Search

A duty of loyalty prohibits fiduciaries from appropriating business opportunities from their companies. Starting in 2000, Delaware, followed by several other states, allowed boards to waive their duty. We show that public firms covered by waiver laws invest less in R&D, produce fewer and less valuable patents, and exhibit abnormally high inventor departures. Remaining innovation activities contribute less to firm value, a fact confirmed by the market reaction when firms reveal their curtailed internal growth opportunities by announcing acquisitions. Consistent with the laws’ intent to provide contracting flexibility to emerging firms, we find evidence of positive impacts for small firms. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.

PURPOSE To inform precision oncology, methods are needed to use electronic health records (EHRs) to identify patients with cancer who are experiencing clinical inflection points, consistent with worsening prognosis or a high propensity to change treatment, at specific time points. Such patients might benefit from real-time screening for clinical trials. METHODS Using serial unstructured imaging reports for patients with solid tumors or lymphoma participating in a single-institution precision medicine study, we trained a deep neural network natural language processing (NLP) model to dynamically predict patients' prognoses and propensity to start new palliative-intent systemic therapy within 30 days. Model performance was evaluated using Harrell's c-index (for prognosis) and the area under the receiver operating characteristic curve (AUC; for new treatment and new clinical trial enrollment). Associations between model outputs and manual annotations of cancer progression were also evaluated using the AUC. RESULTS A deep NLP model was trained and evaluated using 302,688 imaging reports for 16,780 patients. In a held-out test set of 34,770 reports for 1,952 additional patients, the model predicted survival with a c-index of 0.76 and initiation of new treatment with an AUC of 0.77. Model-generated prognostic scores were associated with annotation of cancer progression on the basis of manual EHR review (n = 1,488 reports for 110 patients with lung or colorectal cancer) with an AUC of 0.78, and predictions of new treatment were associated with annotation of cancer progression on the basis of manual EHR review with an AUC of 0.84. CONCLUSION Training a deep NLP model to identify clinical inflection points among patients with cancer is feasible. This approach could identify patients who may benefit from real-time targeted clinical trial screening interventions at health system scale.

Tumor molecular profiling of single gene-variant (‘first-order’) genomic alterations informs potential therapeutic approaches. Interactions between such first-order events and global molecular features (for example, mutational signatures) are increasingly associated with clinical outcomes, but these ‘second-order’ alterations are not yet accounted for in clinical interpretation algorithms and knowledge bases. We introduce the Molecular Oncology Almanac (MOAlmanac), a paired clinical interpretation algorithm and knowledge base to enable integrative interpretation of multimodal genomic data for point-of-care decision making and translational-hypothesis generation. We benchmarked MOAlmanac to a first-order interpretation method across multiple retrospective cohorts and observed an increased number of clinical hypotheses from evaluation of molecular features and profile-to-cell line matchmaking. When applied to a prospective precision oncology trial cohort, MOAlmanac nominated a median of two therapies per patient and identified therapeutic strategies administered in 47% of patients. Overall, we present an open-source computational method for integrative clinical interpretation of individualized molecular profiles. Van Allen and colleagues develop a data-integration framework with an underlying knowledge base supporting clinical decision making and also serving as a hypothesis-generating platform, which the authors benchmark and validate across several retrospective cohorts and a prospective precision oncology trial.

The determination of molecular features that mediate clinically aggressive phenotypes in prostate cancer remains a major biological and clinical challenge1,2. Recent advances in interpretability of machine learning models as applied to biomedical problems may enable discovery and prediction in clinical cancer genomics3–5. Here we developed P-NET—a biologically informed deep learning model—to stratify patients with prostate cancer by treatment-resistance state and evaluate molecular drivers of treatment resistance for therapeutic targeting through complete model interpretability. We demonstrate that P-NET can predict cancer state using molecular data with a performance that is superior to other modelling approaches. Moreover, the biological interpretability within P-NET revealed established and novel molecularly altered candidates, such as MDM4 and FGFR1, which were implicated in predicting advanced disease and validated in vitro. Broadly, biologically informed fully interpretable neural networks enable preclinical discovery and clinical prediction in prostate cancer and may have general applicability across cancer types. A biologically informed, interpretable deep learning model has been developed to evaluate molecular drivers of resistance to cancer treatment, predict clinical outcomes and guide hypotheses on disease progression.

Immune checkpoint inhibitors (ICIs) have minimal therapeutic effect in hormone receptor-positive (HR+ ) breast cancer. We present final overall survival (OS) results (n = 88) from a randomized phase 2 trial of eribulin ± pembrolizumab for patients with metastatic HR+ breast cancer, computationally dissect genomic and/or transcriptomic data from pre-treatment tumors (n = 52) for molecular associations with efficacy, and identify cytokine changes differentiating response and ICI-related toxicity (n = 58). Despite no improvement in OS with combination therapy (hazard ratio 0.95, 95% CI 0.59–1.55, p = 0.84), immune infiltration and antigen presentation distinguished responding tumors, while tumor heterogeneity and estrogen signaling independently associated with resistance. Moreover, patients with ICI-related toxicity had lower levels of immunoregulatory cytokines. Broadly, we establish a framework for ICI response in HR+ breast cancer that warrants diagnostic and therapeutic validation. ClinicalTrials.gov Registration: NCT03051659., A randomized phase 2 clinical trial has recently shown no benefit of the combination eribulin and pembrolizumab over pembrolizumab alone in HR + metastatic breast cancer patients (NCT03051659). Here, the authors are reporting the final OS data and biomarker analyses on a subset of samples to analyze molecular correlates

Several risk factors have been established for colorectal cancer, yet their direct mutagenic effects in patients' tumors remain to be elucidated. Here, we leveraged whole-exome sequencing data from 900 colorectal cancer cases that had occurred in three U.S.-wide prospective studies with extensive dietary and lifestyle information. We found an alkylating signature that was previously undescribed in colorectal cancer and then showed the existence of a similar mutational process in normal colonic crypts. This alkylating signature is associated with high intakes of processed and unprocessed red meat prior to diagnosis. In addition, this signature was more abundant in the distal colorectum, predicted to target cancer driver mutations KRAS p.G12D, KRAS p.G13D, and PIK3CA p.E545K, and associated with poor survival. Together, these results link for the first time a colorectal mutational signature to a component of diet and further implicate the role of red meat in colorectal cancer initiation and progression. Significance: Colorectal cancer has several lifestyle risk factors, but the underlying mutations for most have not been observed directly in tumors. Analysis of 900 colorectal cancers with whole-exome sequencing and epidemiologic annotations revealed an alkylating mutational signature that was associated with red meat consumption and distal tumor location, as well as predicted to target KRAS p.G12D/p.G13D. This article is highlighted in the In This Issue feature, p. 2355

Purpose: Enzalutamide is a second-generation androgen receptor (AR) inhibitor that has improved overall survival (OS) in metastatic castration-resistant prostate cancer (CRPC). However, nearly all patients develop resistance. We designed a phase II multicenter study of enzalutamide in metastatic CRPC incorporating tissue and blood biomarkers to dissect mechanisms driving resistance. Patients and Methods: Eligible patients with metastatic CRPC underwent a baseline metastasis biopsy and then initiated enzalutamide 160 mg daily. A repeat metastasis biopsy was obtained at radiographic progression from the same site when possible. Blood for circulating tumor cell (CTC) analysis was collected at baseline and progression. The primary objective was to analyze mechanisms of resistance in serial biopsies. Whole-exome sequencing was performed on tissue biopsies. CTC samples underwent RNA sequencing. Results: A total of 65 patients initiated treatment, of whom 22 (33.8%) had received prior abiraterone. Baseline biopsies were enriched for alterations in AR (mutations, amplifications) and tumor suppression genes (PTEN, RB1, and TP53), which were observed in 73.1% and 92.3% of baseline biopsies, respectively. Progression biopsies revealed increased AR amplifications (64.7% at progression vs. 53.9% at baseline) and BRCA2 alterations (64.7% at progression vs. 38.5% at baseline). Genomic analysis of baseline and progression CTC samples demonstrated increased AR splice variants, AR-regulated genes, and neuroendocrine markers at progression. Conclusions: Our results demonstrate that a large proportion of enzalutamide-treated patients have baseline and progression alterations in the AR pathway and tumor suppressor genes. We demonstrate an increased number of BRCA2 alterations post-enzalutamide, highlighting the importance of serial tumor sampling in CRPC.

