Race in a molecular tumor board compared to cancer registry population

103

Background: Disparities in cancer care due to race and ethnicity are prevalent in both the care patients receive and patient outcomes. The evaluation of next generation sequencing (NGS) results from patients with advanced cancer by a molecular tumor board (MTB) has become standard practice in many institutions for the identification of additional treatment options and targeted therapies. We sought to compare the racial distribution of patients evaluated by our MTB with our institutional cancer registry (CR). Methods: We tabulated the racial distribution of 560 MTB patients chosen for presentation in a bimonthly case conference based on physician request or clinical interest from more than 2,500 NGS reports of patients with advanced cancer from 2016 through 2020. Self-identified race from patients with stage 4 cancer within our institutional CR from the same time interval was compared to the MTB population from each year using the Chi-Squared test. The Cochran-Mantel-Haenszel test was used to analyze the relationship between race and MTB/CR after controlling for year. Race categories were defined as Asian, Black/African-American (AA), White/Caucasian, and other. Results: We identified 4,151 CR patients with stage 4 cancer from 2016 through 2020, 573 of whom identified as Black/AA (13.8%). Of the 560 MTB patients, 55 were Black/AA (9.8%). When controlling for year, Black/AA patients were less frequently included in the MTB compared to the CR (p = 0.0128). Conclusions: Black/AA patients with advanced cancer are under-represented in our MTB. Larger studies are warranted to examine underlying causes of this discrepancy including implicit bias, generalizability of this finding to other minorities and institutions, and potential remedies to ensure equitable access to state-of-the-art cancer care.[Table: see text]

Homologous recombination deficiency (HRD) alterations as a predictor of responsiveness to combination pancreatic cancer in advanced pancreatic ductal adenocarcinoma (PDAC)

e16259

Background: Advanced PDAC is associated with a poor prognosis and median survivals of less than one year. FOLFIRINOX and Gemcitabine/Nab-Paclitaxel are both considered standard of care in the first line setting, though there is considerable variability among oncologists regarding sequencing of these regimens. Pre-clinical and clinical data support the use of platinum therapy in HRD+ PDAC, but little is known about how this translates into best practices regarding sequencing of treatment or clinical benefit relevant to Gemcitabine based therapies in this population. Methods: Using the cancer registry and Inova molecular tumor board database, we retrospectively reviewed all patients from January 2015 until November 2021 with unresectable PDAC who had NGS performed and stratified by the presence or absence of HRD mutation. HRD mutations included by BRCA1, BRCA2, PALB2, ATM, ATR, CHEK2, ARID1A, FANCA, FANCL. We analyzed for RR to Folfirinox/Folfox and Gemcitabine based therapies, adjusted time to progression on Folfirinox/Folfox (time from first dose of oxaliplatin based combination to first dose of gemcitabine to account for those who stopped oxaliplatin early for neuropathy or received maintenance therapy), and time on gemcitabine based therapy. Results: 142 consecutive patients with advanced PDAC were examined, including 20 with HRD variants. When excluding those who eventually received surgery and those who never received platinum therapy, we were left with 87 (74 in Control and 13 in HRD) for analysis. RR to Folfirinox/Folfox was 60% in HRD+ and 31.25% in control (p = 0.71) (Table). Mean adjusted time to progression on Folfirinox/Folfox was 320.0 days in HRD+ group and 252.9 in the control group (p = 0.57) (Table 2). The time on treatment for gemcitabine based combination therapy was 260.7 in HRD+ and 165.1 in control group. Conclusions: HRD+ patients had longer durations of treatment on both 5-FU based therapy and Gemcitabine based combinations. These data challenges the importance of first line platinum therapy in HRD+ patients, and further exploration should be considered.[Table: see text]

Constitutional MLH1 promotor hypermethylation: Clinical characteristics and testing frequency of a poorly recognized mechanism for Lynch-associated malignancies

