [Table: see text]

A phase 1b/2 study of batiraxcept (AVB-S6-500) in combination with cabozantinib in patients with advanced or metastatic clear cell renal cell carcinoma (ccRCC)

666

Background: Activation of the GAS6/AXL pathway promotes tumor cell proliferation; high levels of GAS6 and AXL have been associated with drug resistance and decreased survival. Batiraxcept is an ultra-high affinity decoy protein that captures GAS6, preventing it from activating AXL signaling. Serum soluble AXL (sAXL)/GAS6 ratio correlates with response to batiraxcept in advanced ovarian cancer. Batiraxcept in combination with cabozantinib is being evaluated in patients (pts) with metastatic clear cell renal cell carcinoma (ccRCC) in this study. Prior studies of cabozantinib in similar pt populations have shown an overall response rate (ORR) ranging from 15 – 32.5% and progression-free survival (PFS) from 3.7-7.2 months (mo). Methods: Key eligibility criteria include previously treated ccRCC pts, excluding prior cabozantinib. Objectives were to determine the safety of batiraxcept plus cabozantinib 60 mg, the recommended phase 2 dose (RP2D), ORR, PFS, duration of response. Correlative endpoints included assessment of baseline serum sAXL/GAS6 with ORR and PFS. Results: In Phase 1b (P1b), 26 pts received at least one dose of batiraxcept with cabozantinib (16 pts - 15 mg/kg; 10 pts - 20 mg/kg). Median age was 60 (range 40-81), male 85% (n=22), IMDC intermediate/poor risk 77% (n=20). All pts received prior immuno-oncology (IO) therapy, 54% (n=14) also received VEGF-TKI; 85% (n=22) received 1 to 2 prior lines of therapy (range 1-6). Median follow-up time was 11.6 mo (range 3.7-18). The RP2D is 15 mg/kg. Related adverse events (AE) any grade and grade ≥ 3 were 100% and 39% (n=10), respectively. Cabozantinib dose reductions occurred in 54% (n=14). AEs of special interest were grade 1 or 2 infusion related reactions, 27% (n=7). Batiraxcept was discontinued due to disease progression 62% (n=16) or toxicity 12% (n=3). The table shows P1b efficacy, ORR of 42%, median PFS of 9.3 mo, and DCR of 84%. Twenty (77%) pts had high baseline sAXL/GAS6 levels; ORR was 55% (11/20) with mPFS of 9.3 mo, compared to 0% ORR and mPFS 6.4 mo in the low sAXL/GAS6 pts. Conclusions: Batiraxcept was well tolerated when combined with cabozantinib. No new safety signals were noted. Efficacy for the combination is encouraging compared to historical cabozantinib monotherapy data. Higher GAS6 levels were predictive for treatment response. The P2 accrual is complete and efficacy data are maturing. A P3 trial plans to evaluate the efficacy in ccRCC patients who have progressed on IO-based and VEGF-TKI therapies. Clinical trial information: NCT04300140 . [Table: see text]

Correlating HLA variants and the emergence of immune-related adverse events (irAEs) from ipilimumab/nivolumab (I/N) in patients (pts) treated on CheckMate 214

659

Background: Combination therapy with I/N can achieve long-term benefit in a sizable proportion of RCC pts, but broad application has been challenged by a high incidence of irAEs. Host features affecting antigen presentation are candidate determinants of irAE pathogenesis, and isolated reports have linked Human Leukocyte Antigen (HLA) variations to organ- specific irAEs. We analyzed data for pts treated on a phase 3 trial to investigate association of class 1 HLA variants with development of irAEs from I/N and develop a tool to predict risk for toxicity. Methods: We analyzed germline HLA + clinical data for 472 pts receiving I/N vs sunitinib (SUT). Based on allelic variants for HLA-A and B pts were categorized into 12 established HLA "Super-Types" (ST), sets of HLA variants with largely overlapping peptide binding specificity. We conducted uni- and multivariate (UV; MV) Cox proportional hazards regression to correlate HLA-ST variables with time to treatment-related AEs (TTrAE, grade 2+) in I/N pts using the Kaplan Meier method. Several MV HLA-ST models were developed on a discovery set (DISC; 2/3 random sample of I/N pts without replacement ) and compared using model concordance (c-index) on a validation set (VAL; remaining 1/3). Model coefficients were used to create a “HLA-ST score” that could be computed for individual pts. Results: 235 pts and 237 pts received I/N vs SUT and had available HLA data, respectively. On UV analysis for I/N, HLA-B07 ST had protective association TTrAE (HR= 0.65, 95% CI: 0.46,0.90; p= 0.010), while B62 associated adversely (HR=1.64, 95% CI: 1.12, 2.40; p=0.014). Relevant to the development of our model we identified interactions between several pairs of HLA ST. The HLA ST model with best performance in DISC (c-index= 0.606) and VAL (c-index=0.595) integrates B07, B62, A01, B08, and two interactions: B07-B08 and B07-A01. The model-generated HLA-ST score was significantly associated with TTrAE after adjusting for race, BMI, region, PDL1 status, and MSKCC risk score (DISC p<0.001; VAL p=0.028). No association with TTrAE was seen in SUT treated pts (p=0.655), neither in UV nor MV model. I/N-treated patients with HLA-ST score dichotomized >= vs < median had significantly different TTrAE both in DISC (p=0.004) and VAL (p=0.009); again, no difference was observed in the SUT arm (p=0.5). The weighted HLA-ST score had no association with PFS or OS for I/N pts. Conclusions: In this large sample of I/N-treated patients, class I HLA variants were associated with the risk of developing irAEs. We developed a HLA ST based score that correlated with treatment toxicity independent of relevant clinical and demographic features. No association with TTrAE was seen in SUT-treated pts. These results highlight the potential of characterizing germline features to predict immune related toxicity upfront and deserves further study. Clinical trial information: NCT02231749 .

P01.01.A Lesion-Function Analysis from Multimodal Imaging and Normative Brain Atlases for Prediction of Cognitive Deficits in Glioma Patients

Background

Cognitive deficits are common in glioma patients following multimodality therapy, but the relative impact of different types and locations of treatment-related brain damage and recurrent tumors on cognition is not well understood.

Material and Methods

In 121 WHO Grade III/IV glioma patients, structural MRI, O-(2-[18F]fluoroethyl)-L-tyrosine FET-PET, and neuropsychological testing were performed at a median interval of 14 months (range, 1-214 months) after therapy initiation. Resection cavities, T1-enhancing lesions, T2/FLAIR hyperintensities, and FET-PET positive tumor sites were semiautomatically segmented and elastically registered to a normative, resting state (RS) fMRI-based functional cortical network atlas and to the JHU atlas of white matter (WM) tracts, and their influence on cognitive test scores relative to a cohort of matched healthy subjects was assessed.

Results

T2/FLAIR hyperintensities presumably caused by radiation therapy covered more extensive brain areas than the other lesion types and significantly impaired cognitive performance in many domains when affecting left-hemispheric RS-nodes and WM-tracts as opposed to brain tissue damage caused by resection or recurrent tumors. Verbal episodic memory proved to be especially vulnerable to T2/FLAIR abnormalities affecting the nodes and tracts of the left temporal lobe.

Conclusion

In order to improve radiotherapy planning, publicly available brain atlases, in conjunction with elastic registration techniques, should be used, similar to neuronavigation in neurosurgery.

P01.02.B Case Report: Disruption of Resting-State Networks and Cognitive Deficits After Whole Brain Irradiation for Singular Brain Metastasis

Background

Long-term survivors of whole brain radiation (WBRT) are at significant risk for developing cognitive deficits, but knowledge about the underlying pathophysiological mechanisms is limited. Therefore, we here report a rare case with a singular brain metastasis treated by resection and WBRT that survived for more than 10 years where we investigated the integrity of brain networks using resting-state functional MRI.

Material and Methods

A female patient with a left frontal non-small cell lung cancer (NSCLC) brain metastasis had resection and postoperative WBRT (30.0 in 3.0Gy fractions) and stayed free from brain metastasis recurrence for a follow-up period of 11 years. Structural magnetic resonance imaging (MRI) and amino acid [O-(2-[18F]fluoroethyl)-L-tyrosine] positron emission tomography (FET PET) were repeatedly acquired. At the last follow up, neurocognitive functions and resting-state functional connectivity (RSFC) using resting-state fMRI were assessed. Within-network and inter-network connectivity of seven resting-state networks were computed from a connectivity matrix. All measures were compared to a matched group of 10 female healthy subjects.

Results

At the 11-year follow-up, T2/FLAIR MR images of the patient showed extended regions of hyper-intensities covering mainly the white matter of the bilateral dorsal frontal and parietal lobes while sparing most of the temporal lobes. Compared to the healthy subjects, the patient performed significantly worse in all cognitive domains that included executive functions, attention and processing speed, while verbal working memory, verbal episodic memory, and visual working memory were left mostly unaffected. The connectivity matrix showed a heavily disturbed pattern with a widely distributed, scattered loss of RSFC. The within-network RSFC revealed a significant loss of connectivity within all seven networks where the dorsal attention and fronto-parietal

control networks were affected most severely. The inter-network RSFC was significantly reduced for the visual, somato-motor, and dorsal and ventral attention networks.

Conclusion

As demonstrated here in a patient with a metastatic NSCLC and long-term survival, WBRT may lead to extended white matter damage and cause severe disruption of the RSFC in multiple resting state networks. In consequence, executive functioning which is assumed to depend on the interaction of several networks may be severely impaired following WBRT apart from the well-recognized deficits in memory function.

KS05.5.A Alterations in white matter fiber density associated with structural MRI and metabolic PET lesions following multimodal therapy in glioma patients

Background

In glioma patients, multimodal therapy and recurrent tumor result in local brain tissue changes, characterized by pathologic findings in structural MRI and metabolic PET images. Little is known about these different lesion types’ impact on the local white matter fiber architecture and clinical outcome.

Patients and Methods

This study included data from 121 pretreated patients (median age, 52 years; ECOG, 01) with histomolecularly characterized glioma (WHO grade IV glioblastoma, n=81; WHO grade III anaplastic astrocytoma, n=28; WHO grade III anaplastic oligodendroglioma, n=12), who had a resection, radiotherapy, alkylating chemotherapy, or combinations thereof. After a median time of 14 months (range, 1-214 months), post-therapeutic structural and metabolic findings were evaluated using anatomical MRI and O-(2-[18F]fluoroethyl)-L-tyrosine (FET) PET acquired on a 3T hybrid PET/MR scanner. Local fiber density was estimated from tractography based on highangular resolution diffusion-weighted imaging. A cohort of 121 healthy subjects selected from the 1000BRAINS study and matched for age, gender and education served as a control group.