Biases in data used to train machine learning (ML) models can inflate their prediction performance and confound our understanding of how and what they learn. Although biases are common in biological data, systematic auditing of ML models to identify and eliminate these biases is not a common practice when applying ML in the life sciences. Here we devise a systematic, principled, and general approach to audit ML models in the life sciences. We use this auditing framework to examine biases in three ML applications of therapeutic interest and identify unrecognized biases that hinder the ML process and result in substantially reduced model performance on new datasets. Ultimately, we show that ML models tend to learn primarily from data biases when there is insufficient signal in the data to learn from. We provide detailed protocols, guidelines, and examples of code to enable tailoring of the auditing framework to other biomedical applications., Fatma-Elzahraa Eid et al. illustrate a principled approach for identifying biases that can inflate the performance of biological machine learning models. When applied to three biomedical prediction problems, they identify previously unrecognized biases and ultimately show that models are likely to learn primarily from data biases when there is insufficient learnable signal in the data.

The management of advanced-stage renal cell carcinoma (RCC) has been transformed by the development of immune-checkpoint inhibitors (ICIs). Nonetheless, most patients do not derive durable clinical benefit from these agents. Importantly, unlike other immunotherapy-responsive solid tumours, most RCCs have only a moderate mutational burden, and paradoxically, high levels of tumour CD8+ T cell infiltration are associated with a worse prognosis in patients with this disease. Building on the successes of antibodies targeting the PD-1 and CTLA4 immune checkpoints, multiple innovative immunotherapies are now in clinical development for the treatment of patients with RCC, including ICIs with novel targets, co-stimulatory pathway agonists, modified cytokines, metabolic pathway modulators, cell therapies and therapeutic vaccines. However, the successful development of such novel immune-based treatments and of immunotherapy-based combinations will require a disease-specific framework that incorporates a deep understanding of RCC immunobiology. In this Review, using the structure provided by the well-described cancer–immunity cycle, we outline the key steps required for a successful antitumour immune response in the context of RCC, and describe the development of promising new immunotherapies within the context of this framework. With this approach, we summarize and analyse the most encouraging targets of novel immune-based therapies within the RCC microenvironment, and review the landscape of emerging antigen-directed therapies for this disease. Renal cell carcinomas (RCCs) are generally immunogenic, but only a subset of patients receiving currently approved immune-checkpoint inhibitors have long-term disease remission. In this Review, the authors provide a conceptual framework for developing novel immunotherapy approaches, including an overview of the most promising novel immune checkpoints and antigen-directed therapies, and highlighting the potential of these agents to further improve the outcomes in patients with RCC.

Despite advances in the screening and treatment of prostate cancer, the therapy options available, particularly for later stages of the disease, remain limited, and the treatment-resistant setting represents a serious unmet medical need. Moreover, disease heterogeneity and disparities in patient access to medical advances result in considerable variability in outcomes across patients. Disease classification based on genomic sequencing is a promising approach for identifying patients whose tumors exhibit actionable targets and for making more informed treatment decisions. Here we discuss how precision oncology can be accelerated to inform broader genomically driven clinical decisions for men with advanced prostate cancer, to inform drug development and, ultimately, to contribute to new treatment paradigms. Beltran and colleagues discuss the challenges in treating metastatic prostate cancer and strategies to accelerate precision oncology and improve therapy and clinical decisions in this setting.

PURPOSE Evidence-based somatic and germline sequencing has transformed cancer care and improves patient outcomes. However, patients’ low genetic literacy and misunderstanding of their own genomic results poses a threat to the realization of precision oncology. To optimize patient genomic comprehension, we developed a Web-based, patient-directed, genomic sequencing education and return-of-results tool, HOPE-Genomics. METHODS The HOPE-Genomics prototype included somatic and germline sequencing results, embedded multimedia genomic education, and interactive features (eg, request for genetic counseling). Between January and April 2018, we elicited feedback on tool usability and comprehensiveness through participant surveys, 4 focus groups of patients with cancer and their family members, and 3 provider focus groups (comprising 8 patients, 5 family members, and 19 providers). RESULTS We identified themes in patient/family tool-related responses, including the desire to view a patient-friendly report, a desire to receive multiple types of genomic information (eg, prognostic and uncertain), high acceptability of report content, and interest in tool-enabled access to genetic counseling. Major themes from the clinician focus groups included believing the tool could help patients formulate questions and facilitate patients’ communication of results to family members. However, there were diverse responses from all participants in terms of tool implementation (ie, timing and nature of report release). Some participants preferred report release before meeting with the provider, and others preferred it during the appointment. Additionally, some clinicians were concerned about providing prognostic and treatment information through the tool. CONCLUSION There was high acceptability and interest from patients, family members, and providers in a patient-directed genomics report. Future work will determine whether direct-to-patient reporting of genomic results improves patient knowledge, care engagement, and compliance with genomically guided interventions.

Background The phosphatidyl 3-inositol kinase (PI3K)/Akt/mechanistic target of rapamycin (mTOR) pathway frequently is activated in patients with urothelial carcinoma (UC). In the current study, the authors performed a phase 2 study evaluating the efficacy of the pan-isoform class I PI3K inhibitor buparlisib in patients with platinum-refractory metastatic UC. Methods Two cohorts were recruited: an initial genetically unselected cohort and a subsequent expansion cohort of patients with PI3K/Akt/mTOR pathway-altered tumors. The primary endpoint was the 2-month progression-free survival rate. A rate of ≥80% was considered promising using a Simon 2-stage minimax design. Secondary endpoints included safety and correlation of markers of PI3K pathway activation with outcome. Results Six of 13 evaluable patients within the initial cohort demonstrated stable disease and 1 demonstrated a partial response, which was below the cutoff of 9 patients required to proceed to stage 2. Three of the patients with stable disease and the patient with a partial response harbored somatic TSC1 alterations. Four patients subsequently were recruited onto an expansion cohort: 3 patients with TSC1 alterations and 1 patient with a PIK3CA-activating mutation. No patient achieved disease control at 8 weeks and accrual was halted. Of the 19 patients evaluable for toxicity, 17 demonstrated treatment-related toxicities, 2 of whom had to discontinue therapy. Conclusions Buparlisib was found to demonstrate modest activity in patients with metastatic UC whose tumors harbored TSC1 loss of function alterations; however, this was not a robust predictor of response to buparlisib. The pattern of genetic coalterations likely influences drug sensitivity. Given the modest clinical activity and substantial toxicity of buparlisib, future trials of PI3K inhibitors in patients with UC should focus on isoform-selective PI3K inhibitors in genomically selected patients. Lay summary The phosphatidyl 3-inositol kinase (PI3K)/Akt/mechanistic target of rapamycin (mTOR) signaling pathway frequently is upregulated in patients with metastatic urothelial carcinoma (UC). This trial explored buparlisib, an inhibitor of the pathway, in patients with heavily pretreated metastatic UC. Although the drug was found to have modest efficacy, with 6 patients experiencing stable disease and 1 patient achieving a partial response at 8 weeks on therapy, significant side effects also were observed. Patients with specific genetic alterations responded to treatment. Further studies of PI3K pathway inhibition are warranted using newer agents that have superior toxicity profiles and are more selective inhibitors of the pathway.

Prior data have variably implicated the inactivation of the mammalian SWItch/Sucrose Non-Fermentable (mSWI/SNF) complex with increased tumor sensitivity to immune checkpoint inhibitors (ICI). Herein, we examined the association between mSWI/SNF variants and clinical outcomes to ICIs. We correlated somatic loss-of-function (LOF) variants in a predefined set of mSWI/SNF genes (ARID1A, ARID1B, SMARCA4, SMARCB1, PBRM1, and ARID2) with clinical outcomes in patients with cancer treated with systemic ICIs. We identified 676 patients from Dana-Farber Cancer Institute (DFCI, Boston, MA) and 848 patients from a publicly available database from Memorial Sloan Kettering Cancer Center (MSKCC, New York, NY) who met the inclusion criteria. Multivariable analyses were conducted and adjusted for available baseline factors and tumor mutational burden. Median follow-up was 19.6 (17.6–22.0) months and 28.0 (25.0–29.0) months for the DFCI and MSKCC cohorts, respectively. Seven solid tumor subtypes were examined. In the DFCI cohort, LOF variants of mSWI/SNF did not predict improved overall survival (OS), time-to-treatment failure (TTF), or disease control rate. Only patients with renal cell carcinoma with mSWI/SNF LOF showed significantly improved OS and TTF with adjusted HRs (95% confidence interval) of 0.33 (0.16–0.7) and 0.49 (0.27–0.88), respectively, and this was mostly driven by PRBM1. In the MSKCC cohort, where only OS was captured, LOF mSWI/SNF did not correlate with improved outcomes across any tumor subtype. We did not find a consistent association between mSWI/SNF LOF variants and improved clinical outcomes to ICIs, suggesting that mSWI/SNF variants should not be considered as biomarkers of response to ICIs.