10581

Background: MSI-H colorectal cancer is most often a result of deleterious mutations in mismatch repair genes, but can also occur through repressed gene transcription due to hypermethylation of the MLH1 promoter, often associated with BRAF V600E mutations. However, there is a subset of patients who have a “constitutional epimutation”, resulting in hypermethylation of MLH1 throughout normal tissue. We observed a young patient develop a second primary Lynch associated malignancy (see Table 1) who was found to have a constitutional epimutation in MLH1 that prompted us to review the frequency with which the test was ordered and the positivity rate, as well as outline the clinical history in the positive cases. Methods: We reviewed all of the testing ordered for MLH1 hypermethylation of peripheral blood (MLHPB) at the Mayo Clinic Laboratory between 09/01/2020 and 09/01/2021. To the best of our knowledge, this is the only clinically available testing lab in the United States. We reviewed positive casesfor characteristics including the number of malignancies, age of diagnosis, and family history. Results: 33 MLHPB total tests were ordered in the United States at the Mayo Clinic laboratories between 09/01/2020 and 09/01/2021. Three of the tests were positive, including the single test ordered by our institution, and one of the two additional positive tests was available for detailed review. Our institution’s positive test case was a 41 year old women who developed T4N2M0 colorectal cancer 5 years after being treated for endometrial cancer. She had endometrial cancer with absence of MLH1 and PMS2 staining and had a negative germline cancer risk panel at age 36. Five years later, she developed a T4N2 colorectal cancer after which repeat germline testing with RNA sequencing was negative and she was found to have constitutional MLH1 promoter hypermethylation. We were able to obtain clinical information about two of the three individuals with positive tests (See Table). Conclusions: Recognition of constitutional MLH1 hypermethylation may allow for earlier recognition of Lynch related malignancies in affected patients and families. Testing appears infrequent and this condition often goes unrecognized. Specific consensus guidelines may improve recognition and cancer screening in this population. [Table: see text]

Association of pancreatic adenocarcinoma location (head/body/tail) with DDR mutation status and response to platinum-based therapy

612

Background: Pancreatic adenocarcinoma is an aggressive disease with poor clinical prognosis that can originate from either the head (H) or body/tail (BT). Potential prognostic implications for H versus BT tumors have been reported; however, the molecular underpinnings associated with these differences in survival have not fully been explored. Using a large-scale real-world cohort of H and BT tumors with NGS results available from commercial labs, we retrospectively aim to identify potential differences between H and BT tumors in their response to standard therapies to help understand whether the treatment prioritization for pancreatic adenocarcinoma should take into account anatomical sidedness, as is recognized today with left-sided versus right-sided colorectal cancers. Methods: We analyzed outcomes across 1540 pts with NGS results from Perthera’s Real-World Evidence database who were diagnosed with PDAC originating from the H or BT. Progression-free survival (PFS) was evaluated from initiation of 1st line for advanced disease until discontinuation due to disease progression. Hazard ratios and p-values were computed via Cox regression when comparing PFS between 1st line FOLFIRINOX and gemcitabine/nab-paclitaxel. Differences in frequencies of genomic alterations between proximal and distal were analyzed by Fisher’s exact test. Results: Mutations in BRCA1/ BRCA2/ PALB2 were enriched (unadjusted p-value=0.017) in BT tumors (8.6% of 619) relative to H tumors (5.4% of 921). An expanded set of DDR pathway alterations (e.g. ATM, FANCA, CHEK2, BAP1, BRIP1, etc) were also enriched (unadjusted p-value=0.003) in BT tumors (21.4% of 619) relative to H tumors (15.6% of 921). In BT tumors, mPFS on 1st line FOLFIRINOX was longer (Table) than 1st line gemcitabine/nab-paclitaxel (p=0.0078) but this difference was not observed in H tumors (p=0.34). Overall survival data in these patients and an independent institutional cohort which motivated these analyses will also be discussed. Conclusions: DDR pathway alterations are known predictors of increased benefit from platinums and these real-world insights preliminarily suggest that DDR mutations are more common in BT vs. H. Prospective studies may be warranted to confirm the hypothesis-generating findings that platinum-based regimens should be prioritized in patients with BT tumors while underscoring the importance of routine NGS testing in both BT and H tumors given the prevalence of DDR pathway alterations on both sides of the pancreas.[Table: see text]