Results

The median volume of resection cavities, contrast-enhancing regions, regions with pathologically increased FET uptake, and T2/FLAIR hyperintense regions amounted to 20.9, 7.9, 30.3, and 53.4 mL, respectively. Compared to the control group, the average local fiber density in these regions was significantly reduced (p&lt;0.001). Resection cavities showed the highest reduction, followed by contrast-enhancing lesions and metabolically active tumors on FET PET (relative fiber density reduction, -87%, -65%, -55%, respectively). The local fiber density was inversely related (p=0.005) to the FET uptake in recurrent tumors. T2/FLAIR hyperintense lesions, either assigned to peritumoral edema in recurrent glioma or radiation-induced gliosis, had a comparable impact on reducing fiber density (48% and 41%, respectively). The total fiber loss (average fiber loss multiplied by lesion volume) associated with contrast-enhancing lesions (p=0.006) and T2/FLAIR hyperintense lesions (p=0.013) had a significant impact on the general performance status of the patients (ECOG score).

Conclusions

Our results suggest that apart from resection cavities, reduction in local fiber density is greatest in contrast-enhancing recurrent tumors, but total fiber loss induced by edema or gliosis has an equal detrimental effect on the patients’ performance due to the larger volume affected.

Funding

Funded by the 1000BRAINS study (INM, Research Centre Juelich, Germany), Horizon 2020 (Grant No. 945539 (HBP SGA3; SC)), and Heinz Nixdorf Foundation.

The Risk of Opportunistic Infections and the Role of Antibiotic Prophylaxis in Patients on Checkpoint Inhibitors Requiring Steroids

BACKGROUND

Immune-related adverse events (irAEs) often require treatment with high-dose systemic steroids (SS) and other immunosuppressive agents (ISAs). NCCN Guidelines recommend prophylactic antibiotics for Pneumocystis jirovecii pneumonia (PJP) for patients receiving prolonged SS/ISAs. However, there is a paucity of evidence regarding the incidence of opportunistic infections (OIs) and non-OIs and the role of prophylactic antibiotics in patients on SS/ISAs for irAEs.

METHODS

A retrospective analysis was conducted of patients treated using immune checkpoint inhibitor (ICI) therapy at 5 MedStar Health hospitals from January 2011 to April 2018. OIs were defined per the Infectious Diseases Society of America guidelines for the prevention and treatment of OIs in patients with HIV. The study cohort included patients who received ≥20 mg daily of a prednisone equivalent for ≥4 weeks to manage irAEs.

RESULTS

The study cohort identified 112 (15%) of 758 total patients treated using ICIs. Baseline characteristics included the following: median age was 64 years, 74% (n=82) of patients were White, 89% (n=100) had an ECOG performance status ≤1, 61% (n=68) had melanoma, 19% (n=21) had non-small cell lung cancer, 45% (n=50) were treated using an anti-PD-(L)1 ICI, and 33% (n=37) were treated using an anti-PD-1/anti-CTLA-4 combination. The median starting SS dose was 100 mg of a prednisone equivalent, and 25% of patients required additional ISAs, with infliximab (n=15) and mycophenolate mofetil (n=9) being the most common. We found that 20% (n=22) of patients developed any infection, including 7% (n=8) with OIs (oral candidiasis [n=4], nondisseminated varicella zoster infection [n=2], PJP [n=1], and Listeria monocytogenes endophthalmitis [n=1]) and 13% (n=14) with non-OIs (most common: Clostridium difficile and pneumonia [n=5 each]). PJP prophylaxis with sulfamethoxazole/trimethoprim was given to 13% (n=14) patients, of whom 43% (n=6) developed OIs/non-OIs.

CONCLUSIONS

Our study highlights the fundamental issues for patients on ICI therapy who require SS/ISAs for irAEs: the degree of immunosuppression and the relative risk of OI. We noted a low incidence of OIs overall and breakthrough infections despite PJP prophylaxis. We question whether PJP prophylaxis is efficacious or necessary. Prospective trials are required to answer these questions.

Matched Molecular Profiling of Cell-Free DNA and Tumor Tissue in Patients With Advanced Clear Cell Renal Cell Carcinoma

PURPOSE

The clinical utility of cell-free DNA (cfDNA) as a biomarker for advanced clear cell renal cell carcinoma (ccRCC) remains unclear. We evaluated the validity of cfDNA-based genomic profiling in a large cohort of patients with ccRCC with matched next-generation sequencing (NGS) from primary tumor tissues.

MATERIALS AND METHODS

We performed paired NGS of tumor DNA and plasma cfDNA using the MSK-IMPACT platform in 110 patients with metastatic ccRCC. Tissues were profiled for variants and copy number alterations with germline comparison. Manual cross-genotyping between cfDNA and tumor tissue was performed. Deep sequencing with a higher sensitivity platform, MSK-ACCESS, was performed on a subset of cfDNA samples. Clinical data and radiographic tumor volumes were assessed to correlate cfDNA yield with treatment response and disease burden.

RESULTS

Tumor tissue MSK-IMPACT testing identified 582 genomic alterations (GAs) across the cohort. Using standard thresholds for de novo variant calling in cfDNA, only 24 GAs were found by MSK-IMPACT in cfDNA in 7 of 110 patients (6%). With manual cross-genotyping, 210 GAs were detectable below thresholds in 74 patients (67%). Intrapatient concordance with tumor tissue was limited, including VHL (31.6%), PBRM1 (24.1%), and TP53 (52.9%). cfDNA profiling did not identify 3p loss because of low tumor fractions. Tumor volume was associated with cfDNA allele frequency, and VHL concordance was superior for patients with greater disease burden.

CONCLUSION

cfDNA-based NGS profiling yielded low detection rates in this metastatic ccRCC cohort. Concordance with tumor profiling was low, even for truncal mutations such as VHL, and some findings in peripheral blood may represent clonal hematopoiesis. Routine cfDNA panel testing is not supported, and its application in biomarker efforts must account for these limitations.

Contribution of the Skin-Gut Axis to Immune-Related Adverse Events with Multi-System Involvement

Simple Summary

Increasing numbers of cancer patients are treated with immunotherapy that activates their immune systems to control or even eliminate tumors. However, a substantial proportion of patients experience adverse events mediated by the unleashed immune system. The skin is one of the most frequently affected organs, with toxicities typically manifesting as distinct types of rashes. The gastrointestinal (GI) tract is also commonly affected, with a wide spectrum of symptom manifestations that can range from self-limited diarrhea to life-threatening colitis. Here we present the relationship between skin and GI adverse events among cancer patients receiving treatment with immune checkpoint blockade, which has not been well-studied.

Abstract

Immune-related adverse events (irAEs) frequently complicate treatment with immune checkpoint blockade (ICB) targeting CTLA-4, PD-1, and PD-L1, which are commonly used to treat solid and hematologic malignancies. The skin and gastrointestinal (GI) tract are most frequently affected by irAEs. While extensive efforts to further characterize organ-specific adverse events have contributed to the understanding and management of individual toxicities, investigations into the relationship between multi-organ toxicities have been limited. Therefore, we aimed to conduct a characterization of irAEs occurring in both the skin and gut. A retrospective analysis of two cohorts of patients treated with ICB at Memorial Sloan Kettering Cancer Center was conducted, including a cohort of patients with cutaneous irAEs (ircAEs) confirmed by dermatologists (n = 152) and a cohort of patients with biopsy-proven immune-related colitis (n = 246). Among both cohorts, 15% (61/398) of patients developed both skin and GI irAEs, of which 72% (44/61) patients had ircAEs preceding GI irAEs (p = 0.00013). Our study suggests that in the subset of patients who develop both ircAEs and GI irAEs, ircAEs are likely to occur first. Further prospective studies with larger sample sizes are needed to validate our findings, to assess the overall incidence of co-incident irAEs, and to determine whether ircAEs are predictors of other irAEs. This analysis highlights the development of multi-system dermatologic and gastrointestinal irAEs and underscores the importance of oncologists, gastroenterologists, and dermatologists confronted with an ircAE to remain alert for additional irAEs.

A phase 1b/2 study of batiraxcept (AVB-S6-500) in combination with cabozantinib, cabozantinib and nivolumab, and as monotherapy in patients with advanced or metastatic clear cell renal cell carcinoma (NCT04300140)

TPS4599

Background: In clear cell renal cell carcinoma (ccRCC) the constitutive expression of hypoxia induced factor 1-α leads to increased expression of AXL. AXL overexpression has been associated with the development of resistance to VEGF inhibitors and suppression of the innate immune response through inhibition of macrophage-driven inflammation. Batiraxcept is a recombinant fusion protein dimer containing an extracellular region of human AXL combined with the human immunoglobulin G1 heavy chain (Fc), which demonstrates highly potent, specific AXL inhibition. In preclinical studies using the 786-O, M62, and SN12L1 tumors, batiraxcept monotherapy resulted in a significant reduction of tumor growth compared to control. In healthy volunteer and ovarian cancer clinical studies, batiraxcept was well tolerated with no dose-related adverse events, and a maximum tolerated dose was not reached. Therefore, batiraxcept could be tested as either a monotherapy or in combination with standard of care drugs in patients with metastatic ccRCC. The Phase 1b dose-escalation portion of this study evaluated batiraxcept in combination with standard of care cabozantanib in patients who progressed on or after first line therapy. No DLT was observed at either of two batiraxcept doses evaluated. The recommended Phase 2 dose of batiraxcept has been identified as 15 mg/kg every 2 weeks (q2w) with cabozantinib 60 mg based upon safety, PK/PD, and preliminary efficacy data. Methods: This Phase 2, multi-center, open-label study includes three parts: Part A) batiraxcept 15 mg/kg q2w in combination with cabozantinib 60 mg daily for ccRCC subjects who have progressed on or after one line of therapy, n=25. Part B) batiraxcept 15 mg/kg q2w with cabozantinib 40 mg daily and nivolumab at the investigator’s choice (240 mg q2w or 480 mg q4w) for first line treatment of advanced or metastatic ccRCC subjects, n=20. If no safety signals are observed in the first 6 subjects enrolled, 10 subjects will be enrolled in the first stage of a Simon 2-stage minmax statistical design. If ≥ 6/10 subjects achieve PR or CR, stage 2 will open to enroll up to 20 total subjects. Part C) batiraxcept 15 mg/kg q2w monotherapy for subjects with advanced/metastatic ccRCC ineligible for curative intent therapies, n=10. The primary objective for each arm is objective response rate by RECIST v1.1. Secondary objectives include safety, duration of response, clinical benefit rate, progression free survival by RECIST v1.1, and overall survival. Exploratory objectives include pharmacokinetic and pharmacodynamic assessments. The Phase 2 portion of this Ph1b/2 study is currently enrolling. Clinical trial information: NCT04300140.