PURPOSE Smoldering multiple myeloma (SMM) is a precursor condition of multiple myeloma (MM) with a 10% annual risk of progression. Various prognostic models exist for risk stratification; however, those are based on solely clinical metrics. The discovery of genomic alterations that underlie disease progression to MM could improve current risk models. METHODS We used next-generation sequencing to study 214 patients with SMM. We performed whole-exome sequencing on 166 tumors, including 5 with serial samples, and deep targeted sequencing on 48 tumors. RESULTS We observed that most of the genetic alterations necessary for progression have already been acquired by the diagnosis of SMM. Particularly, we found that alterations of the mitogen-activated protein kinase pathway ( KRAS and NRAS single nucleotide variants [SNVs]), the DNA repair pathway (deletion 17p, TP53, and ATM SNVs), and MYC (translocations or copy number variations) were all independent risk factors of progression after accounting for clinical risk staging. We validated these findings in an external SMM cohort by showing that patients who have any of these three features have a higher risk of progressing to MM. Moreover, APOBEC associated mutations were enriched in patients who progressed and were associated with a shorter time to progression in our cohort. CONCLUSION SMM is a genetically mature entity whereby most driver genetic alterations have already occurred, which suggests the existence of a right-skewed model of genetic evolution from monoclonal gammopathy of undetermined significance to MM. We identified and externally validated genomic predictors of progression that could distinguish patients at high risk of progression to MM and, thus, improve on the precision of current clinical models.

Multiple myeloma (MM) is a plasma-cell neoplasm that is treated with high-dose chemotherapy, autologous stem cell transplant (ASCT) and long-term immunomodulatory drug (IMiD) maintenance. The presence of somatic mutations in the peripheral blood is termed clonal hematopoiesis of indeterminate potential (CHIP) and is associated with adverse outcomes. Targeted sequencing of the stem cell product from 629 MM patients treated by ASCT at the Dana-Farber Cancer Institute (2003–2011) detects CHIP in 136/629 patients (21.6%). The most commonly mutated genes are DNMT3A, TET2, TP53, ASXL1 and PPM1D. Twenty-one from fifty-six patients (3.3%) receiving first-line IMiD maintenance develop a therapy-related myeloid neoplasm (TMN). However, regardless of CHIP status, the use of IMiD maintenance associates with improved PFS and OS. In those not receiving IMiD maintenance, CHIP is associated with decreased overall survival (OS) (HR:1.34, p = 0.02) and progression free survival (PFS) (HR:1.45, p, Multiple myeloma (MM) is treated with induction chemotherapy, autologous stem cell transplant (ASCT) and long-term immunomodulatory drug (IMiD) maintenance. Here, the authors show that the presence of clonal haematopoiesis of indeterminate potential (CHIP) at time of ASCT is associated with adverse outcomes in MM patients.

Alterations in DNA damage response (DDR) genes are common in advanced prostate tumors and are associated with unique genomic and clinical features. ATM is a DDR kinase that has a central role in coordinating DNA repair and cell-cycle response following DNA damage, and ATM alterations are present in approximately 5% of advanced prostate tumors. Recently, inhibitors of PARP have demonstrated activity in advanced prostate tumors harboring DDR gene alterations, particularly in tumors with BRCA1/2 alterations. However, the role of alterations in DDR genes beyond BRCA1/2 in mediating PARP inhibitor sensitivity is poorly understood. To define the role of ATM loss in prostate tumor DDR function and sensitivity to DDR-directed agents, we created a series of ATM-deficient preclinical prostate cancer models and tested the impact of ATM loss on DNA repair function and therapeutic sensitivities. ATM loss altered DDR signaling, but did not directly impact homologous recombination function. Furthermore, ATM loss did not significantly impact sensitivity to PARP inhibition but robustly sensitized to inhibitors of the related DDR kinase ATR. These results have important implications for planned and ongoing prostate cancer clinical trials and suggest that patients with tumor ATM alterations may be more likely to benefit from ATR inhibitor than PARP inhibitor therapy. Significance: ATM loss occurs in a subset of prostate tumors. This study shows that deleting ATM in prostate cancer models does not significantly increase sensitivity to PARP inhibition but does sensitize to ATR inhibition. See related commentary by Setton and Powell, p. 2085

PD-1 blockade has transformed the management of advanced clear cell renal cell carcinoma (ccRCC), but the drivers and resistors of the PD-1 response remain incompletely elucidated. Here, we analyzed 592 tumors from patients with advanced ccRCC enrolled in prospective clinical trials of treatment with PD-1 blockade by whole-exome and RNA sequencing, integrated with immunofluorescence analysis, to uncover the immunogenomic determinants of the therapeutic response. Although conventional genomic markers (such as tumor mutation burden and neoantigen load) and the degree of CD8+ T cell infiltration were not associated with clinical response, we discovered numerous chromosomal alterations associated with response or resistance to PD-1 blockade. These advanced ccRCC tumors were highly CD8+ T cell infiltrated, with only 27% having a non-infiltrated phenotype. Our analysis revealed that infiltrated tumors are depleted of favorable PBRM1 mutations and enriched for unfavorable chromosomal losses of 9p21.3, as compared with non-infiltrated tumors, demonstrating how the potential interplay of immunophenotypes with somatic alterations impacts therapeutic efficacy. A pooled genetic, transcriptomic and immunopathologic analysis of over 500 tumors from patients with advanced renal cell cancer suggests that response to PD-1 blockade depends on both CD8+ T cell infiltration and enrichment of tumor-intrinsic somatic alterations.

BACKGROUND: Inherited germline TP53 pathogenic and likely pathogenic mutations (gTP53) cause autosomal dominant multicancer predisposition including Li-Fraumeni syndrome (LFS). However, there is no known association of prostate cancer with gTP53. OBJECTIVE: To determine whether gTP53 predisposes to prostate cancer. DESIGN, SETTING, AND PARTICIPANTS: This multi-institutional retrospective study characterizes prostate cancer incidence in a cohort of LFS males and gTP53 prevalence in a prostate cancer cohort. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We evaluated the spectrum of gTP53 variants and clinical features associated with prostate cancer. RESULTS AND LIMITATIONS: We identified 31 prostate cancer cases among 163 adult LFS males, including 26 of 54 aged ≥50 yr. Among 117 LFS males without prostate cancer at the time of genetic testing, six were diagnosed with prostate cancer over a median (interquartile range [IQR]) of 3.0 (1.3–7.2) yr of follow-up, a 25-fold increased risk (95% confidence interval [CI] 9.2–55; p < 0.0001). We identified gTP53 in 38 of 6850 males (0.6%) in the prostate cancer cohort, a relative risk 9.1-fold higher than that of population controls (95% CI 6.2–14; p < 0.0001; gnomAD). We observed hotspots at the sites of attenuated mutations not associated with classic LFS. Two-thirds of available gTP53 prostate tumors had somatic inactivation of the second TP53 allele. Among gTP53 prostate cancer cases in this study, the median age at diagnosis was 56 (IQR: 51–62) yr, 42% had Gleason ≥8 tumors, and 30% had advanced disease at diagnosis. CONCLUSIONS: Complementary analyses of prostate cancer incidence in LFS males and gTP53 prevalence in prostate cancer cohorts suggest that gTP53 predisposes to aggressive prostate cancer. Prostate cancer should be considered as part of LFS screening protocols and TP53 considered in germline prostate cancer susceptibility testing. PATIENT SUMMARY: Inherited mutations in the TP53 gene are likely to predispose men to aggressive prostate cancer.

Bladder cancer is a genetically heterogeneous disease and novel therapeutic strategies are needed to expand treatment options and improve clinical outcomes. Here we identified a unique subset of urothelial tumors with focal amplification of the RAF1 (CRAF) kinase gene. RAF1-amplified tumors had activation of the RAF/MEK/ERK signaling pathway and exhibited a luminal gene expression pattern. Genetic studies demonstrated that RAF1-amplified tumors were dependent upon RAF1 activity for survival, and RAF1-activated cell lines and patient-derived models were sensitive to available and emerging RAF inhibitors as well as combined RAF plus MEK inhibition. Furthermore, we found that bladder tumors with HRAS or NRAS activating mutations were dependent on RAF1-mediated signaling and were sensitive to RAF1-targeted therapy. Together, these data identified RAF1 activation as a novel dependency in a subset comprising nearly 20% of urothelial tumors and suggested that targeting RAF1-mediated signaling represents a rationale therapeutic strategy.