Real-world clinical outcomes and molecular features of lung-specific and liver-specific metastases in pancreatic ductal adenocarcinoma (PDAC)

532

Background: PDAC remains one of the most lethal malignancies following metastatic presentation, typically to the liver or lung. Previous studies have observed that advanced PDAC patients have variable outcomes depending on site of involvement. Here, we aim to understand survival outcomes and molecular features for PDAC based on involvement of lung vs liver. Methods: We retrospectively analyzed longitudinal clinical outcomes across 787 patients with PDAC with next generation sequencing (NGS) from Perthera’s Real-World Evidence database whose tumors first metastasized to either the lung or the liver. Median overall survival (mOS) was measured from either the date of initial diagnosis (resectable cases only, stage I-III) or advanced diagnosis (stage IV) until death. Differences in survival and frequencies of mutations were evaluated between patients with lung-specific and liver-specific metastases using Cox regression and Fisher's exact test, respectively. Results: Among resectable PDAC, mOS from initial diagnosis was significantly shorter in patients that developed liver only metastasis (Table, left) compared to those patients that developed lung only metastasis (p=2.4e-08, HR=3.04 [2.06-4.49]). In the advanced PDAC cohort, mOS from diagnosis of advanced disease was also significantly shorter (Table, right) in liver only versus lung only metastasis (p=0.0013, HR=1.62 [1.21-2.18]). Differences in treatment-specific outcomes were not significant supporting a potential prognostic role for lung only metastases. PDAC tumors presenting to the liver first were modestly enriched (unadjusted p<0.05) for TP53 mutations (81.4% in liver vs 69.2% in lung), MYC amplifications (8.6% vs 3.0%), and inactivating CDK2NA alterations (51.5% vs 39.1%) whereas lung-specific mutation frequencies were higher for STK11 mutations (2.4% in liver vs 7.5% in lung), CCND1 amplifications (0.5% vs 3.0%), GNAS alterations (2.0% vs 8.5%). No differences in KRAS mutations nor specific isoforms were noted between lung vs liver only metastasis. Conclusions: Lung only metastasis in both resectable and advanced PDAC confers a significant survival advantage compared to liver only metastasis. Deeper investigation into the molecular drivers of site-specific metastases is warranted.[Table: see text]

Concurrent BRAFV600E and BRCA mutations in microsatellite stable (MSS) metastatic colorectal cancer (mCRC): Prevalence and case series of mCRC (pts) with prolonged overall survival (OS)

3561

Background: BRAF V600E+, MSS mCRC patients comprise up to 10% of advanced CRC. They have a poor prognosis with median survivals typically <1 year. Despite use of multi-agent first-line chemotherapy regimens and combination targeted therapies, outcomes are still poor. In our Institutional Molecular Tumor Board database, we identified 3 consecutive mCRC pts with MSS/ BRAF V600E who also had a BRCA1 or BRCA2 co-mutation and had prolonged overall survival. Prior studies suggested that BRCA mutations are uncommon in CRC and we queried the Foundation Medicine (FM) genomic database to evaluate the prevalence of these cases as well as those with co-mutations in other homologous recombination genes. Methods: 36,966 CRC pts were sequenced by FMI using hybrid capture comprehensive genomic profiling (CGP) to evaluate all classes of genomic alterations (GA) for pathogenic BRAF mutations and/or a mutation in BRCA1/2 or a co-mutation in other homologous recombination (HR) genes ( BARD1, CDK12, FANCL, PALB2, ATM, RAD54L, CHEK2, BRAF, BRIP1, RAD51D, RAD51C, RAD51B, CHEK1). Selected cohort analysis were BRAF V600E co-mutated with BRCA1 and BRCA2, separated into MSI-H and MSS cohorts. The clinicopathological features and genomic loss of heterozygosity (gLOH) of those with a BRAF V600E and a BRCA1/BRCA2 mutation are described along with 3 consecutive cases of CRC patients, identified through the Inova Schar Cancer Institute (ISCI) molecular tumor board (MTB) registry, whom had prolonged OS. Results: Of 36,966 colorectal cancer pts, 6.6% were BRAF V600E+ and 1.5% had any co-occurring HR gene mutation(s) with 0.6% having co-mutations of BRAF V600E and BRCA1/2. BRCA co-mutations were higher in MSI-High BRAF V600E, however 24.1% of these occurred in MSS BRAF V600E. BRCA1 co-mutated was more commonly associated with MSS BRAF V600E and was associated with a higher gLOH than MSI-H BRAF V600E (18.7% vs 2.8%; p <0.001 ). In our institutional MTB database, (3/241;1.2%) CRC patients were MSS, BRAF V600E+ with BRCA1 or BRCA2 co-mutations, one confirmed germline and 2 somatic in origin, and had average gLOH of 21.4% with overall survivals of 72+(alive), 17+(alive), and 30 months, respectively. Conclusions: Co-existence of BRAF V600E/ BRCA1/2 may represent a unique subset of advanced MSS CRC that may have a better prognosis and represent an opportunity to test novel targeted therapies. Larger prospective clinical validation trials in this subset is warranted.[Table: see text]