A phase 1b study (STELLAR-002) of XL092 administered in combination with nivolumab (NIVO) with or without ipilimumab (IPI) or bempegaldesleukin (BEMPEG) in patients (pts) with advanced solid tumors

TPS4600

Background: XL092 is a novel oral inhibitor of receptor tyrosine kinases, including MET, VEGFR2, and TAM kinases (AXL, MER), which are implicated in tumor growth, metastasis, angiogenesis, and immune suppression of the tumor microenvironment. XL092 has a relatively short half-life (̃21h) to support convenient daily dosing and help manage tolerability. Preclinical studies of XL092 with an anti‒PD-1 immune checkpoint inhibitor (ICI) demonstrated antitumor activity in tumor models, and BEMPEG (IL-2 pathway agonist) showed synergy with anti‒PD-L1 and anti‒CTLA-4 agents. This phase 1b trial will evaluate the safety and clinical activity of XL092 alone and in combination with NIVO (anti‒PD-1 mAb) ±IPI (anti‒CTLA-4 mAb) or ±BEMPEG in pts with advanced solid tumors including genitourinary cancers. Presented here is the study design. Methods: This multicenter phase 1b, open-label study (NCT05176483) will enroll pts with unresectable advanced or metastatic solid tumors in dose-escalation and expansion stages. In the dose-escalation stage, ̃36 pts will be enrolled in three XL092 combination therapy cohorts using a rolling 6 design. Cohort A: XL092 (starting dose [SD] 100 mg PO QD) + NIVO (360 mg IV Q3W); Cohort B: XL092 (SD 80 mg PO QD) + NIVO (3 mg/kg IV Q3W × 4, then 480 mg IV Q4W) + IPI (1 mg/kg Q3W × 4); Cohort C: XL092 (SD 100 mg PO QD) + NIVO (360 mg IV Q3W) + BEMPEG (0.006 mg/kg IV Q3W). The primary objective of the dose-escalation stage is to determine the recommended doses of XL092 with the NIVO regimens to be used in the expansion stage. The expansion stage will include cohorts of advanced genitourinary tumors: Cohort 1, clear-cell renal cell carcinoma (ccRCC), no prior systemic therapy; Cohort 2, ccRCC, 1 prior ICI combination regimen; Cohort 3, metastatic castration-resistant prostate cancer (mCRPC), 1 prior novel-hormonal therapy; Cohort 4, urothelial carcinoma (UC), 1 prior platinum-based regimen, ICI-naïve; Cohort 5, UC, ≤2 prior systemic regimens, ICI-experienced; Cohort 6, non-ccRCC, no prior systemic therapy. In each cohort, pts will be randomized to one of the following treatments (based on tumor cohort): single-agent XL092 (Cohorts 2‒6); XL092+NIVO (Cohorts 1‒6); NIVO+IPI (Cohort 1); XL092+NIVO+IPI (Cohorts 1, 3, 6); NIVO+BEMPEG (Cohort 1), XL092+NIVO+BEMPEG (Cohort 1, 2, 4‒6). Thirty pts will be enrolled in each single-agent XL092 arm and 40 pts in each combination therapy arm. Expansion stage objectives are to assess preliminary efficacy, safety, and pharmacokinetics of XL092 alone or in combination in each tumor-specific cohort. Primary efficacy endpoints include objective response rate by investigator per RECIST v1.1 and progression-free survival by blinded independent radiology committee per Prostate Working Group 3 criteria (mCRPC cohort only). The study is currently enrolling pts. Clinical trial information: NCT05176483.

Long-term outcomes of adrenal insufficiency (AI) due to anti–PD(L)-1 immune checkpoint inhibitors (ICI) among patients with cancer

12092

Background: Endocrine immune-related adverse events (irAEs) to anti-PD-(L)1 ICI, including hypophysitis, autoimmune insulin-dependent diabetes mellites (ICI-DM), and primary adrenal insufficiency (AI), are rare but often associated with long term co-morbidities. AI from anti-PD-(L)1 ICI without CTLA-4 blockade is not well characterized in the literature. Long term outcomes for AI including need for replacement steroids and cortisol levels are unknown. Methods: We performed a single center retrospective analysis of patients treated with ICI including anti-programmed cell death protein-1 (Anti-PD-1: nivolumab or pembrolizumab), anti-programmed death-ligand-1 (anti-PD-L1: atezolizumab or cemiplimab) and diagnosed with ICI induced AI by a board-certified endocrinologist from 1/1/2011 to 12/31/2021. Patients treated with anti-CTLA-4 based therapy were not included. Patient baseline characteristics, presenting symptoms at the time of AI diagnosis, and treatment with corticosteroids were obtained by chart review. Baseline labs were collected at the time of AI and during routine patient follow-up. Descriptive statistics are reported for the cohort. Results: Twenty-nine patients were identified, 27 diagnosed with secondary and 2 diagnosed with primary AI. The median age was 63 (range 39-84), 18 (62%) males and 11 (38%) females, 14 (48%) receiving anti-PD(L)-1 monotherapy and 22 receiving anti-PD(L)-1 combination therapy either with chemotherapy (7, 24%), targeted therapy (6, 21%) or other/investigational therapy (2, 7%). The common tumor types were 6 renal cell carcinoma, 4 urothelial carcinoma, 4 melanoma, 15 others. The common presenting symptoms were 26 (90%) with fatigue,16 (55%) weakness, 11 (38%) nausea/vomiting, 4 (14%) headaches and 2 (7%) arthralgias. The common concomitant irAEs were 16 (55%) hypothyroidism, 8 (28%) acute kidney injury, 7 (24%) colitis, 6 (21%) joint pain, 3 (10%) pneumonitis & dermatitis. Eleven (38%) patients were treated with high dose steroids. Median follow-up time for survivors in the cohort was 24 months (range 7-68). Median cortisol level at time of AI diagnosis was 1.1 (IQR 0.9, 2.7; n = 27) and 1.8 (IQR 0.6, 6.4; n = 10) at last follow-up (median lab follow-up 39 months, range 29-58); 7 of 10 patients with available data had cortisol < 5.0 at last follow-up. For secondary AI patients, median ACTH at time of diagnosis was 2.3 (IQR 2.0, 5.0; n = 21) and 2.0 (IQR 2.0, 7.2; n = 7) at last follow-up. Twenty-two of 23 patients with available data continued replacement steroids at last follow-up. Conclusions: AI associated with anti-PD(L)-1 ICI is primarily secondary. Cortisol and ACTH levels remain low even during long-term follow-up. Most patients need long-term replacement steroids. Systemic and comprehensive follow-up for patients who develop AI due to anti-PD(L)-1 ICI is needed to confirm these findings.

A phase 1b/2 study of batiraxcept (AVB-S6-500) in combination with cabozantinib in patients with advanced or metastatic clear cell renal cell (ccRCC) carcinoma who have received front-line treatment (NCT04300140)

4511

Background: AXL is up-regulated by hypoxia-inducible factor-1 signaling in both VHL-deficient and hypoxic tumor cells and plays a critical role in the metastatic phenotype of ccRCC. Batiraxcept is a recombinant fusion protein containing an extracellular region of human AXL combined with the human immunoglobulin G1 heavy chain (Fc), demonstrating highly potent, specific AXL inhibition. Methods: Batiraxcept at doses of 15 and 20 mg/kg, plus cabozantinib 60 mg daily, was evaluated using a 3+3 dose escalation study design. The primary objective was safety; secondary and exploratory objectives included identification of the recommended phase 2 dose (RP2D), overall response rate (ORR), and duration of response (DOR). Correlation of serum soluble AXL (sAXL)/GAS6 with ORR was evaluated. Key eligibility criteria include previously treated (2L+) ccRCC patients; prior treatment with cabozantanib was not allowed. sAXL/GAS6 was evaluated at baseline. Results: Data as of 4-February-2022, Phase 1b enrolled 26 patients, 16 patients treated with 15 mg/kg and 10 patients with 20 mg/kg dose of batiraxcept. Baseline characteristics: median age 60 (40-81); male 22 (85%); median prior line of therapy 1 (1-5); IMDC risk group of favorable 6 (23%); prior VEGF inhibitor 15 (58%); 100% with prior immunotherapy. At median follow up of 4.9 months, 92% (n=24) patients remained on the study. No dose limiting toxicities were observed at either 15 mg/kg or 20 mg/kg dose. Batiraxcept and cabozantinib related adverse events (AEs) occurred in 17 subjects (65%). Most common related AE include decreased appetite 31% (n=8), diarrhea and fatigue 23% (n=6). Grade 3 related AEs occurred in 4 patients (15%) including diarrhea, thromboembolism, hypertension, small bowel obstruction, and thrombocytopenia (n=1, 4% each) being most common. No grade 4 or 5 related AEs were observed. The ORR was 46% (n=12, partial response [PR]; Table). No patients had primary progressive disease. Among the patients who had baseline sAXL/GAS6 ratio of ≥ 2.3, the ORR was 67% (12/18). Regardless of baseline sAXL/GAS6 ratio, 3-month DOR was 100%; and 6-month progression free survival was 79%. Batiraxcept PK levels were similar across both doses and GAS6 levels suppressed through the dosing period. Conclusions: Batiraxcept plus cabozantinib is well tolerated. The RP2D of batiraxcept was identified as 15 mg/kg. Early efficacy signals were observed including 100% DOR at 3 months. Baseline sAXL/GAS6 may serve as a potential biomarker to enrich the population. Clinical trial information: NCT04300140. [Table: see text]

Healthcare resource utilization (HCRU) and costs for patients (pts) with metastatic renal cell carcinoma (mRCC) receiving first-line (LOT1) pembrolizumab plus axitinib (P+A) or ipilimumab plus nivolumab (I+N)

4528

Background: Approval of immuno-oncology (IO) agents have changed treatment paradigm for mRCC pts. While IO-based therapies have demonstrated improved survival, these can be associated with considerable HCRU and costs necessitating their examination in real-world practice. Methods: This retrospective claims analyses utilizing Optum Research Database included adult pts with mRCC diagnosis from July-2017 to Aug-2020 and received P+A or I+N as LOT1 from Jan-2018 to May-2020 (first claim=index date). All eligible pts required continuous enrollment for minimum of 6-months prior and 3-months post the index date unless death occurred. All-cause HCRU counts, and associated costs were examined during the first 90 days (LOT1-90) and entire LOT1 duration and reported as overall and per-pt-per-month (PPPM) estimates. Results: The study identified 507 pts (P+A=126, I+N=381). Average age of the entire cohort was 67 years, 71% were male, mean NCI Charlson score was 2.4, and lung (55%) and bone (33%) were the most common metastatic sites. Mean (SD) distance from mRCC diagnosis to index date was 97 (172) days. Pts with P+A and I+N had similar baseline characteristics. Total % of pts with ambulatory visits was similar for P+A and I+N for LOT1-90 and entire LOT1 (99.2 vs. 100.0%, p=0.082 for both). During LOT1-90, we observed a lower % of pts on P+A with ER visits and inpatient (IP) stay compared to I+N (34 vs. 48, p=0.008; 19 vs. 38, p<0.001, respectively). We also observed a shorter mean (SD) IP stay for P+A vs. I+N during LOT1-90 (1.9 (6.5) vs. 5.6 (13.24) days, p<0.001). Similarly, P+A had lower mean PPPM ambulatory visits, IP stay, and ICU stay during both LOT1-90 and entire LOT1 (Table). In addition, mean PPPM total (medical + pharmacy) and mean PPPM medical costs were lower for P+A compared to I+N, but mean PPPM pharmacy costs were higher for P+A for both LOT1-90 and entire LOT-1 (Table). Conclusions: This study noted significantly higher HCRU with I+N including higher mean PPPM ambulatory visits, IP stays, and ICU stays compared to P+A. Although, P+A had higher mean PPPM pharmacy costs, the total medical plus pharmacy costs were significantly lower compared to I+N. [Table: see text]