Imaging datasets in cancer research are growing exponentially in both quantity and information density. These massive datasets may enable derivation of insights for cancer research and clinical care, but only if researchers are equipped with the tools to leverage advanced computational analysis approaches such as machine learning and artificial intelligence. In this work, we highlight three themes to guide development of such computational tools: scalability, standardization, and ease of use. We then apply these principles to develop PathML, a general-purpose research toolkit for computational pathology. We describe the design of the PathML framework and demonstrate applications in diverse use-cases. PathML is publicly available at www.pathml.com.

Computer vision (CV) approaches applied to digital pathology have informed biological discovery and development of tools to help inform clinical decision-making. However, batch effects in the images have the potential to introduce spurious confounders and represent a major challenge to effective analysis and interpretation of these data. Standard methods to circumvent learning such confounders include (i) application of image augmentation techniques and (ii) examination of the learning process by evaluating through external validation (e.g., unseen data coming from a comparable dataset collected at another hospital). Here, we show that the source site of a histopathology slide can be learned from the image using CV algorithms in spite of image augmentation, and we explore these source site predictions using interpretability tools. A CV model trained using Empirical Risk Minimization (ERM) risks learning this source-site signal as a spurious correlate in the weak-label regime, which we abate by using a training method with abstention. We find that a patch based classifier trained using abstention outperformed a model trained using ERM by 9.9, 10 and 19.4% F1 in the binary classification tasks of identifying tumor versus normal tissue in lung adenocarcinoma, Gleason score in prostate adenocarcinoma, and tumor tissue grade in clear cell renal cell carcinoma, respectively, at the expense of up to 80% coverage (defined as the percent of tiles not abstained on by the model). Further, by examining the areas abstained by the model, we find that the model trained using abstention is more robust to heterogeneity, artifacts and spurious correlates in the tissue. Thus, a method trained with abstention may offer novel insights into relevant areas of the tissue contributing to a particular phenotype. Together, we suggest using data augmentation methods that help mitigate a digital pathology model’s reliance on potentially spurious visual features, as well as selecting models that can identify features truly relevant for translational discovery and clinical decision support.

INTRODUCTION AND OBJECTIVE:Radiation associated muscle-invasive bladder cancer (RA-MIBC) is a potentially lethal late toxicity following definitive radiation for prostate cancer. RA-MIBC can be dif...

In recent years, therapy with immune modulating antibodies, termed immune checkpoint blockade (ICB), has revolutionized the treatment of advanced metastatic melanoma, yielding long-lasting clinical responses in a subgroup of patients. But despite this remarkable progress, resistance to therapy represents a major clinical challenge. ICB efficacy is critically dependent on cytotoxic CD8+ T cells targeting tumor cells in an HLA class I (HLA-I) antigen-dependent manner. Transcriptional suppression of the HLA-I antigen processing and presentation machinery (HLA-I APM) in melanoma cells leads to HLA-I-low/-negative tumor cell phenotypes escaping CD8+ T cell recognition and contributing to ICB resistance. In general, HLA-I-low/-negative tumor cells can be re-sensitized to T cells by interferons (IFN), augmenting HLA-I APM expression. However, this mechanism fails when melanoma cells acquire resistance to IFN, which recently turned out as a key resistance mechanism in ICB, besides HLA-I APM suppression. Seeking for a strategy to overcome these barriers, we identified a novel mechanism that restores HLA-I antigen presentation in tumor cells independent of IFN (Such et al. (2020) J Clin Invest, doi: 10.1172/JCI131572). We demonstrated that tumor cell-intrinsic activation of the cytosolic innate immunoreceptor RIG-I by its synthetic ligand 3pRNA overcomes transcriptional HLA-I APM suppression in patient-derived IFN-resistant melanoma cells. De novo HLA-I APM expression is IRF1/IRF3-dependent and re-sensitizes melanoma cells to autologous cytotoxic CD8+ T cells. Notably, synthetic RIG-I ligands and ICB synergize in T cell activation, suggesting combinational therapy could be an efficient strategy to improve patient outcomes in melanoma.

Introduction: mTNBC progresses rapidly on first-line chemotherapy. For instance, in the TNT trial, median progression-free survival (PFS) was 3 months on first-line carboplatin. Selective inhibition of Wee1, a negative regulator of the G2 cell cycle checkpoint, may enhance the efficacy of DNA-damaging agents by reducing DNA damage repair. We report the results of the first phase II study assessing the efficacy of the Wee1 inhibitor adavosertib (AZD-1775) with cisplatin in mTNBC. Methods: mTNBC patients (pts) with 0-1 prior lines of chemotherapy in the metastatic setting were eligible. Pts received cisplatin 75 mg/m2 IV followed by combination therapy 21 d later with cisplatin plus adavosertib 200 mg oral twice daily x 5 doses over 2.5 d every 21 d. Tumor biopsies were mandatory for patients with safely accessible tumors, and occurred 5-48 h after monotherapy and 5-8 h after the last adavosertib dose in the first combination therapy cycle. The study used a Simon optimal two-stage design that had 90% power to detect the difference between null (20%) and alternative (40%) response rates with a one-sided type I error of 0.1. The primary endpoint was objective response rate (ORR): an ORR > 30% would identify the regimen as worthy of further study. Key secondary endpoints were PFS and overall survival (OS). Exploratory analyses evaluated the association of response, defined as clinical benefit rate (CBR) > 6 months (CR + PR + SD > 6 mos), with transcriptional and immunostaining profiles of on-treatment tumors (n = 16), as well as with targeted panel genomic alterations in archival tissue. Tumor immune cell composition was assessed by CIBERSORT. Results: 34 pts initiated protocol therapy; median age was 56 yrs, 2 (6%) pts had known BRCA2 mutations, and 14 (41%) had 1 prior line of chemotherapy. Median follow-up was 13 mos. ORR was 29% (3 CR + 7 PR), median PFS 4.9 months (95% CI, 2.3-5.8 mos), CBR 35%, and preliminary OS 14.0 mos (95% CI, 11.8-21.8 mos). All-cause AEs occurred in 100% of pts (G3-4, 56%; most commonly diarrhea, 21%), including one death due to sepsis possibly related to study therapy. The primary tumor of the longest responding patient had a possible biallelic loss-of-function alteration (missense mutation and loss of heterozygosity) in the homologous recombination-related gene FANCM. In Hallmark gene set enrichment analyses of 26 on-treatment biopsies across 16 pts, responding tumors (n = 4) demonstrated enriched expression of immune response gene sets (allograft rejection FDR q < 0.001, IL2-STAT5 signaling FDR q = 0.001, inflammatory response FDR q = 0.003), while non-responding tumors (n = 12) showed enrichment of cell cycle gene sets (E2F targets FDR q < 0.001, G2M checkpoint FDR q < 0.001). In the same on-treatment biopsies, responding tumors (n = 4) had higher tumor infiltrating lymphocytes (46% v. 29%, Mann-Whitney p = 0.04) and lower M0 macrophages (3% v. 17%, Mann-Whitney p = 0.04) than non-responding tumors (n =12). Updated data, including OS and tissue immunostaining results, will be presented. Conclusions: Among mTNBC pts, the combination of adavosertib and cisplatin was associated with a 29% ORR, failing to meet the pre-determined ORR of > 30% required to support further investigation. Responses were associated with immune-related gene expression and TILs detected by RNA sequencing. Citation Format: Tanya E. Keenan, Zhenying Tan-Wasielewski, Lorenzo Trippa, Bose Kochupurakkal, Jennifer L. Guerriero, Rie K. Tahara, Grace Winship, Wafa Osmani, Chelsea Andrews, Jake R. Conway, Meng X He, Ricardo Pastorello, Adam Tracy, Robert E. Godin, Beth A. Overmoyer, Eric P. Winer, Elizabeth A. Mittendorf, Geoffrey I Shapiro, Eliezer M Van Allen, Sara M Tolaney. A phase II study of cisplatin and the Wee1 inhibitor adavosertib in metastatic triple-negative breast cancer (mTNBC) [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P3-10-08.

