Phenotyping Hepatic Immune-Related Adverse Events in the Setting of Immune Checkpoint Inhibitor Therapy

PURPOSE

We present and validate a rule-based algorithm for the detection of moderate to severe liver-related immune-related adverse events (irAEs) in a real-world patient cohort. The algorithm can be applied to studies of irAEs in large data sets.

METHODS

We developed a set of criteria to define hepatic irAEs. The criteria include: the temporality of elevated laboratory measurements in the first 2-14 weeks of immune checkpoint inhibitor (ICI) treatment, steroid intervention within 2 weeks of the onset of elevated laboratory measurements, and intervention with a duration of at least 2 weeks. These criteria are based on the kinetics of patients who experienced moderate to severe hepatotoxicity (Common Terminology Criteria for Adverse Events grades 2-4). We applied these criteria to a retrospective cohort of 682 patients diagnosed with hepatocellular carcinoma and treated with ICI. All patients were required to have baseline laboratory measurements before and after the initiation of ICI.

RESULTS

A set of 63 equally sampled patients were reviewed by two blinded, clinical adjudicators. Disagreements were reviewed and consensus was taken to be the ground truth. Of these, 25 patients with irAEs were identified, 16 were determined to be hepatic irAEs, 36 patients were nonadverse events, and two patients were of indeterminant status. Reviewers agreed in 44 of 63 patients, including 19 patients with irAEs (0.70 concordance, Fleiss' kappa: 0.43). By comparison, the algorithm achieved a sensitivity and specificity of identifying hepatic irAEs of 0.63 and 0.81, respectively, with a test efficiency (percent correctly classified) of 0.78 and outcome-weighted F1 score of 0.74.

CONCLUSION

The algorithm achieves greater concordance with the ground truth than either individual clinical adjudicator for the detection of irAEs.

Baseline correlates of frailty and its association with survival in United States veterans with acute myeloid leukemia

Frailty is an important construct to measure in acute myeloid leukemia (AML). We used the Veterans Affairs Frailty Index (VA-FI) - calculated using readily available data within the VA's electronic health records - to measure frailty in U.S. veterans with AML. Of the 1166 newly diagnosed and treated veterans with AML between 2012 and 2022, 722 (62%) veterans with AML were classified as frail (VA-FI > 0.2). At a median follow-up of 252.5 days, moderate-severely frail veterans had significantly worse survival than mildly frail, and non-frail veterans (median survival 179 vs. 306 vs. 417 days, p < .001). Increasing VA-FI severity was associated with higher mortality. A model with VA-FI in addition to the European LeukemiaNet (ELN) risk classification and other covariates statistically outperformed a model containing the ELN risk and other covariates alone (p < .001). These findings support the VA-FI as a tool to expand frailty measurement in research and clinical practice for informing prognosis in veterans with AML.

Machine learning-based natural language processing to extract PD-L1 expression levels from clinical notes

Introduction: PD-L1 expression is used to determine oncology patients' response to and eligibility for immunologic treatments; however, PD-L1 expression status often only exists in unstructured clinical notes, limiting ability to use it in population-level studies. Methods: We developed and evaluated a machine learning based natural language processing (NLP) tool to extract PD-L1 expression values from the nationwide Veterans Affairs electronic health record system. Results: The model demonstrated strong evaluation performance across multiple levels of label granularity. Mean precision of the overall PD-L1 positive label was 0.859 (sd, 0.039), recall 0.994 (sd, 0.013), and F1 0.921 (0.024). When a numeric PD-L1 value was identified, the mean absolute error of the value was 0.537 on a scale of 0 to 100. Conclusion: We presented an accurate NLP method for deriving PD-L1 status from clinical notes. By reducing the time and manual effort needed to review medical records, our work will enable future population-level studies in cancer immunotherapy.

Prognostic value of the veterans affairs frailty index in older patients with non-small cell lung cancer

BACKGROUND

Older patients with non-small cell lung cancer (NSCLC) are a heterogeneous population with varying degrees of frailty. An electronic frailty index such as the Veterans Affairs Frailty Index (VA-FI) can potentially help identify vulnerable patients at high risk of poor outcomes.

METHODS

NSCLC patients ≥65 years old and diagnosed in 2002-2017 were identified using the VA Central Cancer Registry. The VA-FI was calculated using administrative codes from VA electronic health records data linked with Medicare and Medicaid data. We assessed associations between the VA-FI and times to mortality, hospitalization, and emergency room (ER) visit following diagnosis by Kaplan-Meier analysis and multivariable stratified Cox models. We also evaluated the change in discrimination and calibration of reference prognostic models after adding VA-FI.

RESULTS

We identified a cohort of 42,204 older NSCLC VA patients, in which 55.5% were classified as frail (VA-FI >0.2). After adjustment, there was a strong association between VA-FI and the risk of mortality (HR = 1.23 for an increase of four deficits or, equivalently, an increase of 0.129 on VA-FI, p < 0.001), hospitalization (HR = 1.16 for four deficits, p < 0.001), and ER visit (HR = 1.18 for four deficits, p < 0.001). Adding VA-FI to baseline prognostic models led to statistically significant improvements in time-dependent area under curves and did not have a strong impact on calibration.

CONCLUSION

Older NSCLC patients with higher VA-FI have significantly elevated risks of mortality, hospitalizations, and ER visits following diagnosis. An electronic frailty index can serve as an accessible tool to identify patients with vulnerabilities to inform clinical care and research.

Association of COVID-19 Vaccination With SARS-CoV-2 Infection in Patients With Cancer: A US Nationwide Veterans Affairs Study

Question

What is the effectiveness of SARS-CoV-2 vaccination in patients with cancer?

Findings

In this cohort study of US Veterans Affairs patients who received systemic therapy for cancer between August 15, 2010, and May 4, 2021, a proxy measure for effectiveness of the vaccine starting 14 days after the second dose was 58%. The measure of effectiveness starting 14 days after the second dose was 85% in patients who had not received systemic therapy within the 6 months prior to vaccination and 76% among those receiving hormonal treatment.

Meaning

Results suggest that SARS-CoV-2 vaccination associated with lower infection rates in patients with cancer, especially in those not receiving current systemic therapy and those receiving hormonal treatment.

Importance

Patients with cancer are at increased risk for severe COVID-19, but it is unknown whether SARS-CoV-2 vaccination is effective for them.

Objective

To determine the association between SARS-CoV-2 vaccination and SARS-CoV-2 infections among a population of Veterans Affairs (VA) patients with cancer.

Design, Setting, and Participants

Retrospective, multicenter, nationwide cohort study of SARS-CoV-2 vaccination and infection among patients in the VA health care system from December 15, 2020, to May 4, 2021. All adults with solid tumors or hematologic cancer who received systemic cancer-directed therapy from August 15, 2010, to May 4, 2021, and were alive and without a documented SARS-CoV-2 positive result as of December 15, 2020, were eligible for inclusion. Each day between December 15, 2020, and May 4, 2021, newly vaccinated patients were matched 1:1 with unvaccinated or not yet vaccinated controls based on age, race and ethnicity, VA facility, rurality of home address, cancer type, and treatment type/timing.

Exposures

Receipt of a SARS-CoV-2 vaccine.

Main Outcomes and Measures

The primary outcome was documented SARS-CoV-2 infection. A proxy for vaccine effectiveness was defined as 1 minus the risk ratio of SARS-CoV-2 infection for vaccinated individuals compared with unvaccinated controls.