Progression-free survival after second line of therapy (PFS-2) for metastatic clear cell renal cell carcinoma (ccRCC) in patients treated with first-line immunotherapy combinations

4536

Background: Front-line therapy with immunotherapy combinations is standard of care for metastatic ccRCC, with ipilimumab/nivolumab (IO/IO) and several combinations of a VEGFR-targeted tyrosine kinase inhibitor with a PD-1 inhibitor (TKI/IO) showing superior efficacy to TKI monotherapy. PFS-2 evaluates the ability to be salvaged by 2nd line therapy and is a surrogate for overall survival (OS). PFS-2 was compared in patients receiving 1st line IO/IO vs TKI/IO for metastatic ccRCC. Methods: A retrospective analysis was performed on patients with ccRCC treated at Memorial Sloan Kettering Cancer Center between 1/1/2014 and 12/30/2020, in cohorts defined by 1st line: IO/IO or TKI/IO. PFS-2 is defined as time from start of 1st line to progression on next therapy, or death. Patients without a PFS-2 event were censored at a prespecified cutoff date. Objective response rate to 1st (ORR1st) and 2nd (ORR2nd) line are compared with the Fisher’s exact test. OS, PFS-2, and time on therapy are estimated with the Kaplan-Meier method and compared with the log-rank test. Results: One hundred seventy-three patients received 1st line IO/IO (N = 90) or 1st line TKI/IO (N = 83); respectively, 52 and 40 patients had a PFS-2 event. 1st line TKI/IO regimens included: 34% axitinib/pembrolizumab, 29% lenvatinib/pembrolizumab, 25% axitinib/avelumab, 11% other. More IO/IO patients had brain metastases and intermediate/poor MSKCC risk category (respectively p = 0.007, p < 0.001). ORR1st and median months on 1st line were higher with TKI/IO vs IO/IO (65% vs 39%, p < 0.001; 16.1 vs 5.1, p < 0.001). ORR2nd was higher with IO/IO vs TKI/IO (47% vs 13%, p < 0.001), and median months on 2nd line was not significantly different (7.7 vs 7.1, p = 0.30). Median PFS-2 for TKI/IO was 44 months (95% CI: 27, 53) vs 23 months (95% CI: 16, 47) for IO/IO, p = 0.13. For TKI/IO and IO/IO groups, respective PFS-2 at 12 months was 86% (95% CI 77, 92) and 74% (95% CI 63, 82); PFS-2 at 36 months was 51% (95% CI 39, 63) and 42% (95% CI 30, 53). OS was not significantly different (p = 0.32; 3 year OS: IO/IO 60%, 95% CI 47, 71; TKI/IO 62%, 95% CI 49, 73). (Table) Conclusions: In patients receiving 1st line IO/IO or TKI/IO, ORR2nd was higher with IO/IO and median PFS-2 was numerically higher with TKI/IO, but no statistically significant difference in PFS-2 or OS was seen. These findings suggest that IO/IO and TKI/O are both acceptable 1st line treatment strategies in ccRCC. [Table: see text]

Phase II Trial of Cabozantinib Plus Nivolumab in Patients With Non-Clear-Cell Renal Cell Carcinoma and Genomic Correlates

PURPOSE

To assess the efficacy and safety of cabozantinib plus nivolumab in a phase II trial in patients with non-clear-cell renal cell carcinoma (RCC).

PATIENTS AND METHODS

Patients had advanced non-clear-cell renal carcinoma who underwent 0-1 prior systemic therapies excluding prior immune checkpoint inhibitors. Patients received cabozantinib 40 mg once daily plus nivolumab 240 mg once every 2 weeks or 480 mg once every 4 weeks. Cohort 1 enrolled patients with papillary, unclassified, or translocation-associated RCC; cohort 2 enrolled patients with chromophobe RCC. The primary end point was objective response rate (ORR) by RECIST 1.1; secondary end points included progression-free survival, overall survival, and safety. Next-generation sequencing results were correlated with response.

RESULTS

A total of 47 patients were treated with a median follow-up of 13.1 months. Objective response rate for cohort 1 (n = 40) was 47.5% (95% CI, 31.5 to 63.9), with median progression-free survival of 12.5 months (95% CI, 6.3 to 16.4) and median overall survival of 28 months (95% CI, 16.3 to not evaluable). In cohort 2 (n = 7), no responses were observed; one patient had stable disease > 1 year. Grade 3/4 treatment-related adverse events were observed in 32% treated patients. Cabozantinib and nivolumab were discontinued because of toxicity in 13% and 17% of patients, respectively. Common mutations included NF2 and FH in cohort 1 and TP53 and PTEN in cohort 2. Objective responses were seen in 10/12 patients with either NF2 or FH mutations.

CONCLUSION

Cabozantinib plus nivolumab showed promising efficacy in most non-clear-cell RCC variants tested in this trial, particularly those with prominent papillary features, whereas treatment effects were limited in chromophobe RCC. Genomic findings in non-clear-cell RCC variants warrant further study as predictors of response.

Real-world assessment of changing treatment patterns and sequence for patients with metastatic renal cell carcinoma (mRCC) in the first-line (1L) setting

302

Background: Several immune-oncology (IO) agents and/or tyrosine kinase inhibitors (TKIs) have received approval for treatment of mRCC in 1L setting by Food and Drug Administration (FDA) over last few years. Limited data exists on evolving real-world treatment patterns and sequence in mRCC patients receiving these agents, especially at the community oncology setting. Methods: We used data from The US Oncology Network of over 1,300 providers from over 480 sites across United States from 01/01/2018 to 12/31/2020 (study period). Eligible study population included mRCC patients who received ipilimumab + nivolumab (Ipi+nivo) (IO+IO); pembrolizumab + axitinib (Pembro+axi) (IO+TKI); and axitinib (Axi) or cabozantinib (Cabo) or pazopanib (Pazo) or sunitinib (Suni) (TKIs monotherapy) in 1L setting until 09/30/2020. Descriptive statistics were used for cohort characterization. Results: We identified 3,756 mRCC patients, of which 1,538 were eligible including 42% (n=641) IO+IO, 18% (n=279) IO+TKI, and 40% (n=618) TKI monotherapy. The median age for the entire cohort was 67.1 years (range 25.0, 93.3), 70% (n=1,076) were male, 70% (n=1,081) were white, 38% (n=587) had BMI ≥ 30 and 79% (n=1,208) had clear cell histology. Among entire cohort, 87% (n=1,338) had intermediate/poor risk score as per International mRCC Database Consortium risk model. We noted a trend towards increased utilization of IO+IO and IO+TKI following their respective FDA approvals (IO+IO: April 2018, IO+TKI: April 2019) (Table). During the study period, overall, 35% (n=535), 12% (n=184), and 4% (n=62) mRCC patients received second-line (2L), third-line (3L) and fourth-line (4L) treatments, respectively. Cabo (49%) and pazo (12%); cabo (51%) and ipi+nivo (23%); and nivo (45%) and ipi+nivo (20%) were the most common 2L treatments in IO+IO, IO+TKI, and TKI monotherapy cohorts, respectively. Conclusions: This large real-world study examined use of new FDA approved mRCC treatments and their impact on treatment paradigm. The results show a rapid adaptation of these newer treatments in the community oncology settings. A longer follow-up is needed to assess their clinical impact and optimal treatment strategy in subsequent setting.[Table: see text]

A phase II study of talazoparib and avelumab in VHL deficient clear cell renal cell carcinoma

347

Background: VHL inactivation may lead to impairments in the DNA damage response pathway and increased replication stress, which consequently can increase the genomic instability of clear cell renal cell carcinoma (RCC). Other molecularly driven RCC subtypes including FH and SDH deficient RCC show impaired DNA repair through oncometabolite accumulation. Synthetic lethality with PARP inhibitors (PARPi) has been suggested in preclinical models, and PARPi may potentiate the effects of immune checkpoint blockade (ICB) therapy. Methods: We conducted a single center, investigator-initiated phase II trial to evaluate combined PARPi + PD-L1 inhibition in two genomically selected advanced RCC cohorts: 1) VHL altered RCC; 2) FH, SDH-RCC and renal medullary carcinoma. For Cohort 1, patients had to have previously been treated with VEGFR TKI and ICB therapy with maximum 3 prior lines of therapy. Patients received talazoparib 1mg daily plus avelumab 800mg IV every 14 days. The primary endpoint was objective response rate (ORR) by iRECIST, and secondary endpoints included progression-free survival (PFS), safety and tolerability. Results from Cohort 1 are reported here. Results: From 2019-2021, 10 advanced RCC patients were enrolled per the first stage of a Simon 2-stage design for cohort 1. The median age was 63 years (R: 42-71), and median number of prior therapies was 2 (R: 1-3). All patients had VHL loss detected by tissue NGS sequencing via MSK-IMPACT, and 3/10 (30%) of patients had co-occurring somatic or germline alterations in DDR specific genes. No objective tumor responses were seen, and the disease control rate was 60% (6/10) with those patients achieving stable disease as best response. All patients experienced disease progression and the median PFS was 3.6 months (95% CI 1.4-4.9). Adverse events were in keeping with reported toxicities for individual agents without emerging safety signals. Conclusions: This is the first clinical study of combination PARPi and ICB therapy in advanced clear cell RCC. In our cohort of VHL deficient TKI/ICB refractory RCC, the study did not meet its pre-defined efficacy threshold to continue enrollment, and our results do not support further study of this regimen in this population. Exploratory efforts to evaluate this treatment approach in other biomarker selected populations of RCC, including cohort 2 of this study, are ongoing. Clinical trial information: NCT04068831.