Background In metastatic urothelial carcinoma (mUC), predictive biomarkers that correlate with response to immune checkpoint inhibitors (ICIs) are lacking. Here, we interrogated genomic and clinical features associated with response to ICIs in mUC. Methods Sixty two mUC patients treated with ICI who had targeted tumour sequencing were studied. We examined associations between candidate biomarkers and clinical benefit (CB, any objective reduction in tumour size) versus no clinical benefit (NCB, no change or objective increase in tumour size). Both univariable and multivariable analyses for associations were conducted. A comparator cohort of 39 mUC patients treated with taxanes was analysed by using the same methodology. Results Nine clinical and seven genomic factors correlated with clinical outcomes in univariable analysis in the ICI cohort. Among the 16 factors, neutrophil-to-lymphocyte ratio (NLR) ≥5 (OR = 0.12, 95% CI, 0.01–1.15), visceral metastasis (OR = 0.05, 95% CI, 0.01–0.43) and single-nucleotide variant (SNV) count Conclusions This three-factor model includes genomic (SNV count >9) and clinical (NLR

Purpose To date, there has not been a large, systematic evaluation of the prevalence of germline risk variants in urothelial carcinoma (UC). Methods We evaluated the frequency of germline pathogenic and likely pathogenic variants in 1038 patients with high-risk UC who underwent targeted clinical germline testing. Case–control enrichment analysis was performed to screen for pathogenic variant enrichment in 17 DNA repair genes in 1038 UC patients relative to cancer-free individuals. Results Among 1038 patients with UC, the cumulative frequency of patients with pathogenic variants was 24%; 18.6% of patients harbored ≥1 actionable germline variant with preventive or therapeutic utility. MSH2 (34/969, 3.5%) and BRCA1/2 (38/867, 4.4%) germline variants had the highest frequency. Germline variants in DNA damage repair genes accounted for 78% of pathogenic germline variants. Compared to the cancer-free cohort, UC patients had significant variant enrichment in MSH2 (odds ratio [OR]: 15.4, 95% confidence interval [CI]: 7.1–32.7, p

Despite remarkable success of immune checkpoint inhibitors, the majority of cancer patients have yet to receive durable benefits. Here, in order to investigate the metabolic alterations in response to immune checkpoint blockade, we comprehensively profile serum metabolites in advanced melanoma and renal cell carcinoma patients treated with nivolumab, an antibody against programmed cell death protein 1 (PD1). We identify serum kynurenine/tryptophan ratio increases as an adaptive resistance mechanism associated with worse overall survival. This advocates for patient stratification and metabolic monitoring in immunotherapy clinical trials including those combining PD1 blockade with indoleamine 2,3-dioxygenase/tryptophan 2,3-dioxygenase (IDO/TDO) inhibitors., Immune-checkpoint inhibition therapy has achieved success in a subset of patients. Here the authors profiled about 200 relevant metabolites in patient serum samples from three independent immunotherapy trials and found the serum kynurenine/tryptophan ratio increases to be associated with worse overall survival.

Background TP53, PTEN, and RB1 tumor suppressor genes (TSGs) are recurrently altered in treatment-resistant prostate cancer. Cooperative loss of two or more TSGs may drive more aggressive disease. Objective To determine clinical outcomes of single and compound TSG alterations across the spectrum of prostate cancer. Design, setting, and participants Massively parallel targeted sequencing using castration-sensitive prostate cancer (CSPC; localized [L] and metastatic [M1]) and castration-resistant prostate cancer (CRPC) specimens (n = 285). TSG altered (TSG-alt) was any copy number loss or deleterious mutation of one or more TSGs (TP53, PTEN, and RB1). Outcome measurements and statistical analysis For L-CSPC, event-free survival (EFS) and time to CRPC were estimated. For M1-CSPC and M1-CRPC, overall survival (OS) was estimated. Cox regression models assessed the association between cumulative TSG hits (zero hits vs one hit vs two to three hits) and outcomes with multivariable analyses adjusted for clinicopathological factors. Results and limitations TSG variants increased with advanced disease (L-CSPC: 39%; M1-CSPC: 63%, M1-CRPC: 92%). TSG-alt L-CSPC had shorter EFS (median 2.6 yr, hazard ratio [HR] 1.95, 95% confidence interval [CI] 1.22–3.13) and time to CRPC (median 9.5 mo, HR 3.36, 95% CI 1.01–11.16). Cumulative gene hits led to an incremental risk of relapse (EFS: one gene, HR 1.69, 95% CI 0.99–2.87; two to three genes, HR 2.70, 95% CI 1.43–5.08; both versus zero genes, p = 0.004). There was evidence of inferior OS with increasing TSG hits in the metastatic cohorts. Only four (8%) patients in the M1-CRPC cohort were TSG-neg, one of whom died after 5.2 yr. Multivariable analyses adjusting for mutational and copy number burden did not demonstrate a significant independent association of increasing gene hits and poorer outcomes. Conclusions Deleterious TSG variants are associated with an increased risk of relapse (L) and death (M1) in CSPC. Poorer outcomes are seen with compound gene hits in both early and advanced disease, and this may in part reflect increasing global genomic instability. Patient summary Men with prostate tumors with compound tumor suppressor gene mutations have poorer outcomes. These findings help identify patients with aggressive features who may benefit from intensified treatment.

The emergence of immune checkpoint blockade therapies over the last decade has transformed cancer treatment in a wide range of tumor types. Unprecedented and durable clinical responses in difficult-to-treat cancer histologies have been observed. However, despite these promising long-term responses, the majority of patients fail to respond to immune checkpoint blockade, demonstrating primary resistance. Additionally, many of those who initially respond to treatment eventually experience relapse secondary to acquired resistance. Both primary and acquired resistance are a result of complex and constantly evolving interactions between cancer cells and the immune system. Many mechanisms of resistance have been characterized to date, and more continue to be uncovered. By elucidating and targeting mechanisms of resistance, treatments can be tailored to improve clinical outcomes. This review will discuss the landscape of immune checkpoint blockade response data, different resistance mechanisms, and potential therapeutic strategies to overcome resistance.

Purpose: The purpose of this study is to characterize the mutational landscape across the spectrum of urothelial carcinoma (UC) to identify mutational features and potential therapeutic targets. Experimental Design: Using targeted exome sequencing (n = 237 genes), we analyzed the mutation spectra of 82 low-grade nonmuscle-invasive bladder cancers (LG-NMIBC), 126 high-grade (HG) NMIBC, 199 muscle-invasive bladder cancers (MIBC), 10 LG-upper tract urothelial cancers (LG-UTUC), and 55 HG-UTUC. Results: FGFR3 and KDM6A mutations were significantly more common in LG-NMIBC (72% and 44%, respectively) versus other bladder subtypes. FGFR3 alterations were also enriched in LG-UTUC versus HG-UTUC tumors (80% vs. 16%). In contrast, TP53 and RB1 mutations were significantly more frequent in all 3 HG urothelial carcinoma subtypes than in LG-NIMBC (45%–58% vs. 4%; 9%–22% vs. 0; respectively). Among LG-NMIBC tumors, KDM6A mutations were more common in women than in men (71% vs. 38%). HG-NMIBC and MIBC had higher tumor mutational burden (TMB) than LG-NMIBC (P = 0.001 and P < 0.01, respectively). DNA-damage repair (DDR) alterations were associated with a higher TMB in HG-NMIBC and MIBC tumors, and these two tumor types were also enriched for an APOBEC mutational signature compared with LG-NMIBC and HG-UTUC. Alterations in FGFR3, PIK3CA, and EP300 correlated with worse overall survival in HG-UTUC and occurred concurrently. Conclusions: Our analysis suggests that a fraction of MIBCs likely arise from precursor lesions other than LG-NMIBC. KDM6A mutations are twice as common in women with LG-NIMBC than those in men. DDR gene mutations and APOBEC mutagenesis drive mutations in HG-NMIBC and MIBC. UTUC has a distinct mutation profile from bladder cancer.

This single-arm phase II study investigated the efficacy and safety of cabozantinib combined with nivolumab in metastatic triple-negative breast cancer (mTNBC). The primary endpoint was objective response rate (ORR) by RECIST 1.1. Biopsies at baseline and after cycle 1 were analyzed for tumor-infiltrating lymphocytes (TILs), PD-L1, and whole-exome and transcriptome sequencing. Only 1/18 patients achieved a partial response (ORR 6%), and the trial was stopped early. Toxicity led to cabozantinib dose reduction in 50% of patients. One patient had a PD-L1-positive tumor, and three patients had TILs > 10%. The responding patient had a PD-L1-negative tumor with low tumor mutational burden but high TILs and enriched immune gene expression. High pretreatment levels of plasma immunosuppressive cytokines, chemokines, and immune checkpoint molecules were associated with rapid progression. Although this study did not meet its primary endpoint, immunostaining, genomic, and proteomic studies indicated a high degree of tumor immunosuppression in this mTNBC cohort.