Results

A total of 184 485 patients met eligibility criteria, and 113 796 were vaccinated. Of these, 29 152 vaccinated patients (median [IQR] age, 74.1 [70.2-79.3] years; 95% were men; 71% were non-Hispanic White individuals) were matched 1:1 to unvaccinated or not yet vaccinated controls. As of a median 47 days of follow-up, 436 SARS-CoV-2 infections were detected in the matched cohort (161 infections in vaccinated patients vs 275 in unvaccinated patients). There were 17 COVID-19–related deaths in the vaccinated group vs 27 COVID-19–related deaths in the unvaccinated group. Overall vaccine effectiveness in the matched cohort was 58% (95% CI, 39% to 72%) starting 14 days after the second dose. Patients who received chemotherapy within 3 months prior to the first vaccination dose were estimated to have a vaccine effectiveness of 57% (95% CI, –23% to 90%) starting 14 days after the second dose vs 76% (95% CI, 50% to 91%) for those receiving endocrine therapy and 85% (95% CI, 29% to 100%) for those who had not received systemic therapy for at least 6 months prior.

Conclusions and Relevance

In this cohort study, COVID-19 vaccination was associated with lower SARS-CoV-2 infection rates in patients with cancer. Some immunosuppressed subgroups may remain at early risk for COVID-19 despite vaccination, and consideration should be given to additional risk reduction strategies, such as serologic testing for vaccine response and a third vaccine dose to optimize outcomes.

Multi-scale Pan-cancer Integrative Analyses Identify the STAT3-VSIR Axis as a key Immunosuppressive Mechanism in Head and Neck Cancer

PURPOSE

VSIR is a novel immune checkpoint protein whose expression on tumor cells across cancers remains largely uncharacterized. Here we purposed to decode the pan-cancer biologic and clinical significance of VSIR over-expression in the tumor compartment.

EXPERIMENTAL DESIGN

We performed multi-omics integrative analyses of 9735 tumor samples to identify cancers with non-leukocytic expression of VSIR (VSIR-High), followed by association with overall survival and immune cell infiltration levels. Orthogonal assessments of VSIR protein expression and lymphocytic infiltration were performed using Quantitative immunofluorescence.

RESULTS

Integrative modeling identified a subset of cancer types as being enriched for VSIR-High tumors. VSIR-High tumors were associated with significantly poorer overall survival in immunogenic Ovarian Serous Adenocarcinoma (SA) and Oral Cavity Squamous Cell Carcinoma (SCC). QIF assessments in an independent validation cohort confirmed over-expression of VSIR as being associated with poorer overall survival within immunogenic Oral Cavity SCC. VSIR over-expression was associated with lower CD4 Helper T-cell infiltration in both Ovarian SA and Oral Cavity SCC, but did not impact CD8 T-cell infiltration. VSIR over-expressing tumors in both cancer types exhibited significantly higher STAT3 signaling activity. Pharmacologic inhibition of STAT3 signaling resulted in dose-dependent reduction of VSIR expression in Ovarian SA & Oral Cavity SCC cells.

CONCLUSIONS

The STAT3-VSIR axis is a potentially significant immuno-modulatory mechanism in oral cavity and ovarian cancers, whose activation is associated with poorer survival and an immune microenvironment marked by decreased CD4 helper T-cell activity. The role of VSIR as a tumor-intrinsic modulator of resistance to immunotherapy warrants further exploration.

Prevalence and Outcome of COVID-19 Infection in Cancer Patients: A National Veterans Affairs Study

Background

Emerging data suggest variability in susceptibility and outcome to coronavirus disease 2019 (COVID-19) infection. Identifying risk factors associated with infection and outcomes in cancer patients is necessary to develop healthcare recommendations.

Methods

We analyzed electronic health records of the US Veterans Affairs Healthcare System and assessed the prevalence of COVID-19 infection in cancer patients. We evaluated the proportion of cancer patients tested for COVID-19 who were positive, as well as outcome attributable to COVID-19, and stratified by clinical characteristics including demographics, comorbidities, cancer treatment, and cancer type. All statistical tests are 2-sided.

Results

Of 22 914 cancer patients tested for COVID-19, 1794 (7.8%) were positive. The prevalence of COVID-19 was similar across age. Higher prevalence was observed in African American (15.0%) compared with White (5.5%; P < .001) and in patients with hematologic malignancy compared with those with solid tumors (10.9% vs 7.8%; P < .001). Conversely, prevalence was lower in current smokers and patients who recently received cancer therapy (<6 months). The COVID-19–attributable mortality was 10.9%. Higher attributable mortality rates were observed in older patients, those with higher Charlson comorbidity score, and in certain cancer types. Recent (<6 months) or past treatment did not influence attributable mortality. Importantly, African American patients had 3.5-fold higher COVID-19–attributable hospitalization; however, they had similar attributable mortality as White patients.

Conclusion

Preexistence of cancer affects both susceptibility to COVID-19 infection and eventual outcome. The overall COVID-19–attributable mortality in cancer patients is affected by age, comorbidity, and specific cancer types; however, race or recent treatment including immunotherapy do not impact outcome.

Real-world usage and outcomes for patients treated with immune checkpoint inhibitors across multiple treatment indications in the Veterans Affairs system

e15063

Background: Increasingly broad patient groups are being treated with immune checkpoint inhibitors (ICIs) in clinical practice, but few studies have assessed their usage and outcomes in large, comprehensive real-world cohorts. In this study, we identified patients who received ICI therapy through April 2019 in the Veterans Affairs (VA) system. We described patient characteristics and assessed survival outcomes for patients across multiple treatment indications based on the cancer type and line of therapy. Methods: We conducted a retrospective analysis using electronic health record data captured from VA facilities nationwide. Patients treated with ICI were identified based on pharmacy and orders records. The associated cancer type and line of therapy (first line 1L or later line 2L+) was ascertained based on data on patients’ treatment history, administrative codes, and VA cancer registry records. The distribution of time from ICI treatment initiation to death was separately assessed by treatment indication using Kaplan-Meier estimates. Results: We identified 13,276 patients treated with ICI. The mean age was 69 years old. A majority of patients were male (97.2%) and white (76.9%), and a substantial minority were African-American (16.9%). Among patients treated with ICI for an identifiable cancer type, the most common cancer types included non-small lung cancer (NSCLC; 47.2%), melanoma (22.2%), renal cell carcinoma (RCC; 6.7%), squamous cell carcinoma of the head and neck (SCCHN; 10.4%), and urothelial cancer (6.4%). For major treatment indications, VA patients generally exhibited lower rates of survival than those reported in many ICI clinical trials, but the results are more comparable to those in trials among a subgroup of non-frail patients. Conclusions: This study describes ICI utilization and outcomes across multiple tumor types in a real-world population at the VA. The results offer evidence on the utilization and performance of ICIs in a real-world population. In addition, this work establishes a platform for further analyses of ICI outcomes in real-world practice at the VA. [Table: see text]

Predictive value of an electronic frailty index (FI) in U.S. Veterans with newly diagnosed non-small cell lung cancer (NSCLC)