Real-world clinical outcomes of patients (pts) with metastatic renal cell carcinoma (mRCC) in current immune-oncology (IO) and tyrosine kinase inhibitors (TKIs) era

331

Background: IO agents and TKIs have revolutionized treatment landscape of mRCC pts. Despite robust clinical trials’ data for these agents, real-world (rw) clinical outcomes data, especially from community setting in the US is limited. Methods: This retrospective cohort study included mRCC pts who received 1L treatment with either, pembrolizumab + axitinib (P+A) (IO+TKI), ipilimumab + nivolumab (I+P) (IO+IO) or cabozantinib, sunitinib, pazopanib and axitinib (TKI monotherapy (mono)) between 1/1/2018 and 9/30/2020 from The US Oncology Network of 480 sites. Patients were followed until 12/31/2020 to collect data on rw-time on treatment (rwToT), rw-time to next treatment (rwTTNT) and overall survival (OS). Kaplan-Meier analyses were performed to examine clinical outcomes. Results: We identified 1,538 eligible pts, of which 18% (n = 279) received P+A, 42% (n = 641) I+N and 40% (n = 618) TKI mono. The median follow-up duration for P+A, I+N and TKI mono cohort was 7.2 (range 0.0 - 20.5), 8.5 (range 0.0 - 32.3) and 7.8 (range 0.0 - 35.3) months, respectively. Majority of pts had clear cell histology (P+A - 82%, I+N - 86%, and TKI mono - 72%) and intermediate/poor IMDC risk score (P+A - 87%, I+N - 94%, and TKI mono - 81%). Median OS was not reached for P+A and was similar for I+N and TKI mono cohort (NR, 27.6, and 26.9 months, p=0.237, respectively). The median rwToT (13.6 vs. 5.8 vs. 3.4 months, p<0.0001) and rwTTNT (16.4 vs. 8.3 vs. 8.4 months, p<0.0001) was higher for P+A cohort compared to I+N and TKI mono cohort, respectively. Similar clinical outcomes results were noted for subgroup of pts with intermediate and/or poor IMDC risk score (Table). Conclusions: In this rw US community oncology-based study, longer treatment exposure (rwToT, rwTTNT) was noted in P+A compared to I+N or TKI mono. These rw endpoints may reflect treatment exposure, tolerability and/or compliance.[Table: see text]

Incidence and clinical outcomes of HER2-altered bladder cancer (BC) patients (pts)

556

Background: Bladder cancer has one of the highest rates of human epidermal growth factor receptor 2 (HER2) alteration. Novel HER2-directed agents are being investigated in metastatic BC. We sought to define the incidence and clinical characteristics of HER2-altered BC across disease states. Methods: We retrospectively analyzed our single-institution, clinically annotated cohort of urothelial BC pts with available genomic profiling data (MSK-IMPACT). We quantified the prevalence of HER2 alterations, defined as driver mutation (based on OncoKB), and/or amplification, across BC disease states. We examined the association between HER2 alteration and disease progression and survival. The Kaplan-Meier method was used for time-to-event analyses. Results: A total of 1073 BC pts underwent MSK IMPACT profiling of tumor tissue derived from the following disease states: 36% (n = 380) non-muscle invasive (NMI)BC, 41% (n = 443) muscle invasive (MI)BC, and 23% (n = 250) (met)BC. At initial diagnosis, the median age was 67 years, 77% (n = 822) were male, 86% (n = 928) were white, and 66% (n = 710) were smokers. Overall, 16% (n = 177) of pts had any oncogenic HER2 alteration (Table), including 11% with a HER2 driver mutation and 7% with HER2 amplification The most frequent mutations were S310F (40%, n = 53) and S310Y (14%, n = 19). The rate of HER2 amplification was different among the three groups (p = 0.002), 9% in MIBC and metBC compared to 3% in NMIBC. Among 514 pts with NMIBC, the median time to progression (TTP) to MIBC or metBC was 111.6 months (95% Cl: 85.7-NR). Among NMIBC pts, TTP was significantly shorter for HER2-amplified (n = 17) vs. non-amplified (n = 497) (HR = 1.99, 95%CI: 1.05-3.76, p = 0.034, median 26 vs. 114 months). Among pts with metBC receiving frontline platinum-based chemotherapy (n = 143), the median overall survival (OS) was 25.3 months (95%CI: 18.5-33.9). OS was numerically higher in pts with any oncogenic HER2 alteration (n = 26) compared to wild-type (n = 117) (HR = 0.59, 95% Cl: 0.33-1.07, p = 0.082), though this difference was not statistically significant. The median OS for platinum-refractory metBC pts receiving 2nd line immunotherapy (n = 63) was 10.3 months (95%CI: 7.2-31.6), and the association between OS and HER2 alteration was not significant (HR = 0.57, 95%CI: 0.24-1.37, p = 0.2). Conclusions: HER2 amplification is more frequent in MIBC and metBC than in NMIBC. In NMIBC, HER2 amplification is associated with shorter TTP to MIBC or metBC. HER2 alteration in metBC is associated with a non-significant trend towards improved OS in frontline platinum-treated pts, though this analysis is limited by small sample size.[Table: see text]

Germline Variants Identified in Patients with Early-onset Renal Cell Carcinoma Referred for Germline Genetic Testing

BACKGROUND

Despite guidelines recommending genetic counseling for patients with early-onset renal cell carcinoma (RCC), studies interrogating the spectrum of germline mutations and clinical associations in patients with early-onset RCC are lacking.

OBJECTIVE

To define the germline genetic spectrum and clinical associations for patients with early-onset RCC diagnosed at age ≤46 yr who underwent genetic testing.

DESIGN, SETTING, AND PARTICIPANTS

We retrospectively identified patients with early-onset RCC who underwent germline testing at our institution from February 2003 to June 2020.

OUTCOME MEASUREMENT AND STATISTICAL ANALYSIS

The frequency and spectrum of pathogenic/likely pathogenic (P/LP) variants were determined. Clinical characteristics associated with mutation status were analyzed using two-sample comparison (Fisher's exact or χ2 test).

RESULTS AND LIMITATIONS

Of 232 patients with early-onset RCC, 50% had non-clear-cell histology, including unclassified RCC (12.1%), chromophobe RCC (9.7%), FH-deficient RCC (7.0%), papillary RCC (6.6%), and translocation-associated RCC (4.3%). Overall, 43.5% had metastatic disease. Germline P/LP variants were identified in 41 patients (17.7%), of which 21 (9.1%) were in an RCC-associated gene and 20 (8.6%) in a non-RCC-associated gene, including 17 (7.3%) in DNA damage repair genes such as BRCA1/2, ATM, and CHEK2. Factors associated with RCC P/LP variants include bilateral/multifocal renal tumors, non-clear-cell histology, and additional extrarenal primary malignancies. In patients with only a solitary clear-cell RCC, the prevalence of P/LP variants in RCC-associated and non-RCC-associated genes was 0% and 9.9%, respectively.

CONCLUSIONS

Patients with early-onset RCC had high frequencies of germline P/LP variants in genes associated with both hereditary RCC and other cancer predispositions. Germline RCC panel testing has the highest yield when patients have clinical phenotypes suggestive of underlying RCC gene mutations. Patients with early-onset RCC should undergo comprehensive assessment of personal and family history to guide appropriate genetic testing.

PATIENT SUMMARY

In this study of 232 patients with early-onset kidney cancer who underwent genetic testing, we found a high prevalence of mutations in genes that increase the risk of cancer in both kidneys and other organs for patients and their at-risk family members. Our study suggests that patients with early-onset kidney cancer should undergo comprehensive genetic risk assessment.

Association Between Androgen Deprivation Therapy and Mortality Among Patients With Prostate Cancer and COVID-19

IMPORTANCE

Androgen deprivation therapy (ADT) has been theorized to decrease the severity of SARS-CoV-2 infection in patients with prostate cancer owing to a potential decrease in the tissue-based expression of the SARS-CoV-2 coreceptor transmembrane protease, serine 2 (TMPRSS2).

OBJECTIVE

To examine whether ADT is associated with a decreased rate of 30-day mortality from SARS-CoV-2 infection among patients with prostate cancer.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study analyzed patient data recorded in the COVID-19 and Cancer Consortium registry between March 17, 2020, and February 11, 2021. The consortium maintains a centralized multi-institution registry of patients with a current or past diagnosis of cancer who developed COVID-19. Data were collected and managed using REDCap software hosted at Vanderbilt University Medical Center in Nashville, Tennessee. Initially, 1228 patients aged 18 years or older with prostate cancer listed as their primary malignant neoplasm were included; 122 patients with a second malignant neoplasm, insufficient follow-up, or low-quality data were excluded. Propensity matching was performed using the nearest-neighbor method with a 1:3 ratio of treated units to control units, adjusted for age, body mass index, race and ethnicity, Eastern Cooperative Oncology Group performance status score, smoking status, comorbidities (cardiovascular, pulmonary, kidney disease, and diabetes), cancer status, baseline steroid use, COVID-19 treatment, and presence of metastatic disease.

EXPOSURES

Androgen deprivation therapy use was defined as prior bilateral orchiectomy or pharmacologic ADT administered within the prior 3 months of presentation with COVID-19.

MAIN OUTCOMES AND MEASURES

The primary outcome was the rate of all-cause 30-day mortality after COVID-19 diagnosis for patients receiving ADT compared with patients not receiving ADT after propensity matching.

RESULTS

After exclusions, 1106 patients with prostate cancer (before propensity score matching: median age, 73 years [IQR, 65-79 years]; 561 (51%) self-identified as non-Hispanic White) were included for analysis. Of these patients, 477 were included for propensity score matching (169 who received ADT and 308 who did not receive ADT). After propensity matching, there was no significant difference in the primary end point of the rate of all-cause 30-day mortality (OR, 0.77; 95% CI, 0.42-1.42).

CONCLUSIONS AND RELEVANCE

Findings from this cohort study suggest that ADT use was not associated with decreased mortality from SARS-CoV-2 infection. However, large ongoing clinical trials will provide further evidence on the role of ADT or other androgen-targeted therapies in reducing COVID-19 infection severity.

PD-L1 quantification across tumor types using the reverse phase protein microarray: implications for precision medicine

BACKGROUND

Anti-programmed cell death protein 1 and programmed cell death ligand 1 (PD-L1) agents are broadly used in first-line and second-line treatment across different tumor types. While immunohistochemistry-based assays are routinely used to assess PD-L1 expression, their clinical utility remains controversial due to the partial predictive value and lack of standardized cut-offs across antibody clones. Using a high throughput immunoassay, the reverse phase protein microarray (RPPA), coupled with a fluorescence-based detection system, this study compared the performance of six anti-PD-L1 antibody clones on 666 tumor samples.

METHODS

PD-L1 expression was measured using five antibody clones (22C3, 28-8, CAL10, E1L3N and SP142) and the therapeutic antibody atezolizumab on 222 lung, 71 ovarian, 52 prostate and 267 breast cancers, and 54 metastatic lesions. To capture clinically relevant variables, our cohort included frozen and formalin-fixed paraffin-embedded samples, surgical specimens and core needle biopsies. Pure tumor epithelia were isolated using laser capture microdissection from 602 samples. Correlation coefficients were calculated to assess concordance between antibody clones. For two independent cohorts of patients with lung cancer treated with nivolumab, RPPA-based PD-L1 measurements were examined along with response to treatment.