A Phase II, Single-Arm, Open-Label, Bayesian Adaptive Efficacy and Safety Study of PBI-05204 in Patients with Stage IV Metastatic Pancreatic Adenocarcinoma

LESSONS LEARNED

This trial evaluating a novel plant extract, PBI-05204, did not meet its primary endpoint of overall survival but did show signals of efficacy in heavily pretreated mPDA. PBI-05204 was generally well tolerated, with the most common side effects related to treatment being vomiting (23.7%), nausea (18.4%), decreased appetite (18.4%), and diarrhea (15.8%). Additional trials are needed to explore the role of PBI-05204 in cancer treatment.

BACKGROUND

Survival for metastatic pancreatic ductal adenocarcinoma (mPDA) is dismal, and novel agents are needed. PBI-05204 is a modified supercritical carbon dioxide extract of Nerium oleander leaves. Oleandrin, the extract's major cytotoxic component, is a cardiac glycoside that has demonstrated antitumor activity in various tumor cell lines with a mechanism involving inhibition of Akt phosphorylation and through downregulation of mTOR.

METHODS

A phase II, single-arm, open-label study to determine the efficacy of PBI-05204 in patients with refractory mPDA therapy was conducted. The primary endpoint was overall survival (OS), with the hypothesis that 50% of patients would be alive at 4.5 months. Secondary objectives included safety, progression-free survival (PFS), and overall response rate. Patients received oral PBI-05204 daily until progressive disease (PD), unacceptable toxicity, or patient withdrawal. Radiographic response was assessed every two cycles.

RESULTS

Forty-two patients were enrolled, and 38 were analyzed. Ten patients were alive at 4.5 months (26.3%) with a median PFS of 56 days. One objective response (2.6%) was observed for 162 days. Grade ≥ 3 treatment-emergent adverse events occurred in 63.2% of patients with the most common being fatigue, vomiting, nausea, decreased appetite, and diarrhea.

CONCLUSION

PBI-05204 did not meet its primary endpoint for OS in this study. Recent preclinical data indicate a role for PBI-05204 against glioblastoma multiforme when combined with chemotherapy and radiotherapy. A randomized phase II trial is currently being designed.

Phase II clinical trial of novel agent PBI-05204 in patients with metastatic pancreatic adenocarcinoma (mPDA)