11545

Background: Frailty is a well-established predictor of survival, but assessing it in a real-world healthcare setting is often time-consuming and impractical. We sought to determine if a quantitative measure of frailty, computed automatically from electronic health records using the cumulative deficit model, also predicts overall survival in patients with NSCLC. Methods: We conducted a retrospective cohort study of Veterans identified to have NSCLC in the VA Precision Oncology Data Repository. Frailty was assessed using a previously validated FI, the Frailty Index for Veterans Affairs (VA-FI), consisting of 31 deficits. We categorized non-frail patients into non-frail (0-0.1) and pre-frail (0.1-0.2) groups, and frail patients into mild (0.2-0.3), moderate (0.3-0.4) and severe ( > 0.4) groups. Associations of frailty and survival were assessed using Kaplan-Meier estimates and Cox models. The added value of frailty status for predictive performance was quantified using the continuous net-reclassification index (NRI). Results: The cohort included 359 patients with NSCLC. The mean age was 66 years, and 96.1% were male, including 76.3% Caucasian and 16.2% African-American. The majority had an advanced stage cancer (15.9% and 44.3% with stages III and IV respectively) with minority falling into the early stage category (15.3% and 6.1% with stages I and II respectively). A substantial proportion of patients were identified to be frail (30%). The frailty categories differentiated survival profiles of patients with 2-year survival ranging from 0.41 for non-frail patients to 0.08 for the severely frail patients (log-rank p < 0.001). Frail patients exhibited significantly shorter survival than non-frail group even after adjusting for age, gender, and stage in a multivariate Cox model (hazard ratio 1.7, p < 0.001). Inclusion of frailty status in the Cox model significantly improved 2-year risk estimates of survival (NRI 0.35; 95% CI 0.27–0.47). Conclusions: Frailty as determined by VA-FI is a significant predictor of survival independent of stage and demographic factors among NSCLC patients in this VA cohort. VA-FI is an automated and a practically feasible tool to better estimate life expectancy and help individualize patient care.

A predictive model for survival in non-small cell lung cancer (NSCLC) based on electronic health record (EHR) and tumor sequencing data at the Department of Veterans Affairs (VA)

109

Background: Machine learning tools based on EHR data hold promise to help avoid unnecessary risks associated with lung cancer and its treatment. Additionally, molecular genetic profiling is becoming an integral tool for clinicians to individualize treatment for lung cancer. However, relatively few survival models have been built that integrate this data in individualized predictive models. Here, we combine real-world EHR and tumor sequencing data from the VA Precision Oncology Data Repository (PODR) to build accurate individualized survival predictions in newly-diagnosed NSCLC patients. Methods: We identified a cohort of 356 VA patients newly diagnosed with NSCLC for whom EHR, cancer registry, and targeted tumor sequencing data is available in PODR. We defined 41 features reflecting 15 baseline clinical and demographic characteristics from the EHR and registry, such as age, race, stage, histology, and therapy. We also defined features reflecting 206 clinically actionable somatic variants. We selected 5 important variants for inclusion in the model, as well as the total number of mutations. We trained a random forests algorithm to predict 1-year survival. Precision, recall, and area under the ROC curve (AUC) were assessed using 5-fold cross validation. Results: Mean age at diagnosis was 66 years. The majority of patients had late stage disease (15% stage I, 6% II, 15% III, 44% IV), and 59% of patients received systemic therapy. 45% died within 1 year of diagnosis, and 55% survived past 1 year. Our predictive model for 1-year survival achieves strong results. Cross-validated AUC is 0.79 (SD 0.08), precision is 0.79 (SD 0.07), recall is 0.74 (SD 0.07), suggesting that the trained model combining clinical and genomic features is effective at predicting 1-year survival. Conclusions: By integrating real-world EHR and sequencing data, we built a highly accurate predictive model of 1-year survival in NSCLC patients at the VA. Such a model, after ongoing validation in a larger cohort, offers the ability to make individualized predictions that could inform patient care to improve outcomes.

Endocrinopathies associated with immune checkpoint inhibitors: Standard of care at veterans administration medical centers

e14148

Background: Multiple different immune checkpoint inhibitors (ICI) have now received FDA approval for nearly 70 separate indications covering 14 different tumor types. Patients treated with these agents in clinical trials have an observed incidence of immune related adverse events (irAEs), including endocrinopathies, which may increase morbidity and mortality. Limited information describes the incidence and impact of these events outside of clinical trials. Methods: Retrospective data from the Veterans Health Administration (VA) of patients treated with ICI has been aggregated to understand the impact of these events in a standard of care setting, with a goal of improving patient care through predictive models and contributing to the understanding of the mechanisms and response to treatment. Results: Between October 2015 and December 2018, 10,280 patients were prescribed ICI at VA medical centers, with an average age of 70 years (range 20-99). A total of 11098 ICI orders, allowing for combinations or sequential treatments. Overall, nivolumab was prescribed 6024 times (54.3%), pembrolizumab 3976 (35.8%), ipilimumab 565 (5.1%) and atezolizumab 519 (4.6%). Avelumab (13) and durvalumab (1) had limited use. A candidate set of potential endocrine adverse events was estimated based on selected ICD10 codes recorded for the first time after treatment with ICI (Table). Conclusions: The frequency of endocrine immune related adverse events has been reported to be 5-10%. Here we have identified a cohort of ICI treated patients who may have developed endocrine adverse events. This cohort will be used to evaluate phenotyping, potential biomarkers and models of predictive risk.[Table: see text]

Phase 1 trial of CA-170, a novel oral small molecule dual inhibitor of immune checkpoints PD-1 and VISTA, in patients (pts) with advanced solid tumor or lymphomas

TPS3099

Background: Programmed-death 1 (PD-1) and V-domain Ig suppressor of T-cell activation (VISTA) are independent immune checkpoints that negatively regulate T-cell function and are implicated in various malignancies. Preclinical studies have demonstrated that dual blockade of these pathways is synergistic. CA-170 is a first-in-class oral small molecule that directly targets both PD-1/PD-L1 and VISTA pathways and has shown anti-tumor activity in multiple preclinical models. Methods: The dose escalation phase has a target enrollment of 50 pts with advanced solid tumors or lymphomas onto escalating doses; the first four single-pt cohorts are accelerated titration but then switch to 3+3 design. The dose expansion phase has a target enrollment of 250 pts with select tumor types known to be responsive to anti-PD-1/L1 inhibitors and/or known to express PD-L1 or VISTA. Key eligibility criteria include: age ≥ 18 years, ECOG ≤1, adequate organ function, and ineligible for/did not respond to standard therapy including anti-PD-1/L1 inhibitors, where available. Primary objectives of this first-in-human study: safety, maximum tolerated dose, and recommended phase 2 dose. Secondary objectives: pharmacokinetics (PK) and anti-tumor activity. Exploratory endpoints: biomarkers and pharmacodynamic (PD) effects, which include changes in immune cell and peripheral cytokine populations in tumor (IHC/mRNA) and blood (flow cytometry/mRNA). Oral CA-170 is administered once daily in 21-day cycles. Response will be evaluated every other cycle per RECIST (v1.1) and Immune-related Response Criteria or by Cheson criteria (2007). Patients who discontinue treatment for reasons other than progressive disease will be followed for progression-free survival. Serial plasma, blood, and tumor samples will be collected for PK and PD evaluation. Clinical trial identifier: Clinical trial information: NCT02812875.

Dual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC and Suppresses Growth of MYC-dependent Cancers

Upregulation of MYC is a common driver event in human cancers, and some tumors depend on MYC to maintain transcriptional programs that promote cell growth and proliferation. Preclinical studies have suggested that individually targeting upstream regulators of MYC, such as histone deacetylases (HDAC) and phosphoinositide 3-kinases (PI3K), can reduce MYC protein levels and suppress the growth of MYC-driven cancers. Synergy between HDAC and PI3K inhibition in inducing cancer cell death has also been reported, but the involvement of MYC regulation is unclear. In this study, we demonstrated that HDAC and PI3K inhibition synergistically downregulates MYC protein levels and induces apoptosis in "double-hit" (DH) diffuse large B-cell lymphoma (DLBCL) cells. Furthermore, CUDC-907, a small-molecule dual-acting inhibitor of both class I and II HDACs and class I PI3Ks, effectively suppresses the growth and survival of MYC-altered or MYC-dependent cancer cells, such as DH DLBCL and BRD-NUT fusion-positive NUT midline carcinoma (NMC) cells, and MYC protein downregulation is an early event induced by CUDC-907 treatment. Consistently, the antitumor activity of CUDC-907 against multiple MYC-driven cancer types was also demonstrated in animal models, including DLBCL and NMC xenograft models, Myc transgenic tumor syngeneic models, and MYC-amplified solid tumor patient-derived xenograft (PDX) models. Our findings suggest that dual function HDAC and PI3K inhibitor CUDC-907 is an effective agent targeting MYC and thus may be developed as potential therapy for MYC-dependent cancers. Mol Cancer Ther; 16(2); 285-99. ©2016 AACR.