RESULTS

Median-center PD-L1 dynamic ranged from 0.01 to 39.37 across antibody clones. Correlation coefficients between the six antibody clones were heterogeneous (range: -0.48 to 0.95) and below 0.50 in 61% of the comparisons. In nivolumab-treated patients, RPPA-based measurement identified a subgroup of tumors, where low PD-L1 expression equated to lack of response.

CONCLUSIONS

Continuous RPPA-based measurements capture a broad dynamic range of PD-L1 expression in human specimens and heterogeneous concordance levels between antibody clones. This high throughput immunoassay can potentially identify subgroups of tumors in which low expression of PD-L1 equates to lack of response to treatment.

PET/CT scan and biopsy-driven approach for safe anti-PD-1 therapy discontinuation in patients with advanced melanoma

Background

Limited data exist on safe discontinuation of antiprogrammed cell death protein 1 (PD-1) therapy in responding patients with advanced melanoma. The use of 18fluorodeoxyglucose (¹⁸FDG)-PET/CT scan and tumor biopsy for assessment of active disease may be an effective predictive biomarker to guide such treatment decisions.

Methods

A retrospective study of 122 patients with advanced melanoma treated with anti-PD-1 monotherapy or anti-PD-1/anticytotoxic T-lymphocyte-associated protein 4 combination therapy at Georgetown Lombardi Comprehensive Cancer Center was conducted. Uveal melanoma patients and those receiving concurrent experimental therapy were excluded. Baseline characteristics, treatment outcomes, and survival were analyzed. Patients who decided to come off treatment typically after 12 months using CT scan radiographic complete response (CR), ¹⁸FDG-PET/CT scan complete metabolic response (CMR) or tumor biopsy of a non-CR/CMR tumor site negative for active disease (possible pathological CR) were identified and compared with patients who discontinued treatment due to toxicity while their disease was in control. Event-free survival (EFS) was assessed from the last dose of anti-PD-1 therapy to progression requiring subsequent treatment (surgery, radiation, and/or systemic therapy) or referral to hospice/death due to melanoma.

Results

24 (20%) patients discontinued treatment by choice with no active disease and 28 (23%) patients discontinued treatment due to toxicity with disease control after 12-month and 4-month median treatment durations, respectively. Similar baseline characteristics were observed between cohorts except higher prior receipt of ipilimumab (29% vs 7%; p=0.036) and fewer BRAF mutant positive disease (17% vs 41%; p=0.064) in patients off treatment by choice. Three-year EFS rates were 95% and 71%, respectively. No significant associations between EFS and sex, disease stage, lactate dehydrogenase elevation, BRAF status, prior systemic therapy, ECOG performance status, presence of brain metastases, or combination versus monotherapy were observed. Tumor biopsies led to alternative management in 3/10 patients due to active metastatic melanoma or second malignancy.

Conclusions

Anti-PD-1 therapy discontinuation after 12 months when no active disease is observed on CT scan, PET/CT scan or tumor biopsy may have low rates of disease relapse in patients with advanced melanoma. Biopsy of residual disease may frequently lead to a change in management. These findings are undergoing validation in the EA6192 trial.

Automated Identification of Patients With Immune-Related Adverse Events From Clinical Notes Using Word Embedding and Machine Learning

PURPOSE

Although immune checkpoint inhibitors (ICIs) have substantially improved survival in patients with advanced malignancies, they are associated with a unique spectrum of side effects termed immune-related adverse events (irAEs). To ensure treatment safety, research efforts are needed to comprehensively detect and understand irAEs. Retrospective analysis of data from electronic health records can provide knowledge to characterize these toxicities. However, such information is not captured in a structured format within the electronic health record and requires manual chart review.

MATERIALS AND METHODS

In this work, we propose a natural language processing pipeline that can automatically annotate clinical notes and determine whether there is evidence that a patient developed an irAE. Seven hundred eighty-one cases were manually reviewed by clinicians and annotated for irAEs at the patient level. A dictionary of irAEs keywords was used to perform text reduction on clinical notes belonging to each patient; only sentences with relevant expressions were kept. Word embeddings were then used to generate vector representations over the reduced text, which served as input for the machine learning classifiers. The output of the models was presence or absence of any irAEs. Additional models were built to classify skin-related toxicities, endocrine toxicities, and colitis.

RESULTS

The model for any irAE achieved an average F1-score = 0.75 and area under the receiver operating characteristic curve = 0.85. This outperformed a basic keyword filtering approach. Although the classifier of any irAEs achieved good accuracy, individual irAE classification still has room for improvement.

CONCLUSION

We demonstrate that patient-level annotations combined with a machine learning approach using keywords filtering and word embeddings can achieve promising accuracy in classifying irAEs in clinical notes. This model may facilitate annotation and analysis of large irAEs data sets.

Real-time data quality assurance analysis for real-world, pan-cancer data

e18775

Background: The production of high-quality real-world data requires comprehensive and meticulous data quality assurance (QA) methods to guarantee that adequate standards of accuracy, completeness, and consistency are met. Memorial Sloan Kettering Cancer Center (MSKCC) synthesizes manually curated Electronic Health Record (EHR) data to collect and harmonize the fundamental data elements across all cancer types. Centralized real-time analysis of curated data quality can allow for rigorous review to identify areas of strength and opportunities for improvement in the curation process. Methods: MSKCC built the Core Clinical Data Element (CCDE) data model, which encompasses aspects of PRISSMM, ASCO’s mCODE, and NAACCR tumor registry frameworks, to capture standardized real-world, pan-cancer, pan-specialty data across 11 modules, including cancer genomics, imaging, pathology, surgery, and radiation. A key component within the QA process is source data verification (SDV), the comparison of curated data against source documents to identify inconsistencies. Any discrepancies detected are classified into major and minor violations. Major violations are errors or omissions on core data elements that would impact time interval calculations, such as an incorrect procedure date. Minor violations are errors or omissions on less critical data elements, such as a missing radiation therapy dose. Identifying these inconsistences allows the QA team to recognize patterns in curation errors and distinguish areas for curator retraining. Results: With limited functionality in basic standard data quality checks that exist across various data storage platforms, an interactive application was developed using the R Shiny package to access data as cases are recorded and summarize findings from SDV in real time. The app has two panels, each stratified by CCDE module. The first panel details the total number of forms curated and percentage of forms that underwent SDV, with each form representing one of the 11 modules. The other panel consists of a set of tables that summarize specific major and minor violations based on user selection of a denominator of either patients (e.g. how many patients had a violation on at least one imaging report) or forms (e.g. how many imaging reports had a violation). We will demonstrate the utility of the app and discuss benefits of real time evaluation in large-scale, real-world EHR curation efforts. Conclusions: We recommend automated, user-friendly tools to assess data quality of such efforts. With real-time analysis, the tool allows for ongoing and regular data checks, enabling clarification of directives and retraining of curators as necessary early in the curation process. As the data curation efforts expand to more cancer cohorts, the app examines data quality of each cohort to ensure consistent evaluation. This offers transparency of data quality to ensure usability in real-world data for rigorous research.

A comprehensive Memorial Sloan Kettering Cancer Center real-world data model: Core clinical data elements

e18755

Background: The 2016 21st Century Cures Act supports the use of Real-World Data (RWD) for regulatory decision/approval. Due to technological advances, a vast amount of health-related data are now available, but most are not standardized nor readily useable for research. Also, currently available standardized RWD models are not applicable across cancer types or oncology specialties (surgery, medical oncology, radiation oncology, pathology, radiology, etc.). To address these deficiencies Memorial Sloan Kettering Cancer Center (MSKCC) built a comprehensive, pan-cancer, pan-specialty RWD model. Methods: The Core Clinical Data Element (CCDE) data model incorporates aspects of existing academic and biopharma data models, including PRISSMM framework, ASCO’s mCODE, and NAACCR tumor registry model. The data model encompasses 11 domains that are critical to the understanding of the patient’s cancer journey, including: demographic, comorbidities, diagnosis, pathology, imaging, genomics, cancer surgeries, radiation oncology treatments, medical oncology treatments, cancer status/progression, and additional health information. To align with current standards, we are using ICD-10, ICDO3, CTACE V5.0, HL7, SNOMED and LOINC code sets. Further, this adaptable model allows for 5-10 disease specific elements to accommodate for disease heterogenicity and capture the differences among cancer types. Results: The CCDE database includes 1,126 of total data elements. MSKCC has 52,704 patients with MSK-IMPACT (Next-Generation sequencing platform with 505 genes panel) testing of which, we have identified 1,132 bladder cancer patients with at-least one year of cancer care follow-up for the initial curation cohort. Patients were identified as having an OncoTree bladder tumor type code that is assigned by a pathologist who attests the diagnosis by reviewing results from clinical tests on tumor specimens. To the date, 641 patients including 46,415 curated forms have been curated (Table). Conclusions: The comprehensive MSKCC’s CCDE data model standardizes the common and critical pan-cancer and pan-specialty elements for RWD. The dataset resulting from this curation efforts will provide robust structured and unified genomic and phenomic data across tumor types for future research enabling greater collaboration across various cancer types as well as oncology specialties.[Table: see text]

Real-world outcomes of treatment with immune checkpoint inhibitors in unique patient cohorts: Elderly, non-caucasian race, poor performance status, obese, chronic viral infections, and autoimmune diseases

2641

Background: Immune Checkpoint Inhibitors (ICI) have revolutionized current cancer treatment. Nevertheless, outcomes data across various patient cohorts are lacking. To address this knowledge gap, we conducted a comprehensive analysis of real-world data (RWD) that included patient cohorts traditionally underrepresented in clinical trials. Methods: We identified patients (pts) treated with ICI (anti-CTLA-4, anti-PD(L)1 or their combination at 6 US academic and community hospitals from 1/2011 – 4/2018. Clinical data obtained from EHR and CTCAE V4.03 was used to define immune-related adverse events (irAEs). Results: A total of 1332 pts treated with 1443 unique ICI treatments were included in the cohort. The median age was 66 (21-87), Male 58% (827), Caucasian 70% (1004), African American (AA) 16% (232), other race 14% (207), ECOG PS 0,1 79% (1130), chronic viral infection 5% [hepatitis B (24), hepatitis C (32) and HIV (17)], with BMI > 30 22% (287) and autoimmune disease (AID) 15% (215). Lung cancer (NSCLC) 34% (423), and melanoma 27% (389) were top 2 tumor types and nivolumab 38% (544), pembrolizumab 23% (332), and ipilimumab plus nivolumab 12% (180) were the most common ICI treatments. Overall survival (OS) was worse for patients with ECOG ≥2 (0.34 - 0.63) vs. ECOG 0,1 (1.27 - 1.73, P <0.001), and better with AID (1.21 - 2.63) vs. no AID ( 0.90 - 1.24, P=0.01) and Caucasian (1.02 - 1.45) vs AA (0.72 - 1.30, P=0.02). No difference in OS was noted for sex, other races, h/o chronic viral infection or obesity. We performed an analysis of OS and irAEs restricted to NSCLC patients (n=423); (N=447 unique ICI treatments); age >75 27% (120), AA 28% (124), Female 50% (224), ECOG PS ≥2 23% (104), BMI >30 15% (62), chronic viral infections 10% (44), and AID 14% (62). The ICI therapies were nivolumab 55% (245), pembrolizumab 23% (102), and atezolizumab 6% (27) and 16% (others). Data is contained in the table. Conclusions: Overall, in our RWD, OS appeared to be similar across above cohorts except poor OS for pts with ECOG ≥2. irAEs also appeared to be similar across cohorts except less with ECOG ≥2. In NSCLC cohort, we noted similar findings except less irAEs in Male cohort. Prospective studies are needed to confirm the above findings.[Table: see text]