698

Background: Survival statistics for mPDA are dismal and with limited treatment options novel agents are needed to improve disease outcomes. PBI-05204 (Phoenix Biotechnology, Inc., San Antonio, TX) is a modified supercritical carbon dioxide extract of Nerium oleander leaves. Oleandrin, the extract’s major cytotoxic component, has demonstrated anti-tumor activity in various tumor cell lines. In a human PDA orthotopic model, this preparation reduced tumor burden as monotherapy. Pharmacodynamic studies suggest that PBI-05204’s mechanism of action is through inhibition of the PI3k/Akt/mTOR pathway. Methods: A phase II single-arm, open-label study to determine the efficacy of PBI-05204 in patients (pts) with mPDA refractory to standard therapy was conducted. The primary endpoint was overall survival (OS) with the hypothesis that 50% of pts would be alive at 4.5 months. Secondary objectives included safety, progression-free survival (PFS), and overall response rate. Pts received oral PBI-05204 daily until progressive disease (PD), unacceptable toxicity, or pt withdrawal. Radiographic response was assessed every two cycles. Results: Forty-one pts were enrolled; two never received treatment and one was found to have a neuroendocrine tumor after pathological re-evaluation, leaving 38 pts for analysis. Median age at time of enrollment was 65.0 years. The median time from initial diagnosis to treatment was 16.9 months. The primary reason for withdrawal was PD (45.2%). Ten pts were alive at 4.5 months (26.3%) with a mPFS of 56 days (corresponding to first restaging). One objective response (2.6%) was observed for 162 days. Grade ≥3 treatment-emergent adverse events occurred in 63.2% of pts with the most common attributed to drug (all grades) being fatigue (36.8%), vomiting (23.7%), nausea (18.4%), decreased appetite (18.4%), and diarrhea (15.8%). Conclusions: PBI-05204 did not meet its primary endpoint for OS in this study. Recent preclinical data indicate an efficacious role for PBI-05204 against glioblastoma multiforme when combined with chemotherapy, such as temozolomide, and radiotherapy. A randomized Phase II trial is currently being designed. Clinical trial information: NCT02329717.

Activity of RX-3117, an oral antimetabolite nucleoside, in subjects with advanced urothelial cancer: Preliminary results of a phase IIa study

455

Background: RX-3117 is an oral small molecule antimetabolite that is activated by uridine cytidine kinase 2 (UCK2) which is predominantly expressed in cancer cells. RX-3117 has shown efficacy in xenograft models of gemcitabine resistant pancreatic, bladder and colorectal cancer. Preliminary data from an analysis of a Phase 2a clinical study of RX3117 in advanced urothelial cancer is described. Methods: In the Phase 2a study designed to evaluate safety, tolerability and efficacy, subjects were treated with oral RX-3117 (700 mg) once-daily for 5 consecutive days on and 2 days off for 3 of 4 weeks or all 4 weeks in a 28-day cycle. Eligible subjects (aged ≥ 18 years) had relapsed/refractory metastatic urothelial cancer, ECOG PS of 0 to 1, normal organ function (hepatic, renal and hematology) with no limit on the number of prior therapies. The primary Phase 2a endpoints are progression free survival (PFS) and/or objective clinical response with secondary endpoints of safety, TTP, DOR and ORR. Results: As of October 5, 2018; 33 subjects were treated (23 males and 10 females, median age 67.5 years); 29 subjects were evaluable having completed more than 1 cycle of therapy or discontinued due to a related adverse event. Twenty subjects had received 3 or more prior therapies; 30 received gemcitabine/cisplatin and 25 received a checkpoint inhibitor. The most common related adverse events were anemia (G1-2%, G2-3%, G3-3%), fatigue (G1-6%, G2-3%), neutropenia (G2-2%, G3-5%, G4-2%), diarrhea (G1-4%, G2-2%), and thrombocytopenia (G2-2%, G3-3%, G4-1%). One subject had a complete response after 4 cycles of therapy and continues therapy beyond 10 cycles; 5 subjects had PFS ranging from 133 to 315 days. Conclusions: RX-3117 appears to be safe and well-tolerated in chemotherapy and immunotherapy refractory advanced urothelial cancer with acceptable toxicities. Preliminary results show anti-tumor activity in heavily pre-treated patients. Clinical trial information: NCT02030067.