Estrogen Drives Cellular Transformation and Mutagenesis in Cells Expressing the Breast Cancer-Associated R438W DNA Polymerase Lambda Protein

Repair of DNA damage is critical for maintaining the genomic integrity of cells. DNA polymerase lambda (POLL/Pol λ) is suggested to function in base excision repair (BER) and nonhomologous end-joining (NHEJ), and is likely to play a role in damage tolerance at the replication fork. Here, using next-generation sequencing, it was discovered that the POLL rs3730477 single-nucleotide polymorphism (SNP) encoding R438W Pol λ was significantly enriched in the germlines of breast cancer patients. Expression of R438W Pol λ in human breast epithelial cells induces cellular transformation and chromosomal aberrations. The role of estrogen was assessed as it is commonly used in hormone replacement therapies and is a known breast cancer risk factor. Interestingly, the combination of estrogen treatment and the expression of the R438W Pol λ SNP drastically accelerated the rate of transformation. Estrogen exposure produces 8-oxoguanine lesions that persist in cells expressing R438W Pol λ compared with wild-type (WT) Pol λ-expressing cells. Unlike WT Pol λ, which performs error-free bypass of 8-oxoguanine lesions, expression of R438W Pol λ leads to an increase in mutagenesis and replicative stress in cells treated with estrogen. Together, these data suggest that individuals who carry the rs3730477 POLL germline variant have an increased risk of estrogen-associated breast cancer.

IMPLICATIONS

The Pol λ R438W mutation can serve as a biomarker to predict cancer risk and implicates that treatment with estrogen in individuals with this mutation may further increase their risk of breast cancer. Mol Cancer Res; 14(11); 1068-77. ©2016 AACR.

Abstract 4634: Novel dual HDAC & PI3K inhibitor, CUDC-907, for MYC-driven malignancies

MYC family genes are among the most frequently deregulated oncogenic drivers in human cancer. Pharmacologic inhibition of HDAC activity and blockade of the PI3K pathway have both been shown to suppress MYC-induced transcription.

HDAC activity is critical for MYC gene regulation, as MYC represses transcription of target genes through recruitment of HDACs. HDAC inhibitors have been shown to restore expression of genes suppressed by MYC family members and to induce rapid downregulation of expression of MYC itself.

The PI3K pathway plays a central role in regulating MYC at the post-transcriptional level. Activation of PI3K signaling leads to activation of AKT, which phosphorylates and inhibits GSK3β. As GSK3β normally phosphorylates MYC which facilitates the degradation of MYC, activation of PI3K signaling leads to increased stability of MYC, whereas PI3K inhibitors decrease MYC stability. A recent study has demonstrated addiction to MYC signaling and hypersensitivity to PI3K inhibition in PTEN-deficient diffuse large B-cell (DLBCL) cell lines, suggesting that MYC-driven cancers may be particularly sensitive to PI3K inhibition. As HDACs and PI3K regulate MYC protein levels and functions through nonoverlapping mechanisms, simultaneous HDAC and PI3K inhibition may further enhance MYC suppression.

CUDC-907 is an orally bioavailable, small-molecule dual HDAC and PI3K inhibitor that primarily inhibits class I and II HDACs and the PI3Kα, β, and δ isoforms. CUDC-907 shows greater anti-tumor activity in vitro than single-target HDAC or PI3K inhibitors, especially in MYC-dependent cell types, such as DLBCL and NUT midline carcinoma (NMC). In preclinical testing, CUDC-907 treatment leads to a dose-dependent decrease in MYC protein levels, and is also more potent in decreasing MYC than the HDAC inhibitor panobinostat and the pan-PI3K inhibitor pictilisib alone or in combination. Significant antitumor effects have been consistently observed in MYC-driven DLBCL xenograft and genetically engineered mouse models exposed to CUDC907. In particular, certain MYC translocation (Daudi), double-hit (concurrent MYC and BLC2 translocation, WSUDLCL2 and DOHH2), double-expresser (expression of MYC and BCL2 proteins, U2932) xenograft models, and the Eμ-Myc transgenic mouse model achieve tumor growth inhibition of 100%, 69%, 56%, 97% and 72%, respectively.

These findings raise the possibility that hematologic and solid tumors driven by aberrant overexpression of MYC family genes (e.g., MYC-altered DLBCL and NMC) might be more responsive to simultaneous HDAC and PI3K inhibition with CUDC-907 than they are to single-target therapy. Clinical Phase 1 studies are currently testing CUDC-907 in patients with relapsed/refractory (R/R) DLBCL and advanced MYC-aberrant solid tumors. Preliminary data are encouraging and support the planned Phase 2 study in R/R MYC-altered DLBCL, as well as further testing in other MYC-driven malignancies.

Citation Format: Kaiming Sun, Ruzanna Atoyan, Mylissa A. Borek, Steven Dellarocca, Maria E. Samson, Anna W. Ma, Guangxin Xu, Troy Patterson, David P. Tuck, Jaye L. Viner, Ali Fattaey, Jing Wang. Novel dual HDAC & PI3K inhibitor, CUDC-907, for MYC-driven malignancies. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4634.

Molecular phenotypes in triple negative breast cancer from African American patients suggest targets for therapy

Triple negative breast cancer (TNBC) is characterized by high proliferation, poor differentiation and a poor prognosis due to high rates of recurrence. Despite lower overall incidence African American (AA) patients suffer from higher breast cancer mortality in part due to the higher proportion of TNBC cases among AA patients compared to European Americans (EA). It was recently shown that the clinical heterogeneity of TNBC is reflected by distinct transcriptional programs with distinct drug response profiles in preclinical models. In this study, gene expression profiling and immunohistochemistry were used to elucidate potential differences between TNBC tumors of EA and AA patients on a molecular level. In a retrospective cohort of 136 TNBC patients, a major transcriptional signature of proliferation was found to be significantly upregulated in samples of AA ethnicity. Furthermore, transcriptional profiles of AA tumors showed differential activation of insulin-like growth factor 1 (IGF1) and a signature of BRCA1 deficiency in this cohort. Using signatures derived from the meta-analysis of TNBC gene expression carried out by Lehmann et al., tumors from AA patients were more likely of basal-like subtypes whereas transcriptional features of many EA samples corresponded to mesenchymal-like or luminal androgen receptor driven subtypes. These results were validated in The Cancer Genome Atlas mRNA and protein expression data, again showing enrichment of a basal-like phenotype in AA tumors and mesenchymal subtypes in EA tumors. In addition, increased expression of VEGF-activated genes together with elevated microvessel area determined by the AQUA method suggest that AA patients exhibit higher tumor vascularization. This study confirms the existence of distinct transcriptional programs in triple negative breast cancer in two separate cohorts and that these programs differ by racial group. Differences in TNBC subtypes and levels of tumor angiogenesis in AA versus EA patients suggest that targeted therapy choices should be considered in the context of race.