Late immune-related adverse events with immune checkpoint inhibitors

2635

Background: Immune Checkpoint Inhibitors (ICIs) are associated with unique immune-related adverse events (irAEs). IrAEs can occur at any timepoint of ICI treatment. Late irAEs are not well reported in the literature. Herein, we attempt to characterize irAEs that occur 6-month, one year and two years after ICI treatment initiation. Methods: We identified patients treated with ICIs (anti-CTLA-4, anti-PD(L)-1 either alone or in combination or with chemotherapy) across Hackensack Meridian Health hospital and MedStar Georgetown University Health systems from 12011 to 4/2018. Patients' baseline demographics, treatment history, and irAEs were collected from EHR. CTCAE V4.03 was used to grade irAEs. Results: We identified 1332 patients treated with 1443 unique ICIs. The ICI therapies were nivolumab 38% (543), pembrolizumab 23% (332), ipilimumab plus nivolumab 12% (180), ipilimumab 11% (161), Atezolizumab 3% (47) and others 13% (180). Tumor types were lung cancer 34% (496), melanoma 27% (389), GI cancers 6% (92), kidney cancer 6% (87), and other cancers 26% (379). The median age was 66 (21-87), age >75 37% (541), Caucasian 67% (970). We identified a total of 911 any grade irAEs among 37% (552) therapies. Among, 911 irAEs, grade 1-2, grade ≥3 and unknown grade irAEs were 39% (572), 12% (182) and 11% (157), respectively. The most common any grade irAEs were skin rash 22% (202), colitis 13% (120), and hepatitis 12% (108). 84% of all irAEs and 85% of ≥ Grade 3 irAEs occurred within 6 months of treatment initiation. Of the 350, patients on active treatment at six months, 37 % (132) and 7% (26) developed any grade and grade ≥3 irAEs, respectively. irAEs that had > 10% of their occurrences after six months were skin rash and colitis 14% each. Other common irAEs were hypothyroidism, hepatitis, joint pain, pruritis and pneumonitis at 7% each. Among 170 patients on active treatment at one year, 37% (62) and 7% (12) developed any grade and grade ≥3 irAEs respectively. irAEs with >10% incidence after one year of treatment were rash 19% and hepatitis 13%. Conclusions: Our RWE findings suggest although 85% irAEs occurs within the first six months of treatment, late irAEs can occur with ICI treatment. The incidence and pattern of late irAEs appears similar to early irAEs, (e.g., skin rash, colitis, hypothyroidism and hepatitis) with pneumonitis being a notable exception. It is uncertain if these results will be influenced by changing patterns of ICI use (e.g. different diseases and/or regimens) over time.[Table: see text]

Nivolumab plus cabozantinib in patients with non-clear cell renal cell carcinoma: Results of a phase 2 trial

4509

Background: Cabozantinib plus nivolumab (CaboNivo) improved objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) over sunitinib in a phase 3 trial for metastatic clear cell renal cell carcinoma (RCC). (Choueiri, abstract 6960, ESMO 2020) We report the results of a phase 2 trial of CaboNivo in patients (pts) with non-clear cell RCC. Methods: Pts had advanced non-clear cell RCC, 0 or 1 prior systemic therapies excluding prior immune checkpoint inhibitors, and measurable disease by RECIST. Cabo 40 mg/day plus Nivo 240 mg every 2 weeks or 480 mg every 4 weeks was given across two cohorts. Cohort 1: papillary, unclassified, or translocation associated RCC; Cohort 2: chromophobe RCC. The primary endpoint was ORR by RECIST; secondary endpoints included PFS, OS, and safety. Cohort 1 was a single stage design that met its primary endpoint and was expanded to produce more precise estimates of ORR. Cohort 2 was a Simon two-stage design that closed early for lack of efficacy. Correlative analyses by next generation sequencing were performed and to be presented. Results: A total of 40 pts were treated in Cohort 1, and 7 pts were treated in Cohort 2 (data cutoff: Jan 20, 2021). Median follow up time was 13.1 months (range 2.2 – 28.6). In Cohort 1, 26 (65%) pts were previously untreated, and 14 (35%) pts had 1 prior line: 10 (25%) received prior VEGF-targeted therapy and 8 (20%) received prior mTOR-targeted therapy. ORR for Cohort 1 was 48% (95% CI 31.5–63.9; Table). Median PFS was 12.5 months (95% CI 6.3–16.4) and median OS was 28 months (95% CI 16.3–NE). No responses were seen among 7 patients in Cohort 2 with chromophobe histology (Table). Grade 3/4 treatment emergent adverse events were consistent with that reported in the phase 3 trial; Grade 3/4 AST and ALT were 9% and 15%, respectively. Cabozantinib and nivolumab were discontinued due to toxicity in 17% and 19% of pts, respectively. Conclusions: CaboNivo had an acceptable safety profile and showed promising efficacy in metastatic non-clear cell RCC pts with papillary, unclassified, or translocation associated histologies whereas activity in patients with chromophobe RCC was limited. Clinical trial information: NCT03635892. [Table: see text]

The role of androgen deprivation therapy on the clinical course of COVID-19 infection in men with prostate cancer

41

Background: TMPRSS2, a cell surface protease which is commonly upregulated in prostate cancer (PC) and regulated by androgens, is a necessary component for SARS-CoV2 cellular entry into respiratory epithelial cells. PC patients receiving ADT were reported to have a lower risk of SARS-CoV-2 infection. However, whether ADT may have an impact on the severity of COVID-19 illness in this population is poorly understood. Methods: In this study performed across 7 US medical centers, we retrospectively evaluated patients with active PC and SARS-COV-2 viral detection by PCR between 03/01/20 and 05/31/20. We collected information on demographics; medical comorbidities; medications; PC Gleason score at initial diagnosis; presence of active disease, metastases, and castration resistance; ADT use as defined by GnRH analog or antagonist within 3 months or castration levels of testosterone < 50 ng/dL within 6 months of COVID-19 diagnosis, or history of bilateral orchiectomy; active non-ADT systemic therapies including, but not limited to, androgen-receptor-targeted therapies and chemotherapy; and COVID-19-related outcomes including hospitalization, supplemental oxygen use, mechanical ventilation requirement, WHO COVID-19 ordinal scale for clinical improvement, follow-up duration, and vital status. Multivariable mixed-effect logistic regression was performed to evaluate any difference in COVID-19 clinical outcomes between patients on and not on ADT. Survival analysis was done using adjusted Cox proportion-hazards regression model. All tests were two-sided at 0.05 significance level. Results: We identified 465 evaluable patients with median age of 71 (61-81) years. Median duration of follow-up was 60 (12-114.2) days. In this follow up period, there were 195 (41.9%) hospitalizations and 111 (23.9%) deaths. When adjusted for age, BMI, and PC clinical disease state, overall survival (HR 1.28 [95%CI 0.79-2.08], P = 0.32), hospitalization status (HR 1.07 [0.61-1.87], P = 0.82), supplemental oxygen use (HR 1.29 [0.77-2.17], P = 0.34), and use of mechanical ventilation (HR 1.07 [0.51-2.23], P = 0.87) were not statistically different between ADT and non-ADT cohorts. Similarly, in subgroup analysis, no statistical difference in overall survival was found between ADT and non-ADT cohorts for hospitalized patients (HR 1.42 [0.82-2.47], P = 0.21) and those receiving supplemental oxygen (HR 1.10 [0.65-1.85], P = 0.73). Conclusions: In this retrospective cohort of PC patients, use of ADT prior to COVID-19 diagnosis does not protect against severe COVID-19 illness as defined by hospitalization, supplemental oxygen use, or death. Further preclinical work in understanding TMPRSS2 expression and androgen regulation in respiratory epithelial cells is needed. As well, longer clinical follow-up and additional clinical studies inclusive of prospective data are warranted to fully address this question.

Success and failure of additional immune modulators in steroid-refractory/resistant pneumonitis related to immune checkpoint blockade

BACKGROUND

Pneumonitis related to immune checkpoint blockade is uncommon but can be severe, fatal or chronic. Steroids are first-line treatment, however, some patients are refractory or become resistant to steroids. Like many immune-related adverse events, little is known regarding the outcomes and optimal management of patients in whom steroids are ineffective.

METHODS

We performed a single-center retrospective cohort study at a high-volume tertiary cancer center to evaluate the clinical course, management strategies and outcomes of patients treated for immune checkpoint pneumonitis with immune modulatory medications in addition to systemic steroids. Pharmacy records were queried for patients treated with both immune checkpoint blockade and receipt of additional immune modulators. Records were then manually reviewed to identify patients who received the additional immune modulators for immune checkpoint pneumonitis.

RESULTS

From 2013 to 2020, we identified 26 patients treated for immune checkpoint pneumonitis with additional immune modulators in addition to steroids. Twelve patients (46%) were steroid-refractory and 14 (54%) were steroid-resistant. Pneumonitis severity included grade 2 (42%) or grade 3-4 (58%). Additional immune modulation consisted of tumor necrosis factor-alpha inhibitor (77%) and/or mycophenolate (23%). Durable improvement in pneumonitis following initiation of additional immune modulators occurred in 10 patients (38%), including three patients (12%) in whom pneumonitis resolved and all immunosuppressants ceased. The rate of 90-day all-cause mortality/hospice referral was 50%. At last follow-up, mortality attributable to pneumonitis was 23%. In addition to mortality from pneumonitis and cancer, 3 patients (12%) died due to infections possibly associated with immunosuppression.

CONCLUSIONS

Steroid-refractory or -resistant immune checkpoint pneumonitis is uncommon but associated with significant morbidity and mortality. Additional immunomodulators can yield durable improvement, attained in over one third of patients. An improved understanding of the underlying biology of immune-related pneumonitis will be crucial to guide more precise and effective treatment strategies in the future.