A phase 2 study of prostate specific membrane antigen antibody drug conjugate (PSMA ADC) in patients (pts) with progressive metastatic castration-resistant prostate cancer (mCRPC) following abiraterone and/or enzalutamide (abi/enz)

144

Background: PSMA is a validated target that is overexpressed selectively on prostate cancer cells. PSMA ADC is a fully human IgG1 antibody conjugated to the microtubule disrupting agent MMAE which binds to PSMA-positive cells, inducing cytotoxicity. A phase 1 study showed activity and tolerability at doses from 1.8-2.5 mg/kg. We have enrolled 119 mCRPC pts who progressed following abi/enz in a phase 2 trial of PSMA ADC. Methods: mCRPC pts (83 taxane experienced (TE) and 36 chemo-naïve (CN)) were administered PSMA ADC 2.5 or 2.3 mg/kg IV Q3 wk for up to 8 cycles. 95% of pts received prior abi and/or enz treatment. Safety, antitumor activity (including PSA, CTCs, and tumor imaging) and exploratory biomarkers were assessed. Results: In all treated pts, PSA declines of ≥30% and ≥50% were 30% and 14%, respectively (n=113); CTC counts showed a decline of ≥50% in 78% of pts and conversion from ≥5 to <5 cells/7.5 ml blood in 47% (n=77) at any time during the study. For 2.3 mg/kg pts (n=82), corresponding PSA declines were 35% and 17%; CTC declines of ≥50% were seen in 81% and conversions in 46% (n=54). For CN pts, PSA declines of ≥30% and ≥50% were 31% and 20% (n=35); CTC declines of ≥50% were seen in 89% and conversion in 53% (n=19). Radiologic response by RECIST in 31 pts with measurable target lesions: PR in 4 pts, SD in 19 pts, and PD in 8 pts. Efficacy responses were associated with: low neuroendocrine serum markers (low CgA, low NSE, and high PSA), high PSMA expression (CTCs or tumor tissue). The most common treatment-related AEs ≥CTCAE grade 3 were neutropenia (TE: 25%; CN: 22%), fatigue (20%; 8%), electrolyte imbalance (16%; 11%), anemia (10%; 8%), and neuropathy (8%; 8%). Grade 1-2 neuropathy occurred in 40% (TE) and 50% (CN) of pts. Two 2.5 mg/kg pts (n=34) and one 2.3 mg/kg pt (n=85) died of sepsis. 2.3 mg/kg was better tolerated than 2.5 mg/kg. Conclusions: PSMA ADC was active in abi/enz refractory mCRPC pts. Clinically significant AEs included neutropenia and neuropathy. CTC conversions/reductions, PSA declines, and radiologic evidence of antitumor activity were seen in CN as well as heavily pretreated pts. Clinical trial information: NCT01695044.

A phase II clinical trial of ganetespib (STA-9090) in previously treated patients with advanced esophagogastric cancers

4090

Background: Subsets of esophagogastric (EG) cancers harbor genetic abnormalities, including amplification of HER2 or MET, or mutations in PIK3CA, EGFR, or BRAF. These genes encode clients of the molecular chaperone heat-shock protein 90 (HSP90), and inhibition of HSP90 may promote the degradation of these oncogenic signaling proteins. Ganetespib is a novel triazolone heterocyclic inhibitor of HSP90 that is a biologically rational treatment strategy for advanced EG cancers. Methods: This was a multicenter, single-arm Phase 2 trial. Eligibility: Histologically confirmed advanced EG cancer; progression on ≤ 2 lines of systemic therapy; ECOG PS 0-1. Treatment: Ganetespib 200mg/m2IV on Days 1, 8, and 15 of a 28-day cycle. Primary endpoint: overall response rate (ORR). Results: 26/28 patients enrolled received ≥ 1 dose of drug. The characteristics of the 26 patients were: male 77%, median age 64 years old; ECOG PS 0/1 42/58%; median number of prior therapies 2; esophageal/GEJ/gastric 27/42/31%; prior platinum 92%, prior fluoropyrimidine 88%, prior taxane 38%, prior trastuzumab 15%. Median follow-up was 83 days. The most common drug-related adverse events were: diarrhea (77%), fatigue (65%), elevated ALKP (42%), and elevated AST (38%). The most common Grade 3/4 AEs included: leucopenia (12%), fatigue (12%), diarrhea (8%), and elevated ALKP (8%). 14/26 required ≥ 1 dose modification. 22/26 patients completed at least 2 cycles of ganetespib and were evaluable for response. One complete response was seen, and this patient continues on treatment as of cycle 31 (27.5 mos). Molecular characterization of this patient’s tumor revealed a KRAS mutation in codon 12. The ORR was 1/26 (4%). Two of six patients with HER2-positive disease achieved 12% and 19% tumor reduction from baseline, respectively. TTP was 48 days (1.6 mos) and OS was 83 days (2.8 mos). Conclusions: Ganetespib showed manageable toxicity. While the study was terminated early due to insufficient evidence of single agent activity, the durable CR and 2 minor responses suggest that there may be a subset of EG patients who could benefit from this drug. The molecular determinants of response, however, have yet to be fully characterized. Clinical trial information: CT01167114.