Molecular profiling studies have enabled numerous discoveries for metastatic prostate cancer (MPC), but they have mostly occurred in academic medical institutions focused on select patient populations. We developed the Metastatic Prostate Cancer Project (MPCproject, mpcproject.org), a patient-partnered initiative to empower MPC patients living anywhere in the U.S. and Canada to participate in molecular research and contribute directly to translational discovery. Here we present clinicogenomic results from our partnership with the first 706 MPCproject participants. We found that a patient-centered and remote research strategy enhanced engagement with patients in rural and medically underserved areas. Furthermore, patient-reported data achieved 90% consistency with abstracted health records for therapies and provided a mechanism for patient-partners to share information about their cancer experience not documented in medical records. Among the molecular profiling data from 333 patient-partners (n = 573 samples), whole exome sequencing of 63 tumor samples obtained from hospitals across the U.S. and Canada and 19 plasma cell-free DNA (cfDNA) samples from blood donated remotely recapitulated known findings in MPC and enabled longitudinal study of prostate cancer evolution. Inexpensive ultra-low coverage whole genome sequencing of 318 cfDNA samples from donated blood revealed clinically relevant genomic changes like AR amplification, even in the context of low tumor burden. Collectively, this study illustrates the power of a longitudinal partnership with patients to generate a more representative clinical and molecular understanding of MPC.NoteTo assist our patient-partners and the wider MPC community interpret the results of this study, we have included a glossary of terms in the Supplementary Materials.

Translocation renal cell carcinoma (tRCC) is an aggressive and poorly-characterized subtype of kidney cancer driven by MiT/TFE gene fusions. Here, we define the landmarks of tRCC through an integrative analysis of 152 tRCC patients identified across multiple genomic, clinical trial, and retrospective cohorts. Most tRCCs harbor few somatic alterations apart from MiT/TFE fusions and homozygous deletions at chromosome 9p21.3 (19.2% of cases). Transcriptionally, tRCCs display a heightened NRF2-driven antioxidant response that is associated with resistance to many targeted therapies. Consistently, we find that outcomes for tRCC patients treated with vascular endothelial growth factor receptor inhibitors (VEGFR-TKI) are worse than those treated with immune checkpoint inhibition (ICI). Multiparametric immunofluorescence confirmed the presence of CD8 + tumor-infiltrating T cells compatible with a clinical benefit from ICI and revealed an exhaustion immunophenotype distinct from clear cell RCC. Our findings comprehensively define the clinical and molecular features of tRCC and may inspire new therapeutic hypotheses.

Determination of molecular features that mediate clinically aggressive phenotypes in prostate cancer (PrCa) remains a major biological and clinical challenge. Here, we developed a biologically informed deep learning model (P-NET) to stratify PrCa patients by treatment resistance state and evaluate molecular drivers of treatment resistance for therapeutic targeting through complete model interpretability. Using a molecular cohort of 1,238 prostate cancers, we demonstrated that P-NET can predict cancer state using molecular data that is superior to other modeling approaches. Moreover, the biological interpretability within P-NET revealed established and novel molecularly altered candidates, such as MDM4 and FGFR1, that were implicated in predicting advanced disease and validated in vitro. Broadly, biologically informed fully interpretable neural networks enable preclinical discovery and clinical prediction in prostate cancer and may have general applicability across cancer types.

When the COVID-19 pandemic began, formal frameworks to collect data about affected patients were lacking. The COVID-19 and Cancer Consortium (CCC19) was formed to collect granular data on patients with cancer and COVID-19 at scale and as rapidly as possible. CCC19 has grown from five initial institutions to 125 institutions with >400 collaborators. More than 5,000 cases with complete baseline data have been accrued. Future directions include increased electronic health record integration for direct data ingestion, expansion to additional domestic and international sites, more intentional patient involvement, and granular analyses of still-unanswered questions related to cancer subtypes and treatments.

Background Intense neoadjuvant androgen deprivation therapy (ADT) before radical prostatectomy (RP) is an investigational approach to reduce recurrence rates in men with high-risk localized prostate cancer (PCa). The impact of germline DNA damage repair (gDDR) gene alterations on response to intense neoadjuvant ADT is not known. Objective To evaluate the prevalence of gDDR alterations among men with localized PCa at high risk of recurrence and evaluate their impact on response to intense neoadjuvant ADT. Design, setting, and participants We performed germline panel sequencing for 201 men with intermediate- and high-risk localized PCa from five randomized multicenter clinical trials of intense neoadjuvant ADT before RP. Intervention Intense neoadjuvant ADT followed by RP. Outcome measurements and statistical analysis The prevalence of pathogenic gDDR alterations and their association with exceptional pathologic response (complete response or minimal residual disease, defined as residual tumor with the largest cross-section dimension ≤5 mm) to intense neoadjuvant ADT and rates of post-RP biochemical recurrence. Results and limitations Pathogenic gDDR alterations were detected in 19 (9.5%) of the 201 PCa patients. The most frequently altered genes were BRCA2 (n = 6; 3.0%) and ATM (n = 4; 2.0%). Patients with gDDR alterations exhibited similar rates of exceptional pathologic response (26% vs 22%), pT3 disease (42% vs 53%), lymph node involvement (5.3% vs 10%), extraprostatic extension (35% vs 54%), and positive margins (5.3% vs 13%) to patients without gDDR alterations (all p > 0.05). The 3-yr biochemical recurrence–free survival was also similar at 45% (95% confidence interval 7.9–78%) for men with gDDR alterations and 55% (95% confidence interval 44–64%) for men without gDDR alterations. Conclusions gDDR alterations are common among men with intermediate- and high-risk localized PCa. Men with gDDR alterations appear to have a comparable response to intense neoadjuvant ADT to that among men without gDDR alterations and should not be excluded from consideration for this treatment approach. Patient summary Intense therapy to inhibit the production of androgen hormones (eg, testosterone) before surgery may minimize the risk of cancer recurrence for men with high-risk localized prostate cancer. Inherited mutations in certain DNA repair genes are associated with particularly high rates of recurrence. We found that men with these mutations respond equally well to this intense androgen inhibition before surgery as men without the mutations.

Background High tumor mutational burden (TMB) and PD-L1 expression are associated with improved clinical outcomes in patients (pts) with non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICIs). However, how TMB performs as a predictive biomarker to ICIs in different PD-L1 expression subgroups is not well characterized. Methods We collected clinicopathologic and genomic data from NSCLCs which underwent targeted NGS and TMB assessment at DFCI. An unbiased recursive partitioning (URP) algorithm was used to investigate an optimal TMB cut-off with respect to objective response rate (ORR) in the subset of pts treated with ICIs. This TMB cut-off was then validated in the prospective POPLAR/OAK cohort. Results Among 3560 NSCLCs identified, median TMB was significantly higher among current smokers compared to former (P 25%, suggesting that TMB differentially impacts response to immunotherapy among PD-L1 high versus low NSCLCs (figure 5). Conclusions The impact of TMB may vary across PD-L1 expression subgroups. Rational integration of TMB and PD-L1 expression may identify NSCLCs with the greatest likelihood of response or resistance to ICIs. Ethics Approval Clinicopathologic data were collected from patients with advanced NSCLC who had consented to a correlative research study (DF/HCC protocol #02-180).

Medulloblastoma is among the most common malignant brain tumors in children. Recent studies have identified at least four subgroups of the disease that differ in terms of molecular characteristics and patient outcomes. Despite this heterogeneity, most patients with medulloblastoma receive similar therapies, including surgery, radiation, and intensive chemotherapy. Although these treatments prolong survival, many patients still die from the disease and survivors suffer severe long-term side effects from therapy. We hypothesize that each patient with medulloblastoma is sensitive to different therapies and that tailoring therapy based on the molecular and cellular characteristics of patients' tumors will improve outcomes. To test this, we assembled a panel of orthotopic patient-derived xenografts (PDX) and subjected them to DNA sequencing, gene expression profiling, and high-throughput drug screening. Analysis of DNA sequencing revealed that most medulloblastomas do not have actionable mutations that point to effective therapies. In contrast, gene expression and drug response data provided valuable information about potential therapies for every tumor. For example, drug screening demonstrated that actinomycin D, which is used for treatment of sarcoma but rarely for medulloblastoma, was active against PDXs representing Group 3 medulloblastoma, the most aggressive form of the disease. Functional analysis of tumor cells was successfully used in a clinical setting to identify more treatment options than sequencing alone. These studies suggest that it should be possible to move away from a one-size-fits-all approach and begin to treat each patient with therapies that are effective against their specific tumor. Significance: These findings show that high-throughput drug screening identifies therapies for medulloblastoma that cannot be predicted by genomic or transcriptomic analysis.