Salvage Ipilimumab and Nivolumab in Patients With Metastatic Renal Cell Carcinoma After Prior Immune Checkpoint Inhibitors

PURPOSE

Immune checkpoint inhibitors (ICIs) are standard therapy in metastatic renal cell carcinoma (RCC). The safety and activity of the combination of ipilimumab and nivolumab in patients who have received prior ICI targeting the programmed death 1 (PD-1) pathway remains unknown. We evaluated ipilimumab and nivolumab in patients with metastatic RCC after prior treatment with anti-PD-1 pathway-targeted therapy.

PATIENTS AND METHODS

Patients with metastatic RCC who received prior anti-PD-1 pathway-targeted therapy and subsequently received ipilimumab and nivolumab were reviewed. Objective response rate and progression-free survival per investigator assessment were recorded. Toxicity of ipilimumab and nivolumab was also assessed.

RESULTS

Forty-five patients with metastatic RCC were included. All patients (100%) received prior ICIs targeting the PD-1 pathway. The median age was 62 years (range, 21-82 years). At a median follow-up of 12 months, the objective response rate to ipilimumab and nivolumab was 20%. The median progression-free survival while on ipilimumab and nivolumab was 4 months (range, 0.8-19 months). Immune-related adverse events (irAEs) of any grade with ipilimumab and nivolumab were recorded in 29 (64%) of the 45 patients; grade 3 irAEs were recorded in 6 (13%) of the 45 patients.

CONCLUSION

Ipilimumab and nivolumab demonstrated antitumor activity with acceptable toxicity in patients with metastatic RCC who had prior treatment with checkpoint inhibition.

Impact of treatment line on outcomes with salvage ipilimumab + nivolumab in metastatic renal cell carcinoma (mRCC)

e17092

Background: With the approval of ipilimumab plus nivolumab (I+N) and other immune checkpoint blockade (ICB) based combinations in the first-line setting, the role of I+N for salvage is of high interest for treatment sequencing, yet data is limited. Methods: We conducted a retrospective review of mRCC patients (pts) treated with I+N in the second-line (2L) and beyond settings at MSKCC between 2013-2019. Pt demographics, treatment history and toxicity were compiled. IMDC-risk status was calculated at I+N therapy start. Time to treatment failure (TTF) was defined as earliest date of clinical progression, therapy change or death, and overall survival (OS) was estimated by Kaplan-Meier method. Results: 36 pts received I+N in the 2+L setting, including 31/36 with clear-cell histology. Evaluable IMDC-risk at I+N start was favorable in 1/35 and intermediate-poor in 34/35 pts. The most common 1L therapies were anti-VEGF (22/36) and VEGF + ICB (6/36). 11/36 pts had ICB treatment exposure prior to I+N therapy. I+N therapy in the 2L, 3L and 4L was in 21/36, 8/36 and 7/36 pts, respectively, and 7/36 pts continue I+N at data cut-off. 8/36 pts discontinued I+N due to toxicity, 20/36 pts discontinued therapy due to disease progression, and 1 pt discontinued per pt preference. Cohort median OS was 14.8 months (95%CI: 4.2-44). Overall median TTF was 5.0 months (95%CI: 2.9-14.4), and TTF per 2L, 3L and 4+L was 8.3, 8.9 and 2.5 months, respectively. The number of patients who completed all 4 I+N induction cycles in the 2L, 3L, and 4+L was 11/21 (52%), 5/8 (63%), and 1/7 (14%). The number of patients who subsequently received nivolumab maintenance therapy after induction was 16/21 (76%) in the 2L, 1/8 (13%) in the 3L, and 0/7 (0%) in 4+L. Conclusions: With emerging treatment options for mRCC, this study reveals activity and safety for I+N in 2+L settings. In this limited cohort, completion of induction ipilimumab and use of maintenance nivolumab decline in later-line settings, suggesting limitations as salvage therapy. [Table: see text]

A Single-Scan, Rapid Whole-Brain Protocol for Quantitative Water Content Mapping With Neurobiological Implications

Water concentration is tightly regulated in the healthy human brain and changes only slightly with age and gender in healthy subjects. Consequently, changes in water content are important for the characterization of disease. MRI can be used to measure changes in brain water content, but as these changes are usually in the low percentage range, highly accurate and precise methods are required for detection. The method proposed here is based on a long-TR (10 s) multiple-echo gradient-echo measurement with an acquisition time of 7:21 min. Using such a long TR ensures that there is no T₁ weighting, meaning that the image intensity at zero echo time is only proportional to the water content, the transmit field, and to the receive field. The receive and transmit corrections, which are increasingly large at higher field strengths and for highly segmented coil arrays, are multiplicative and can be approached heuristically using a bias field correction. The method was tested on 21 healthy volunteers at 3T field strength. Calibration using cerebral-spinal fluid values (~100% water content) resulted in mean values and standard deviations of the water content distribution in white matter and gray matter of 69.1% (1.7%) and 83.7% (1.2%), respectively. Measured distributions were coil-independent, as seen by using either a 12-channel receiver coil or a 32-channel receiver coil. In a test-retest investigation using 12 scans on one volunteer, the variation in the mean value of water content for different tissue types was ~0.3% and the mean voxel variability was ~1%. Robustness against reduced SNR was assessed by comparing results for 5 additional volunteers at 1.5T and 3T. Furthermore, water content distribution in gray matter is investigated and regional contrast reported for the first time. Clinical applicability is illustrated with data from one stroke patient and one brain tumor patient. It is anticipated that this fast, stable, easy-to-use, high-quality mapping method will facilitate routine quantitative MR imaging of water content.

Safety and efficacy of immune checkpoint inhibitors (ICIs) in cancer patients with HIV, hepatitis B, or hepatitis C viral infection

Background

Patients with chronic viral infections including human immunodeficiency virus (HIV), hepatitis B (HBV) and hepatitis C (HCV) are at increased risk of developing malignancies. The safety and efficacy of ICI therapy in patients with both cancer and chronic viral infections is not well established as most clinical trials of ICIs excluded these patient populations.

Methods

We performed a retrospective analysis of patients with advanced-stage cancers and HIV, HBV, or HCV infection treated with ICI therapy at 5 MedStar Health hospitals from January 2011 to April 2018.

Results

We identified 50 patients including 16 HIV, 29 HBV/HCV, and 5 with concurrent HIV and either HBV or HCV. In the HIV cohort (n = 21), any grade immune-related adverse events (irAEs) were 24% with grade ≥ 3 irAEs 14%. Among 5 patients with matched pre/post-treatment results, no significant changes in HIV viral load and CD4+ T-cell counts were observed. RECIST confirmed (n = 18) overall response rate (ORR) was 28% with 2 complete responses (CR) and 3 partial responses (PR). Responders included 2 patients with low baseline CD4+ T-cell counts (40 and 77 cells/ul, respectively). In the HBV/HCV cohort (n = 34), any grade irAEs were 44% with grade ≥ 3 irAEs 29%. RECIST confirmed ORR was 21% (6 PR). Among the 6 patients with known pre/post-treatment viral titers (2 HCV and 4 HBV), there was no evidence of viral reactivation.

Conclusions

Our retrospective series is one of the largest case series to report clinical outcomes among HIV, HBV and HCV patients treated with ICI therapy. Toxicity and efficacy rates were similar to those observed in patients without chronic viral infections. Viral reactivation was not observed. Tumor responses occurred in HIV patients with low CD4 T-cell counts. While prospective studies are needed to validate above findings, these data support not excluding such patients from ICI–based clinical trials or treatment.

P14.32 Spatial discrepancies between FET PET and conventional MRI in patients with newly diagnosed glioblastoma

BACKGROUND

In patients with glioblastoma, the tissue showing contrast enhancement (CE) in MRI is usually the target for resection or radiotherapy. However, the solid tumor mass typically extends beyond the area of CE. Amino acid PET can detect tumor parts that show no CE. We systematically investigated tumor volumes delineated by amino acid PET and MRI in newly diagnosed, untreated glioblastoma patients.

MATERIAL AND METHODS

Preoperatively, 50 patients with subsequently neuropathologically confirmed glioblastoma underwent O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) PET, fluid-attenuated inversion recovery (FLAIR), and CE MRI. Areas of CE were manually delineated. FET PET tumor volumes were segmented using a tumor-to-brain ratio ≥ 1.6. The percentage of overlapping volumes (OV), as well as Dice and Jaccard spatial similarity coefficients (DSC; JSC), were calculated. FLAIR images were evaluated visually.

RESULTS

In 86% of patients (n = 43), the FET PET tumor volume was significantly larger than the volume of CE (21.5 ± 14.3 mL vs. 9.4 ± 11.3 mL; P < 0.001). Forty patients (80%) showed both an increased uptake of FET and CE. In these 40 patients, the spatial similarity between FET and CE was low (mean DSC, 0.39 ± 0.21; mean JSC, 0.26 ± 0.16). Ten patients (20%) showed no CE, and one of these patients showed no FET uptake. In 10% of patients (n = 5), increased FET uptake was present outside of areas of FLAIR hyperintensity.

CONCLUSION

Our results show that the metabolically active tumor volume delineated by FET PET is significantly larger than tumor volume delineated by CE. The data strongly suggest that the information derived from FET PET should be integrated into the management of newly diagnosed glioblastoma patients.

FUNDING

This work was supported by the Wilhelm-Sander Stiftung, Germany

A case of checkpoint inhibitor-induced celiac disease

BACKGROUND

Immune checkpoint inhibitors (ICIs) have now become standard of care treatment for many malignancies. ICIs are associated with unique immune mediated adverse events (irAEs) due to dysregulation of immune activation. As treatment with ICIs is becoming more common, rare irAEs are also being recognized. Here we report a case of ICI-induced celiac disease.

CASE

A 74-year-old Caucasian female with metastatic renal carcinoma received second line nivolumab (anti-PD1 antibody) after initial disease progression on sunitinib. Ipilimumab was added after she failed to respond to six cycles of nivolumab monotherapy. One week after her first cycle of combination treatment, she presented with nausea, vomiting, grade 1 diarrhea, and weight loss. She underwent endoscopy, which showed bile stasis in the stomach, normal appearing stomach mucosa, and nonbleeding erythematous mucosa in the duodenal bulb. Stomach biopsy showed moderate active chronic gastritis. Duodenal biopsy showed moderate chronic active duodenitis with focal neutrophilic cryptitis, mucosal erosions, villous atrophy, mildly increased intraepithelial lymphocytes, and moderate chronic inflammation in the lamina propria pathognomonic of celiac disease. Symptoms improved with gluten-free diet, twice-daily omeprazole and anti-emetics and she was able to continue on treatment.

CONCLUSIONS

There has been only one published case reporting ICI-induced celiac disease. Our case report highlights a rare irAE (celiac disease) associated with ICI treatment. It is unclear whether the patient had previously undiagnosed celiac disease or whether ICIs triggered her enteritis. Our patient was able to continue treatment with ICIs with dietary modifications, suggesting correct diagnosis is critical for optimal patient outcome.