A phase I study of ganetespib in advanced hepatocellular carcinoma (HCC)

259

Background: Ganetespib is an Hsp90 inhibitor that downregulates EGFR, VEGFR, HER2, MET, IGF-IR, and other Hsp90 client proteins involved in hepatocarcinogenesis, thereby making it an attractive therapy for HCC. This multicenter Phase I study was performed to establish the safety, tolerability, recommended Phase 2 dose (RP2D), and preliminary activity of ganetespib in patients with advanced HCC. Methods: Thirteen patients with advanced HCC, Child-Pugh A or B cirrhosis, progression on or intolerance to sorafenib, and ECOG PS ≤ 1 were enrolled in a standard 3x3 dose escalation study at ganetespib doses of 100 mg/m2, 150 mg/m2, and 200 mg/m2 IV given on days 1, 8, and 15 of a 28 day cycle. RECIST 1.1 response was evaluated by CT/MRI every 8 weeks. The primary objective was to determine the RP2D, and secondary objectives included assessments of safety, toxicity, pharmacokinetics, median time to progression (TTP), median progression-free survival (PFS), median overall survival (OS), and objective response rate (ORR). Results: Twelve of the 13 patients enrolled received study drug, and enrollment is ongoing for the 200 mg/m2 cohort. Of the 12 patients: male 66%; median age 57 years; median number of prior treatments 2; Asian 33%; HCC etiology (HBV 41.7%, HCV 41.7%, hemachromatosis 8.3%, unknown 16.7%); median baseline AFP 115.3 ng/mL. Median TTP for the 10 evaluable patients was 49 days (1.6 months). No responses were seen, but 2/10 (20.0%) patients had stable disease at 8 weeks. AFP response, defined as reduction from baseline of >50% in patients with an elevated baseline AFP, was seen in 0% of patients. Most common AEs: diarrhea (100%), AST elevation (58.3%), hyperglycemia (58.3%), and fatigue (58.3%). Most common Gr 3/4 AEs: hyperglycemia (25%), anemia (16.7%), lipasemia (16.7%), and ALKP elevation (16.7%). One (8.3%) patient had a fatal AE, septic shock, within 30 days of receiving the drug. One DLT was observed: Gr 3 lipasemia at the 100mg/m2 dose. Conclusions: Ganetespib had a manageable safety profile and demonstrated limited efficacy in patients with advanced HCC. Determination of the R2PD, further assessment of clinical efficacy, and analysis of molecular markers are still pending, and a follow-up Phase II study will be considered based on this data. Clinical trial information: NCT01665937.

Phase I/II study of preoperative (pre-op) short course chemoradiation (CRT) with proton beam therapy (PBT) and capecitabine (cape) followed by early surgery for resectable pancreatic ductal adenocarcinoma (PDAC) of the head