Individual tumor molecular profiling is routinely used to detect single gene-variant (“first-order”) genomic alterations that may inform therapeutic actions -- for instance, a tumor with aBRAFp.V600E variant might be considered for RAF/MEK inhibitor therapy. Interactions between such first-order events (e.g., somatic-germline) and global molecular features (e.g. mutational signatures) are increasingly associated with clinical outcomes, but these “second order” alterations are not yet generally accounted for in clinical interpretation algorithms and knowledge bases. Here, we introduce the Molecular Oncology Almanac (MOAlmanac), a clinical interpretation algorithm paired with a novel underlying knowledge base to enable integrative interpretation of genomic and transcriptional cancer data for point-of-care treatment decision-making and translational hypothesis generation. We compared MOAlmanac to first-order interpretation methodology in multiple retrospective patient cohorts and observed that the inclusion of preclinical and inferential evidence as well as second-order molecular features increased the number of nominated clinical hypotheses. MOAlmanac also performed matchmaking between patient molecular profiles and cancer cell lines to further expand individualized clinical actionability. When applied to a prospective precision oncology trial cohort, MOAlmanac nominated a median of two therapies per patient and identified therapeutic strategies administered in 46% of patient profiles. Overall, we present a novel computational method to perform integrative clinical interpretation of individualized molecular profiles. MOAlmanc increases clinical actionability over conventional approaches by considering second-order molecular features and additional evidence sources, and is available as an open-source framework.

IMPORTANCE: Prior studies have shown that only a small proportion of patients with hormone receptor (HR)–positive metastatic breast cancer (MBC) experience benefit from programmed cell death 1 (PD-1)/programmed cell death ligand 1 (PD-L1) inhibitors given as monotherapy. There are data suggesting that activity may be greater with combination strategies. OBJECTIVE: To compare the efficacy of eribulin plus pembrolizumab vs eribulin alone in patients with HR-positive, ERBB2 (formerly HER2)–negative MBC. DESIGN, SETTING, AND PARTICIPANTS: Multicenter phase 2 randomized clinical trial of patients with HR-positive, ERBB2-negative MBC who had received 2 or more lines of hormonal therapy and 0 to 2 lines of chemotherapy. INTERVENTIONS: Patients were randomized 1:1 to eribulin, 1.4 mg/m(2) intravenously, on days 1 and 8 plus pembrolizumab, 200 mg/m(2) intravenously, on day 1 of a 21-day cycle or eribulin alone. At time of progression, patients in the eribulin monotherapy arm could cross over and receive pembrolizumab monotherapy. MAIN OUTCOMES AND MEASURES: The primary end point was progression-free survival (PFS). Secondary end points were objective response rate (ORR) and overall survival (OS). Exploratory analyses assessed the association between PFS and PD-L1 status, tumor-infiltrating lymphocytes (TILs), tumor mutational burden (TMB), and genomic alterations. RESULTS: Eighty-eight patients started protocol therapy; the median (range) age was 57 (30-76) years, median (range) number of prior lines of chemotherapy was 1 (0-2), and median (range) number of prior lines of hormonal therapy was 2 (0-5). Median follow-up was 10.5 (95% CI, 0.4-22.8) months. Median PFS and ORR were not different between the 2 groups (PFS, 4.1 vs 4.2 months; hazard ratio, 0.80; 95% CI, 0.50-1.26; P = .33; ORR, 27% vs 34%, respectively; P = .49). Fourteen patients started crossover treatment with pembrolizumab; 1 patient experienced stable disease. All-cause adverse events occurred in all patients (grade ≥3, 65%) including 2 treatment-related deaths in the combination group, both from immune-related colitis in the setting of sepsis, attributed to both drugs. The PD-L1 22C3 assay was performed on archival tumor samples in 65 patients: 24 (37%) had PD-L1–positive tumors. Analysis indicated that PD-L1 status, TILs, TMB, and genomic alterations were not associated with PFS. CONCLUSIONS AND RELEVANCE: In this randomized clinical trial of patients with HR-positive, ERBB2-negative MBC, the addition of pembrolizumab to eribulin did not improve PFS, ORR, or OS compared with eribulin alone in either the intention-to-treat or PD-L1–positive populations. Further efforts to explore the benefits of adding checkpoint inhibition to chemotherapy among less heavily pretreated patients are needed. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03051659

High-grade T1 (HGT1) bladder cancer is the highest risk subtype of non–muscle-invasive bladder cancer with unpredictable outcome and poorly understood risk factors. Here, we examined the association of somatic mutation profiles with nonrecurrent disease (GO, good outcome), recurrence (R), or progression (PD) in a cohort of HGT1 patients. Exome sequencing was performed on 62 HGT1 and 15 matched normal tissue samples. Both tumor only (TO) and paired analyses were performed, focusing on 95 genes known to be mutated in bladder cancer. Somatic mutations, copy-number alterations, mutation load, and mutation signatures were studied. Thirty-three GO, 10 R, 18 PD, and 1 unknown outcome patients were analyzed. Tumor mutational burden (TMB) was similar to muscle-invasive disease and was highest in GO, intermediate in PD, and lowest in R patients (P = 0.017). DNA damage response gene mutations were associated with higher TMB (P < 0.0001) and GO (P = 0.003). ERCC2 and BRCA2 mutations were associated with GO. TP53, ATM, ARID1A, AHR, and SMARCB1 mutations were more frequent in PD. Focal copy-number gain in CCNE1 and CDKN2A deletion was enriched in PD or R (P = 0.047; P = 0.06). APOBEC (46%) and COSMIC5 (34%) signatures were most frequent. APOBEC-A and ERCC2 mutant tumors (COSMIC5) were associated with GO (P = 0.047; P = 0.0002). pT1b microstaging was associated with a genomic cluster (P = 0.05) with focal amplifications of E2F3/SOX4, PVRL4, CCNE1, and TP53 mutations. Findings were validated using external public datasets. These findings require confirmation but suggest that management of HGT1 bladder cancer may be improved via molecular characterization to predict outcome. Significance: Detailed genetic analyses of HGT1 bladder tumors identify features that correlate with outcome, e.g., high mutational burden, ERCC2 mutations, and high APOBEC-A/ERCC2 mutation signatures were associated with good outcome.

Purpose: Cisplatin-based chemotherapy is a first-line treatment for muscle-invasive and metastatic urothelial cancer. Approximately 10% of bladder urothelial tumors have a somatic missense mutation in the nucleotide excision repair (NER) gene, ERCC2, which confers increased sensitivity to cisplatin-based chemotherapy. However, a significant subset of patients is ineligible to receive cisplatin-based therapy due to medical contraindications, and no NER-targeted approaches are available for platinum-ineligible or platinum-refractory ERCC2-mutant cases. Experimental Design: We used a series of NER-proficient and NER-deficient preclinical tumor models to test sensitivity to irofulven, an abandoned anticancer agent. In addition, we used available clinical and sequencing data from multiple urothelial tumor cohorts to develop and validate a composite mutational signature of ERCC2 deficiency and cisplatin sensitivity. Results: We identified a novel synthetic lethal relationship between tumor NER deficiency and sensitivity to irofulven. Irofulven specifically targets cells with inactivation of the transcription-coupled NER (TC-NER) pathway and leads to robust responses in vitro and in vivo, including in models with acquired cisplatin resistance, while having minimal effect on cells with intact NER. We also found that a composite mutational signature of ERCC2 deficiency was strongly associated with cisplatin response in patients and was also associated with cisplatin and irofulven sensitivity in preclinical models. Conclusions: Tumor NER deficiency confers sensitivity to irofulven, a previously abandoned anticancer agent, with minimal activity in NER-proficient cells. A composite mutational signature of NER deficiency may be useful in identifying patients likely to respond to NER-targeting agents, including cisplatin and irofulven. See related commentary by Jiang and Greenberg, p. 1833