PAM50 proliferation score as a predictor of weekly paclitaxel benefit in breast cancer

To identify a group of patients who might benefit from the addition of weekly paclitaxel to conventional anthracycline-containing chemotherapy as adjuvant therapy of node-positive operable breast cancer. The predictive value of PAM50 subtypes and the 11-gene proliferation score contained within the PAM50 assay were evaluated in 820 patients from the GEICAM/9906 randomized phase III trial comparing adjuvant FEC to FEC followed by weekly paclitaxel (FEC-P). Multivariable Cox regression analyses of the secondary endpoint of overall survival (OS) were performed to determine the significance of the interaction between treatment and the (1) PAM50 subtypes, (2) PAM50 proliferation score, and (3) clinical and pathological variables. Similar OS analyses were performed in 222 patients treated with weekly paclitaxel versus paclitaxel every 3 weeks in the CALGB/9342 and 9840 metastatic clinical trials. In GEICAM/9906, with a median follow up of 8.7 years, OS of the FEC-P arm was significantly superior compared to the FEC arm (unadjusted HR = 0.693, p = 0.013). A benefit from paclitaxel was only observed in the group of patients with a low PAM50 proliferation score (unadjusted HR = 0.23, p < 0.001; and interaction test, p = 0.006). No significant interactions between treatment and the PAM50 subtypes or the various clinical–pathological variables, including Ki-67 and histologic grade, were identified. Finally, similar OS results were obtained in the CALGB data set, although the interaction test did not reach statistical significance (p = 0.109). The PAM50 proliferation score identifies a subset of patients with a low proliferation status that may derive a larger benefit from weekly paclitaxel.

Evaluation of gene expression by RNA-seq after single dose of trastuzumab (T) reveals predictors of pathologic complete response (pCR) in HER2-positive early breast cancer

10558

Background: The use of a ‘brief exposure’ to single agent T allows the measurement of dynamic changes in the transcriptome that may predict response to T-based combinations. We have shown that most gene expression changes in HER2+ tumors treated with T occur in tumors that ultimately achieve a pCR. Our further analysis suggests several patterns of transcriptional change in pCR tumors suggesting different mechanisms of action of T. RNA-seq analysis provides more in-depth annotation of these mechanisms. Methods: Fresh tumor core biopsies were taken at a 2 week time point after a single dose of T (8mg/m2) from 80 HER2+ early breast cancer patients enrolled on a clinical trial of T>T+C. Nucleic acids were extracted using Qiagen AllPrep and were analyzed with Illumina HT12v3 Beadchip and Illumina 610 QUAD V1 SNP arrays. RNA was also processed for sequencing using the Ovation RNA-Seq System and paired-end sequenced using an Illumina Genome Analyzer IIxRNA-seq data was analyzed with Tophat/Cufflinks. Network analysis was performed with Metacore. Results: Among pCR tumors, distinct patterns of differential expression pre/post T were observed, in both microarray and RNA-seq data. ERBB2 down-regulation was characteristic of pCR in one subgroup by microarray. In this group, differentially expressed genes belonged to interaction networks involved in apoptosis and cell cycle regulation. In contrast, tumors with no change in ErbB2 showed differentially expressed genes that belonged to networks related to chromatin assembly and regulation of immune pathways. NOLC1, RPL41, ZCHHC17, and B2M had altered alternative splicing product distributions in both groups. In the ERBB2 down-regulated group, genes with changed expression were enriched for targets of STAT3 and YY1. Conclusions: RNA-seq and microarray reveal distinct responses in tumors that achieve pCR to T-containing regimens. These methods provide predictive markers for validation in subsequent clinical trials.

Abstract 4719: Dynamic changes in gene expression after one dose of trastuzumab (T) reveal novel biomarkers of response to combination therapy in HER2 positive early stage breast cancer

Introduction: The use of a ‘brief exposure’ to single agent T allows the measurement of dynamic changes in the transcriptome that may predict response to T-based combinations. We have shown that AKT and IGF pathway signatures are significantly perturbed in HER2 positive tumors treated with T, only in tumors that ultimately respond to therapy (ASCO 2010). We now demonstrate that transcriptional changes in responding tumors are unique, and uncover unanticipated mechanisms of action of T in early stage breast cancer.

Patients/Methods: Fresh tumor core biopsies were taken at a 2 week timepoint after a single dose of T (8mg/m2) from 80 HER2-overexpressing, early breast cancer patients enrolled on a clinical trial of T&gt;T+C. Nucleic acids were extracted using Qiagen AllPrep and were analyzed with Illumina HT12v3 Beadchip and Illumina 610QUADv1 SNP arrays. RNA was also processed for sequencing using the Ovation® RNA-Seq System (NuGEN Technologies Inc). The Illumina RNA-Seq Library Preparation and sequencing was performed at the Sequencing Core of the Yale Center for Genome Analysis (West Haven, CT) using the Genome Analyzer IIx sequencer. Gene expression was analyzed in Bioconductor using LIMMA analysis. Gene expression levels from RNA-Seq data was computed using Cufflinks taking mapping results from tophat in SAM format using the Human hg18 RefSeq reference gene set. Genomic biomarkers were compared pathologic complete response (pCR) vs objective response (CR+PR=OR) and non-response (SD+PD=NOR) by RECIST criteria at the surgical timepoint. Pathway analysis was performed using pre-defined gene signatures (IGF-1, AKT) and using DAVID bioinformatics resource.

Results: Gene expression before and after one dose of trastuzumab showed significant changes only in tumors that ultimately achieved pCR. Specifically, AKT and IGF-1 pathways were downregulated in pCR tumors compared with NOR and OR tumors (p=0.04 and p=0.01 respectively). In addition, HER2 and GRB7 were significantly downregulated only in pCR tumors (p=0.003) by both Illumina HT12 arrays and by RNA-sequencing for 17q12 amplicon genes. Analysis of differential expression showed significant changes only in pCR tumors(adjusted pvalue &lt;0.05), compared with NOR and OR tumors. Functional annotation clustering showed enrichment for SH3 domain genes (Enrichment score 2.24) apoptosis pathway genes (ES 1.85;) and differentiation genes (ES 1.3).

Conclusions: Early changes in gene expression can predict pathologic complete response to trastuzumab-containing chemotherapy regimens. Changes in HER2 amplicon genes, AKT, IGF-1 pathways, and genes associated with both apoptosis and differentiation may predict pCR. Brief exposure to trastuzumab may uncover novel mechanisms of action of this life-saving therapy.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4719. doi:10.1158/1538-7445.AM2011-4719

Abstract P6-04-04: Hypoxia Is Associated with Somatic Loss of BRCA1 Protein and Pathway Activity in Triple Negative Breast Cancer