4021

Background: Standard adjuvant 6 week CRT may delay and reduce tolerability of adjuvant gemcitabine-based chemotherapy. We explore the safety and efficacy of a one-week course of pre-op CRT with PBT and cape followed by early pancreaticoduodenectomy (PD). Methods: Patients with radiographically resectable, biopsy-proven PDAC of the head were enrolled from May 2007-December 2010 on this prospective, NCI-sponsored clinical trial. Eligibility included no CT involvement of SMA or celiac artery; No metastatic disease. Dose level 1 consisted of PBT delivered 3 Gy x 10. Pts in subsequent dose levels received 5 Gy x 5 in progressively shortened schedules: level 2 (wk 1 M W F, wk 2 T Th), level 3 (wk 1 M T Th F, wk 2 M), level 4 (wk 1 M-F). PBT was targeted at pancreatic mass with elective nodal coverage. Pts received Cape 825 mg/m2 BID wk 1 and 2 M-F. Surgery was performed 1-6 wks after completion of cape. Patients were recommended to receive 6 mo of gemcitabine after surgery. Genotyping of 15 genes (including KRAS, PIK3CA, BRAF, NRAS, TP53, IDH1) was performed. Results: 50 pts were enrolled on study. 48 patients are eligible for this analysis. 3 pts were treated at each of dose levels 1-3. 6 pts were at dose level 4, which was selected as MTD. No DLTs were observed. 35 patients were treated in the phase II portion at the MTD. Gr 3 toxicity was noted in 2 pts (chest wall pain-1, colitis-1). 38 pts underwent PD. Reasons for no PD were: metastatic disease-9, unresectable tumor-1. Mean post-op length of stay was 7 days (range 5-47). 6/38 (16%) resected pts had positive margins. 28/38 (74%) had positive nodes. Median follow up is 21 months among the 19 patients still alive. 4/48 (8%) local failures in ALL patients. mOS and mPFS are 18 and 10 months respectively for ALL patients and 27 and 14 months for RESECTED patients. Of 28 patients with genotype data available, 22 (79%) had KRAS mutations. 10/22 KRAS mt and 4/6 KRAS wt pts are alive. Conclusions: Pre-op CRT with 1 wk of PBT and capecitabine followed by early surgery is well tolerated and associated with favorable local control. Complete genotype and DPC4 data will be presented at the meeting.

Immediate versus delayed adjuvant chemoradiation for resected pancreatic cancer: An analysis of local control and survival

301

Background: Patients with resected pancreatic cancer have a significant risk of both local and distant failures. Because of the importance of chemotherapy (CT) highlighted by recent trials, the ongoing RTOG 0848 delays chemoradiation (CRT) randomization to after 5 cycles of chemotherapy (CT). However, the impact of delaying CRT on LR is unknown. We evaluated the patterns of failure and outcomes for patients treated with immediate vs. delayed CRT.

Methods: 174 consecutive patients with resectable pancreas cancer who received post-operative CRT at the Massachusetts General Hospital and Brigham & Women’s Hospital/Dana-Farber Cancer Institute between 1998-2010 were retrospectively reviewed on an IRB approved protocol. CRT was delivered to 50.4 Gy with concurrent 5FU or capecitabine. Patient baseline characteristics, pathologic features, and CA19-9 were obtained. Patients were divided into immediate CRT vs. delayed CRT, defined as receiving any CT prior to CRT. LR was independently re-reviewed and confirmed. LR, PFS, and OS were calculated using the Kaplan-Meier method and groups compared using the log-rank test.

Results: Median age was 62 (range, 34-83), 75 were male, and ECOG PS was 0-1 in 95% of patients. Median tumor size was 3 cm, 67% had positive nodes, and 33% had positive margins. Mean pre-operative and post-operative CA 19-9 was 813 U/mL and 19 U/mL, respectively. 123/174 (72%) patients received immediate CRT, of whom 101 received additional chemotherapy (5FU or gemcitabine (GEM)). 51/174 (28%) received delayed CRT, and received a median of 4 cycles of GEM prior to CRT. Median follow-up was 33 months (range, 3.6-67). 6/51 (12%) delayed CRT had LR prior to initiation of CRT. 25/51 in the delayed CRT group had a LR (49%) vs. 35/124 patients (28%) in the immediate CRT group. The 1-year LR rates in the immediate and delayed CRT groups were 18% and 40%, respectively (p=0.0093). There was no difference between immediate or delayed CRT, respectively, in mPFS (12.8 mo vs. 12.2 mo, p=0.6544) or mOS (24.8 mo vs. 26.7 mo, p=0.7163).

Conclusions: Delayed radiation is associated with an increased risk of LR, though this did not appear to impact PFS or OS.