Purpose: We report results from a phase II study assessing the efficacy of the WEE1 inhibitor adavosertib with cisplatin in metastatic triple-negative breast cancer (mTNBC). Patients and Methods: Patients with mTNBC treated with 0–1 prior lines of chemotherapy received cisplatin 75 mg/m2 i.v. followed 21 days later by cisplatin plus adavosertib 200 mg oral twice daily for five doses every 21 days. The study had 90% power to detect the difference between null (20%) and alternative (40%) objective response rates (ORR) with a one-sided type I error of 0.1: an ORR >30% was predefined as making the regimen worthy of further study. RNA sequencing and multiplex cyclic immunofluorescence on pre- and post-adavosertib tumor biopsies, as well as targeted next-generation sequencing on archival tissue, were correlated with clinical benefit, defined as stable disease ≥6 months or complete or partial response. Results: A total of 34 patients initiated protocol therapy; median age was 56 years, 2 patients (6%) had BRCA2 mutations, and 14 (41%) had one prior chemotherapy. ORR was 26% [95% confidence interval (CI), 13–44], and median progression-free survival was 4.9 months (95% CI, 2.3–5.7). Treatment-related grade 3–5 adverse events occurred in 53% of patients, most commonly diarrhea in 21%. One death occurred because of sepsis, possibly related to study therapy. Tumors from patients with clinical benefit demonstrated enriched immune gene expression and T-cell infiltration. Conclusions: Among patients with mTNBC treated with 0–1 prior lines, adavosertib combined with cisplatin missed the prespecified ORR cutoff of >30%. The finding of immune-infiltrated tumors in patients with clinical benefit warrants validation.

PURPOSE Although primary germ cell tumors (GCTs) have been extensively characterized, molecular analysis of metastatic sites has been limited. We performed whole-exome sequencing and targeted next-generation sequencing on paired primary and metastatic GCT samples in a patient cohort enriched for cisplatin-resistant disease. PATIENTS AND METHODS Tissue sequencing was performed on 100 tumor specimens from 50 patients with metastatic GCT, and sequencing of plasma cell-free DNA was performed for a subset of patients. RESULTS The mutational landscape of primary and metastatic pairs from GCT patients was highly discordant (68% of all somatic mutations were discordant). Whereas genome duplication was common and highly concordant between primary and metastatic samples, only 25% of primary-metastasis pairs had ≥ 50% concordance at the level of DNA copy number alterations (CNAs). Evolutionary-based analyses revealed that most mutations arose after CNAs at the respective loci in both primary and metastatic samples, with oncogenic mutations enriched in the set of early-occurring mutations versus variants of unknown significance (VUSs). TP53 pathway alterations were identified in nine cisplatin-resistant patients and had the highest degree of concordance in primary and metastatic specimens, consistent with their association with this treatment-resistant phenotype. CONCLUSION Analysis of paired primary and metastatic GCT specimens revealed significant molecular heterogeneity for both CNAs and somatic mutations. Among loci demonstrating serial genetic evolution, most somatic mutations arose after CNAs, but oncogenic mutations were enriched in the set of early-occurring mutations as compared with VUSs. Alterations in TP53 were clonal when present and shared among primary-metastasis pairs.

Summary Immune checkpoint blockade (ICB) results in durable disease control in a subset of patients with advanced renal cell carcinoma (RCC), but mechanisms driving resistance are poorly understood. We characterize the single-cell transcriptomes of cancer and immune cells from metastatic RCC patients before or after ICB exposure. In responders, subsets of cytotoxic T cells express higher levels of co-inhibitory receptors and effector molecules. Macrophages from treated biopsies shift toward pro-inflammatory states in response to an interferon-rich microenvironment but also upregulate immunosuppressive markers. In cancer cells, we identify bifurcation into two subpopulations differing in angiogenic signaling and upregulation of immunosuppressive programs after ICB. Expression signatures for cancer cell subpopulations and immune evasion are associated with PBRM1 mutation and survival in primary and ICB-treated advanced RCC. Our findings demonstrate that ICB remodels the RCC microenvironment and modifies the interplay between cancer and immune cell populations critical for understanding response and resistance to ICB., Graphical abstract, Highlights • Distinct CD8+ T cell phenotypes are enriched after immune checkpoint blockade (ICB) • Immune checkpoint ligands are upregulated in macrophages and tumor cells after ICB • Two cancer cell subpopulations are conserved across heterogeneous RCC tumors • Cancer cell programs drive distinct immune interactions and predict patient outcomes, Bi et al. dissect the cancer cell and immune microenvironmental transcriptional programs in immune checkpoint blockade-exposed metastatic renal cell carcinoma, revealing immune subpopulation reprogramming and interactions with distinct cancer cell populations in the context of clinical resistance.

Purpose: DNA-damaging agents comprise the backbone of systemic treatment for many tumor types; however, few reliable predictive biomarkers are available to guide use of these agents. In muscle-invasive bladder cancer (MIBC), cisplatin-based chemotherapy improves survival, yet response varies widely among patients. Here, we sought to define the role of the nucleotide excision repair (NER) gene ERCC2 as a biomarker predictive of response to cisplatin in MIBC. Experimental Design: Somatic missense mutations in ERCC2 are associated with improved response to cisplatin-based chemotherapy; however, clinically identified ERCC2 mutations are distributed throughout the gene, and the impact of individual ERCC2 variants on NER capacity and cisplatin sensitivity is unknown. We developed a microscopy-based NER assay to profile ERCC2 mutations observed retrospectively in prior studies and prospectively within the context of an institution-wide tumor profiling initiative. In addition, we created the first ERCC2-deficient bladder cancer preclinical model for studying the impact of ERCC2 loss of function. Results: We used our functional assay to test the NER capacity of clinically observed ERCC2 mutations and found that most ERCC2 helicase domain mutations cannot support NER. Furthermore, we show that introducing an ERCC2 mutation into a bladder cancer cell line abrogates NER activity and is sufficient to drive cisplatin sensitivity in an orthotopic xenograft model. Conclusions: Our data support a direct role for ERCC2 mutations in driving cisplatin response, define the functional landscape of ERCC2 mutations in bladder cancer, and provide an opportunity to apply combined genomic and functional approaches to prospectively guide therapy decisions in bladder cancer. See related commentary by Grivas, p. 907

Cancers of unknown primary (CUP) are among the most common causes of death due to cancer, are associated with a poor prognosis and have few therapeutic options available. Molecularly-guided site-specific treatments were explored based on the assumption that CUP are similar in their response to treatment of predicted primary tumours. Given the discordant results between these studies, a meta-analysis using a random-effects model and the inverse variance method was performed. MEDLINE and conference abstracts of American Society of Clinical Oncology (ASCO) and European Society of Medical Oncology (ESMO) meetings were searched from inception until November 2019. A trend towards improved OS was noted with site-specific versus empiric treatment for CUP (HR = 0.73; 95% confidence interval (CI) 0.52–1.02). There was significant heterogeneity across the four studies (I [2] = 79%; p = 0.002) but no significant difference was noted between the treatment effect in the two subgroups (randomised vs. non-randomised; p = 0.07). The test for overall effect for progression free survival, which had only been reported for the two randomised studies, was not statistically significant (HR = 0.93; 95% CI 0.74–1.17), with little heterogeneity between studies (I [2] = 0%; p = 0.77). The results of this meta-analysis highlight the significant heterogeneity between the prospective studies comparing molecularly tailored to empiric therapy for CUP and the need for other randomised studies including only primary tumors with available effective therapies.

Metastatic castration resistant prostate cancer (mCRPC) is primarily treated with therapies that prevent transcriptional activity of the androgen receptor (AR), cause DNA damage, or prevent cell division. Clinical resistance to these therapies, including second-generation androgen-targeting compounds such as enzalutamide and abiraterone, is nearly universal. Other treatment modalities, including immune checkpoint inhibitors, have provided minimal benefit except in rare subsets of patients1,2. Both tumour intrinsic and extrinsic cellular programs contributing to therapeutic resistance remain areas of active investigation. Here we use full-length single-cell RNA-sequencing (scRNA-seq) to identify the transcriptional states of cancer and immune cells in the mCRPC microenvironment. Within cancer cells, we identified transcriptional patterns that mediate a significant proportion of inherited risk for prostate cancer, extensive heterogeneity inARsplicing within and between tumours, and vastly divergent regulatory programs between adenocarcinoma and small cell carcinoma. Moreover, upregulation of TGF-β signalling and epithelial-mesenchymal transition (EMT) were both associated with resistance to enzalutamide. We found that some lymph node metastases, but no bone metastases, were heavily infiltrated by dysfunctional CD8+T cells, including cells undergoing dramatic clonal expansion during enzalutamide treatment. Our findings suggest avenues for rational therapeutic approaches targeting both tumour-intrinsic and immunological pathways to combat resistance to current treatment options.