Background: Triple negative breast cancer (TNBC) has loss of BRCA1 activity either through germline mutation, epigenetic modification or negative transcriptional regulators. Yale investigators have demonstrated that tumor hypoxia leads to loss of critical DNA repair activities, including BRCA1, RAD-51 and γ-H2AX (Bindra RS, et al. Cancer Res 2005;65(24): 11597-604). This study was designed to explore the relationship of BRCA1 loss and hypoxia using Carbonic Anhydrase IX (CA-IX), a downstream target of hypoxia-induced factor 1a (HIF-1a), an accepted surrogate biomarker for tumor hypoxia, in TNBC. Methods: Two cohorts of breast tumors from Yale archival materials were studied. The first consisted of 660, unselected breast tumors, the second a cohort of 130 TNBC, both with long term clinical followup. Automated Quantitative Analysis (AQUA) was used to detect the intensity of BRCA1 and CAIX within specific subcellular compartments. CA-IX antibody M75 was provided by the J Zavada lab (1:10000), BRCA1 monoclonal antibody M110 (Ab-1) was purchased from Calbiochem (1:1000). The 130 TNBC cohort was also subjected to whole genome expression analysis. In brief, tissue core biopsies from tumor blocks were subjected to nucleic acid extraction using RecoverAll Total Nucleic Acid Isolation kit (Applied Biosystems) and 600ng total RNA were processed by the Keck Microarray Facility for the Illumina DASL platform. Statistical analysis of gene expression data was carried out in Bioconductor/R software. A set of relevant signatures was selected by enrichment analysis of modules identified by principal component analysis. Signature scores were computed as Pearson correlation between the signature vector of gene contributions and each sample's expression profile for these genes. Results: In the cohort of 660 specimens, 22 were found to have high CA-IX expression score (above cut-off value of 10). Cut-off was based on known positive cell lines and corresponding AQUA scores, as well as visual confirmation of positivity. A negative correlation (Rho=-0.6, p=0.0165) of BRCA1 nuclear protein with CA-IX level was found and this data was reproducible on a duplicate array. Of note, 14 out of 22 hypoxic breast tumors were from triple negative breast cancers (TNBC) (p=0.0034). In the triple negative cohort, CAIX staining was positive in 20 % of cases (n=15/75) and in these samples it was associated with the 2002 van ‘t Veer BRCA1 mutant signature (Rho=0.51). Positive CAIX staining was also associated with worse outcome (p=0.046) as was CAIX mRNA expression (p=0.02).

Conclusions: In human breast tumors BRCA1 nuclear expression is negatively correlated with tumor hypoxia, measured by CAIX. A subset of TNBC has higher levels of hypoxia and BRCA1 signatures. This data suggests that CAIX may be a useful biomarker for BRCA1 loss, and possibly for response to PARP inhibitor therapy. This will be evaluated in our recently completed trial of BSI-201+Irinotecan.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P6-04-04.

Distinct human papillomavirus type 16 methylomes in cervical cells at different stages of premalignancy

Human papillomavirus (HPV) gene expression is dramatically altered during cervical carcinogenesis. Because dysregulated genes frequently show abnormal patterns of DNA methylation, we hypothesized that comprehensive mapping of the HPV methylomes in cervical samples at different stages of progression would reveal patterns of clinical significance. To test this hypothesis, thirteen HPV16-positive samples were obtained from women undergoing routine cervical cancer screening. Complete methylation data were obtained for 98.7% of the HPV16 CpGs in all samples by bisulfite-sequencing. Most HPV16 CpGs were unmethylated or methylated in only one sample. The other CpGs were methylated at levels ranging from 11% to 100% of the HPV16 copies per sample. The results showed three major patterns and two variants of one pattern. The patterns showed minimal or no methylation (A), low level methylation in the E1 and E6 genes (B), and high level methylation at many CpGs in the E5/L2/L1 region (C). Generally, pattern A was associated with negative cytology, pattern B with low-grade lesions, and pattern C with high-grade lesions. The severity of the cervical lesions was then ranked by the HPV16 DNA methylation patterns and, independently, by the pathologic diagnoses. Statistical analysis of the two rating methods showed highly significant agreement. In conclusion, analysis of the HPV16 DNA methylomes in clinical samples of cervical cells led to the identification of distinct methylation patterns which, after validation in larger studies, could have potential utility as biomarkers of neoplastic cervical progression.

Activity and mechanism of action of histone deacetylase inhibitors in trastuzumab resistant breast cancer

Abstract #3060

Introduction: The Her2 antibody trastuzumab (T) is the first biological therapy that has proven the value of Her2 blockade in breast cancer. However, not all patients respond to trastuzumab and relapses occur after this treatment. Several studies suggest that histone deacetylase inhibitors (HDACis) are active in Her2-amplified breast cancer but none of them have addressed the role of HDACis in trastuzumab-resistant breast cancer. HDACis act by inhibiting the deacetylation of histones and other non-histone proteins such as Hsp90. Our goal is to study the activity, optimal sequencing and mechanism of action of HDACis in Her2-amplified cell lines, which demonstrate either de novo or acquired resistance to trastuzumab.&#x2028; Methods: T-sensitive and Her2-overexpressing BT474 & SkBr3, T-resistant BT474 and de novo T-resistant cell lines MDA-MB361, MDA-MB453, UACC812 and UACC893 were used for all experiments. The HDACi TSA (0-1uM) and trastuzumab at 100ug/mL were used for proliferation assays applying WST1 reagent (Roche). Western-immunoblotting was performed using cell lysates from untreated and treated cells. Proteolytic cleavage of caspase-3 and PARP was analyzed as indicator for apoptosis, Akt expression and phosphorylation for survival, Her2 and Erk1/2 expression and phosphorylation for proliferation and Hsp90 and Hsp70 were analyzed as key enzymes for chaperone activity.&#x2028; Results: TSA inhibited proliferation in all cell lines [IC50 100-300 nM]. The addition of trastuzumab was antagonistic in de novo T-resistant cell lines MDA MB 453, UACC 812 and 893 but did not change the effect induced by TSA in acquired T-resistant cells (BT474_H100). Her2 expression decreased in TSA and TSA+T treated cells but not in cells treated with trastuzumab alone. Erk1/2 phosphorylation decreased in TSA and TSA+T treated cells but was not affected by trastuzumab alone. Similar phosphorylation patterns were observed regarding Akt. Of note, Akt, Erk1/2, Hsp90 and Hsp70 levels did not change. Finally, TSA induced caspase-3 and PARP cleavage when given as single agent or in combination with trastuzumab.&#x2028; Discussion: This study suggests that HDACis are active in both trastuzumab-sensitive and de novo or acquired trastuzumab-resistant cell lines. However, use of trastuzumab together with TSA may be antagonistic in tumors that are resistant against trastuzumab. Further studies are required in these cell lines in order to elucidate the mechanism of resistance against trastuzumab and the mechanism of the here observed antagonistic effect between trastuzumab and TSA.

Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 3060.

MicroRNA profiling of triple negative breast cancer: predicting outcome and targets

Abstract #2040

Background: Breast tumors have been categorized on their molecular basis into ER- and/or PR positive, HER2 gene amplified and triple-negative tumors. Among these, triple-negative tumors represent the most aggressive subclass of breast tumors that are most prevalent in African American women. Currently, no effective targeted therapies or potential biomarkers have been identified to predict outcome within this group.&#x2028; MicroRNAs(miRNAs) are global regulatory RNAs misregulated in almost every cancer and may function as tumor suppressor genes or proto-oncogenes. Specific miRNAs are aberrantly expressed in breast tumor versus normal tissue. Although miRNA expression profiles for the ER positive and HER2 positive subtype have been reported, the triple negative subtype remains to be studied.&#x2028; Hypothesis: We hypothesize that identifying the miRNA signature of the triple negative tumors as a subclass as well as differences in this signature between triple negative patients with known outcome will lead to the identification of miRNAs predictive of outcome. For these studies we will use a large, well-annotated database of triple negative breast cancer patients with long-term follow-up. We will then test the identified miRNAs in a well-established cell-culture system to determine which miRNAs can act as potential therapeutic agents in triple negative breast tumors.&#x2028; Study Design: We have performed miRNA microarray analysis to identify miRNA expression patterns in over 40 triple negative breast cancers using both fresh frozen tumors and formalin-fixed paraffin embedded (FFPE) samples. We have used the Ambion miRNA Taqman Low Density Array for analysis of these samples. The potential of specific miRNAs to predict outcome is currently being determined using statistical analysis.&#x2028; Results: Our initial studies indicated that the triple negative subgroup has a significantly different miRNA signature than the HER2 positive subtype. We also additionally found unique miRNA expression patterns that may define triple negative cancers as a class. The triple negative tumors are known to have low levels of p53 and our microarray data showed that miR-34a which has been shown to be co-regulated by p53 and has similar expression levels as p53 in tissues, was significantly low in the triple negative compared to HER 2 positive breast cancer. We are currently evaluating the impact of altering mir-34 expression in these cell lines in vitro.

Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 2040.

Quantitative assessment of tissue biomarkers and construction of a model to predict outcome in breast cancer using multiple imputation

Missing data pose one of the greatest challenges in the rigorous evaluation of biomarkers. The limited availability of specimens with complete clinical annotation and quality biomaterial often leads to underpowered studies. Tissue microarray studies, for example, may be further handicapped by the loss of data points because of unevaluable staining, core loss, or the lack of tumor in the histospot. This paper presents a novel approach to these common problems in the context of a tissue protein biomarker analysis in a cohort of patients with breast cancer. Our analysis develops techniques based on multiple imputation to address the missing value problem. We first select markers using a training cohort, identifying a small subset of protein expression levels that are most useful in predicting patient survival. The best model is obtained by including both protein markers (including COX6C, GATA3, NAT1, and ESR1) and lymph node status. The use of either lymph node status or the four protein expression levels provides similar improvements in goodness-of-fit, with both significantly better than a baseline clinical model. Using the same multiple imputation strategy, we then validate the results out-of-sample on a larger independent cohort. Our approach of integrating multiple imputation with each stage of the analysis serves as an example that may be replicated or adapted in future studies with missing values.

Long-term outcomes and clinicopathologic differences of African-American versus white patients treated with breast conservation therapy for early-stage breast cancer

BACKGROUND

African-American (AA) and white patients with early-stage disease who were treated with breast conservation therapy (BCT) were examined to detect differences in clinicopathologic features and outcomes as a function of race.

METHODS

Clinical data from the charts of 2164 white and 207 AA patients treated with BCT, and p53 expression status on 444 patients (from an existing tissue database), were analyzed to detect differences between the 2 cohorts.

RESULTS

The median follow-up was 7 years. There were no differences in the method of tumor detection, lymph nodes excised, surgical margin status, or chemotherapy/radiotherapy delivered, reflecting similar screening and treatment policies for AA women in the study community. Despite this, AA patient presented at a younger age, with higher T and N classifications, and more estrogen and progesterone negative and "triple negative" tumors (all P values < .016). Tumors in AA patients were p53 positive more often than tumors in white patients (P= .0003). At 10 years, AA patients had a higher rate of distant metastasis (20% vs 17%; P= .042), lymph node recurrence (6% vs 2%; P= .004), and breast recurrence (17% vs 13%; P= .036). There was no difference in overall survival between the 2 groups. On multivariate analysis, only lymph node recurrence (risk ratio of 3.140; 95% confidence interval, 1.396-7.063 [P= .0057]) remained significantly higher among AA women.

CONCLUSIONS

In this cohort of uniformly treated patients, the authors found the expected clinicopathologic differences, but race was not found to be an independent predictor of local recurrence for AA patients when other confounding variables were taken into account in the multivariate model. These findings suggest that BCT is a reasonable option for appropriately selected AA patients. To the authors' knowledge, this is the largest study addressing outcomes after BCT for AA women published to date.

MSVM-RFE: extensions of SVM-RFE for multiclass gene selection on DNA microarray data

MOTIVATION

Given the thousands of genes and the small number of samples, gene selection has emerged as an important research problem in microarray data analysis. Support Vector Machine-Recursive Feature Elimination (SVM-RFE) is one of a group of recently described algorithms which represent the stat-of-the-art for gene selection. Just like SVM itself, SVM-RFE was originally designed to solve binary gene selection problems. Several groups have extended SVM-RFE to solve multiclass problems using one-versus-all techniques. However, the genes selected from one binary gene selection problem may reduce the classification performance in other binary problems.

RESULTS

In the present study, we propose a family of four extensions to SVM-RFE (called MSVM-RFE) to solve the multiclass gene selection problem, based on different frameworks of multiclass SVMs. By simultaneously considering all classes during the gene selection stages, our proposed extensions identify genes leading to more accurate classification.

Phase I and Pharmacokinetic Study of AI-850, A Novel Microparticle Hydrophobic Drug Delivery System for Paclitaxel

PURPOSE

AI-850, paclitaxel in a novel polyoxyethylated castor oil-free hydrophobic microparticle delivery system, is being developed based on its favorable preclinical safety and antitumor activity profiles. The objectives of the study were to assess the feasibility and safety of administering AI-850 as a <30-min i.v. infusion without premedication every 3 weeks, determine the maximum tolerated dose and the phase II recommended dose of AI-850, study the pharmacokinetics of paclitaxel in this new formulation, and seek evidence of anticancer activity.

EXPERIMENTAL DESIGN

This was an open-label phase I dose escalation study of AI-850 in patients with advanced solid malignancies. AI-850 doses were escalated according to a modified Fibonacci scheme. Clinical and laboratory toxicity was monitored, and paclitaxel plasma concentrations were measured by liquid chromatography-tandem mass spectrometry.

RESULTS

Twenty-two patients received 56 courses of AI-850 at five dose cohorts ranging from 36 to 250 mg/m(2). Grade 4 neutropenia, either exceeding 5 days or complicated by fever, was dose limiting in two of six patients at 250 mg/m(2) AI-850. Three patients experienced grade 2 to 4 infusion-related adverse reactions. Toxicities, including fatigue, alopecia, nausea and vomiting, neuropathy, anorexia, and myalgia, were mild to moderate, reversible, and not dose related. Pharmacokinetics of free and total paclitaxel showed biexponential plasma decay and dose proportionality for maximum plasma paclitaxel concentration and area under the concentration versus time curve. Antitumor activity was documented in two patients with endometrial and tongue carcinomas.

CONCLUSIONS

The administration of AI-850 as a brief infusion once every 3 weeks was feasible at doses up to 205 mg/m(2). The potential of AI-850 as an alternative to other approved paclitaxel formulations requires further clinical evaluation.

Differentiated cells are more efficient than adult stem cells for cloning by somatic cell nuclear transfer

Since the creation of Dolly via somatic cell nuclear transfer (SCNT), more than a dozen species of mammals have been cloned using this technology. One hypothesis for the limited success of cloning via SCNT (1%-5%) is that the clones are likely to be derived from adult stem cells. Support for this hypothesis comes from the findings that the reproductive cloning efficiency for embryonic stem cells is five to ten times higher than that for somatic cells as donors and that cloned pups cannot be produced directly from cloned embryos derived from differentiated B and T cells or neuronal cells. The question remains as to whether SCNT-derived animal clones can be derived from truly differentiated somatic cells. We tested this hypothesis with mouse hematopoietic cells at different differentiation stages: hematopoietic stem cells, progenitor cells and granulocytes. We found that cloning efficiency increases over the differentiation hierarchy, and terminally differentiated postmitotic granulocytes yield cloned pups with the greatest cloning efficiency.

Cruella: developing a scalable tissue microarray data management system

CONTEXT

Compared with DNA microarray technology, relatively little information is available concerning the special requirements, design influences, and implementation strategies of data systems for tissue microarray technology. These issues include the requirement to accommodate new and different data elements for each new project as well as the need to interact with pre-existing models for clinical, biological, and specimen-related data.

OBJECTIVE

To design and implement a flexible, scalable tissue microarray data storage and management system that could accommodate information regarding different disease types and different clinical investigators, and different clinical investigation questions, all of which could potentially contribute unforeseen data types that require dynamic integration with existing data.

DESIGN

The unpredictability of the data elements combined with the novelty of automated analysis algorithms and controlled vocabulary standards in this area require flexible designs and practical decisions. Our design includes a custom Java-based persistence layer to mediate and facilitate interaction with an object-relational database model and a novel database schema. User interaction is provided through a Java Servlet-based Web interface.

RESULTS

Cruella has become an indispensable resource and is used by dozens of researchers every day. The system stores millions of experimental values covering more than 300 biological markers and more than 30 disease types. The experimental data are merged with clinical data that has been aggregated from multiple sources and is available to the researchers for management, analysis, and export.

CONCLUSION

Cruella addresses many of the special considerations for managing tissue microarray experimental data and the associated clinical information. A metadata-driven approach provides a practical solution to many of the unique issues inherent in tissue microarray research, and allows relatively straightforward interoperability with and accommodation of new data models.

Characterizing disease states from topological properties of transcriptional regulatory networks

Background

High throughput gene expression experiments yield large amounts of data that can augment our understanding of disease processes, in addition to classifying samples. Here we present new paradigms of data Separation based on construction of transcriptional regulatory networks for normal and abnormal cells using sequence predictions, literature based data and gene expression studies. We analyzed expression datasets from a number of diseased and normal cells, including different types of acute leukemia, and breast cancer with variable clinical outcome.

Results

We constructed sample-specific regulatory networks to identify links between transcription factors (TFs) and regulated genes that differentiate between healthy and diseased states. This approach carries the advantage of identifying key transcription factor-gene pairs with differential activity between healthy and diseased states rather than merely using gene expression profiles, thus alluding to processes that may be involved in gene deregulation. We then generalized this approach by studying simultaneous changes in functionality of multiple regulatory links pointing to a regulated gene or emanating from one TF (or changes in gene centrality defined by its in-degree or out-degree measures, respectively). We found that samples can often be separated based on these measures of gene centrality more robustly than using individual links.

We examined distributions of distances (the number of links needed to traverse the path between each pair of genes) in the transcriptional networks for gene subsets whose collective expression profiles could best separate each dataset into predefined groups. We found that genes that optimally classify samples are concentrated in neighborhoods in the gene regulatory networks. This suggests that genes that are deregulated in diseased states exhibit a remarkable degree of connectivity.

Conclusion

Transcription factor-regulated gene links and centrality of genes on transcriptional networks can be used to differentiate between cell types. Transcriptional network blueprints can be used as a basis for further research into gene deregulation in diseased states.

Lineage specificity of gene expression patterns

The hematopoietic system offers many advantages as a model for understanding general aspects of lineage choice and specification. Using oligonucleotide microarrays, we compared gene expression patterns of multiple purified hematopoietic cell populations, including neutrophils, monocytes, macrophages, resting, centrocytic, and centroblastic B lymphocytes, dendritic cells, and hematopoietic stem cells. Some of these cells were studied under both resting and stimulated conditions. We studied the collective behavior of subsets of genes derived from the Biocarta database of functional pathways, hand-tuned groupings of genes into broad functional categories based on the Gene Ontology database, and the metabolic pathways in the Kyoto Encyclopedia of Genes and Genomes database. Principal component analysis revealed strikingly pervasive differences in relative levels of gene expression among cell lineages that involve most of the subsets examined. These results indicate that many processes in these cells behave differently in different lineages. Much of the variation among lineages was captured by the first few principal components. Principal components biplots were found to provide a convenient visual display of the contributions of the various genes within the subsets in lineage discrimination. Moreover, by applying tree-constructing methodologies borrowed from phylogenetics to the expression data from differentiated cells and stem cells, we reconstructed a tree of relationships that resembled the established hematopoietic program of lineage development. Thus, the mRNA expression data implicitly contained information about developmental relationships among cell types.

Induction of Erythropoiesis by MHC-Mediated Cognate Interactions between B- and T-Cells

We have previously shown that the expression of membrane burst-promoting activity (mBPA), an erythropoietic cytokine, by B-lymphocytes is augmented by the addition of allogeneic effector cells to the B-cells. Here, we have examined immune mechanisms involved in the induction/promotion of erythropoiesis as assessed by the capacity of autologous and allogeneic peripheral blood lymphocytes to augment burst-forming unit-erythroid (BFU-E) in normal human bone marrow cells in vitro. Preincubation of mBPA-expressing human B-cells with monoclonal antibodies to major histocompatibility complex (MHC) antigens, abrogated erythropoietic activity of both autologous and allogeneic lymphocytes, suggesting that MHC antigens play a role in regulating the expression of the erythroid growth factor. Inhibition of BFU-E proliferation was also evident when antibodies to MHC class-I or class-II antigens were added directly to marrow culture. Furthermore, addition of anti-CD4 antibody to the cultures of PBL and autologous target BM cells markedly reduced erythroid proliferation induced by PBL. By contrast, anti-CD8 and control (UPC-10) monoclonal antibodies had no effect. These results provide evidence that MHC-mediated cognate interactions between T- and B-lymphocytes may participate in the control of erythropoiesis, either directly or by modulating mBPA function.

Human macrophage colony-stimulating factor is expressed at and shed from the cell surface

Surface membrane-associated growth factors are being recognized as important for developmental processes, including cell assembly, differentiation, and growth. To investigate the role of membrane-bound macrophage colony-stimulating factor (M-CSF) in myelopoiesis, and whether this factor is released from the cell surface in association with shed membrane-derived vesicles, COS-1 cells were transfected with cDNAs for M-CSF-tau (containing the transmembrane domain) or a soluble mutant form of the molecule lacking the transmembrane domain ([s]M-CSF-alpha). COS-1 cells transfected with either cDNA released activity into the spent culture medium. Conditioned medium was separated by centrifugation into supernatants and pellets were found to contain plasma membrane-derived vesicles by transmission electron microscopy. When medium fractions were assayed in marrow cultures, activity was localized to shed plasma membrane-derived vesicles in medium conditioned by cells transfected with cDNA for M-CSF-tau and in the vesicle-free supernatants of medium conditioned by cells transfected with cDNA for [s]M-CSF-alpha. In addition, nuclear, mitochondrial, and plasma membrane subfractions of stably transfected cells were prepared and assayed for activity. Concentration-dependent stimulation of macrophage colony formation was observed with purified plasma membranes (but not nuclear or cytosolic proteins) from cells transfected with cDNA for M-CSF-tau. By contrast, membranes from untransfected cells and cells transfected with cDNA for [s]M-CSF-alpha or control DNA expressed no activity. Together, the data indicate that human M-CSF is expressed at the cell surface and exfoliated in association with surface membrane-derived vesicles.

Modulation of hematopoiesis by lymphocyte membrane-derived components

Membrane bound erythroid burst-promoting activity (mBPA) is an integral membrane protein that is present on normal B-cells and some activated T-cells, that induces burst-forming units-erythroid (BFU-E) when cultured with human erythropoietin (rHuEpo). Plasma membranes and vesicles shed from the leukemic A-1 cell line express mBPA. This activity derived from both A-1 cells and normal B-cells can be immunoadsorbed by the D3A4 antibody raised against mBPA. In this study, we demonstrate that interferon-gamma (IFN-gamma) suppresses BFU-E proliferation when added directly to culture of normal human bone marrow cells and in the absence and presence of A-1 cells. However, FACS analysis reveals that IFN-gamma enhances the surface expression of mBPA on A-1 cells. The role of IFN-gamma in modulating erythropoiesis in vitro is discussed with respect to the role of shedding membrane-derived vesicles from the B-cell surface.