Subcutaneous epcoritamab monotherapy in Japanese adults with relapsed/refractory diffuse large B-cell lymphoma.

Epcoritamab is a subcutaneously administered CD3xCD20 bispecific Ab that showed deep, durable responses with a manageable safety profile in patients with relapsed or refractory (R/R) diffuse large B-cell lymphoma (DLBCL) in the global multicenter pivotal phase II trial EPCORE NHL-1. Here, we present results from the similar EPCORE NHL-3 phase I/II trial evaluating epcoritamab monotherapy in Japanese patients with R/R CD20+ B-cell non-Hodgkin's lymphoma previously treated with two or more lines of therapy. Epcoritamab was dosed subcutaneously in 28-day cycles; once weekly during cycles 1-3, every 2 weeks during cycles 4-9, and every 4 weeks from cycle 10 until disease progression or unacceptable toxicity. Step-up dosing and cytokine release syndrome (CRS) prophylaxis were used during treatment cycle 1. As of January 31, 2022, 36 patients received treatment with 48 mg epcoritamab monotherapy. At a median follow-up of 8.4 months, overall response and complete response rates by independent review committee were 55.6% and 44.4%, respectively. The median duration of response, duration of complete response, and overall survival were not reached at the time of data cut-off. The most common treatment-emergent adverse events of any grade were CRS (83.3%), injection-site reactions (69.4%), infections (44.4%), neutropenia (38.9%), hypokalemia (27.8%), and decreased lymphocyte count (25.0%). Cytokine release syndrome occurrence was predictable; events were primarily low grade (grade 1-2), all resolved, and none led to treatment discontinuation. These encouraging results are consistent with previous findings and support the ongoing clinical evaluation of epcoritamab for the treatment of R/R DLBCL, including in earlier treatment lines.

Reversible suppression of T cell function in the bone marrow microenvironment of acute myeloid leukemia.

Acute myeloid leukemia (AML) is the most common acute leukemia in adults, with approximately four new cases per 100,000 persons per year. Standard treatment for AML consists of induction chemotherapy with remission achieved in 50 to 75% of cases. Unfortunately, most patients will relapse and die from their disease, as 5-y survival is roughly 29%. Therefore, other treatment options are urgently needed. In recent years, immune-based therapies have led to unprecedented rates of survival among patients with some advanced cancers. Suppression of T cell function in the tumor microenvironment is commonly observed and may play a role in AML. We found that there is a significant association between T cell infiltration in the bone marrow microenvironment of newly diagnosed patients with AML and increased overall survival. Functional studies aimed at establishing the degree of T cell suppression in patients with AML revealed impaired T cell function in many patients. In most cases, T cell proliferation could be restored by blocking the immune checkpoint molecules PD-1, CTLA-4, or TIM3. Our data demonstrate that AML establishes an immune suppressive environment in the bone marrow, in part through T cell checkpoint function.

CNV Radar: an improved method for somatic copy number alteration characterization in oncology.

BACKGROUND: Cancer associated copy number variation (CNV) events provide important information for identifying patient subgroups and suggesting treatment strategies. Technical and logistical issues, however, make it challenging to accurately detect abnormal copy number events in a cost-effective manner in clinical studies. RESULTS: Here we present CNV Radar, a software tool that utilizes next-generation sequencing read depth information and variant allele frequency patterns, to infer the true copy number status of genes and genomic regions from whole exome sequencing data. Evaluation of CNV Radar in a public multiple myeloma dataset demonstrated that CNV Radar was able to detect a variety of CNVs associated with risk of progression, and we observed > 70% concordance with fluorescence in situ hybridization (FISH) results. Compared to other CNV callers, CNV Radar showed high sensitivity and specificity. Similar results were observed when comparing CNV Radar calls to single nucleotide polymorphism array results from acute myeloid leukemia and prostate cancer datasets available on TCGA. More importantly, CNV Radar demonstrated its utility in the clinical trial setting: in POLLUX and CASTOR, two phase 3 studies in patients with relapsed or refractory multiple myeloma, we observed a high concordance rate with FISH for del17p, a risk defining CNV event (88% in POLLUX and 90% in CASTOR), therefore allowing for efficacy assessments in clinically relevant disease subgroups. Our case studies also showed that CNV Radar is capable of detecting abnormalities such as copy-neutral loss of heterozygosity that elude other approaches. CONCLUSIONS: We demonstrated that CNV Radar is more sensitive than other CNV detection methods, accurately detects clinically important cytogenetic events, and allows for further interrogation of novel disease biology. Overall, CNV Radar exhibited high concordance with standard methods such as FISH, and its success in the POLLUX and CASTOR clinical trials demonstrated its potential utility for informing clinical and therapeutic decisions.

Daratumumab plus lenalidomide and dexamethasone versus lenalidomide and dexamethasone in relapsed or refractory multiple myeloma: updated analysis of POLLUX.

In the POLLUX study, daratumumab plus lenalidomide/dexamethasone significantly reduced risk of progression/death versus lenalidomide/dexamethasone alone in relapsed/refractory multiple myeloma. We provide one additional year of follow up and include the effect on minimal residual disease and in clinically relevant subgroups. After 25.4 months of follow up, daratumumab plus lenalidomide/dexamethasone prolonged progression-free survival versus lenalidomide/dexamethasone alone (median not reached vs 17.5 months; hazard ratio, 0.41; 95% confidence interval, 0.31-0.53; P<0.0001). The overall response rate was 92.9% versus 76.4%, and 51.2% versus 21.0% achieved a complete response or better, respectively (both P<0.0001). At the 10-5 sensitivity threshold, 26.2% versus 6.4% were minimal residual disease-negative, respectively (P<0.0001). Post hoc analyses of clinically relevant patient subgroups demonstrated that progression-free survival was significantly prolonged for daratumumab plus lenalidomide/dexamethasone versus lenalidomide/dexamethasone regardless of number of prior lines of therapy. Patients previously treated with lenalidomide or thalidomide and those refractory to bortezomib received similar benefits (all P<0.01). Treatment benefit with daratumumab plus lenalidomide/dexamethasone was maintained in high-risk patients (median progression-free survival 22.6 vs 10.2 months; hazard ratio, 0.53; 95% confidence interval, 0.25-1.13; P=0.0921) and patients with treatment-free intervals of >12 and ≤12 months and >6 and ≤6 months. No new safety signals were observed. In relapsed/refractory multiple myeloma patients, daratumumab plus lenalidomide/dexamethasone continued to improve progression-free survival and deepen responses versus lenalidomide/dexamethasone. Trial Registration: clinicaltrials.gov identifier: 02076009.

Daratumumab plus bortezomib and dexamethasone versus bortezomib and dexamethasone in relapsed or refractory multiple myeloma: updated analysis of CASTOR.

Daratumumab, a CD38 human monoclonal antibody, demonstrated significant clinical activity in combination with bortezomib and dexamethasone versus bortezomib and dexamethasone alone in the primary analysis of CASTOR, a phase 3 study in relapsed and/or refractory multiple myeloma. A post hoc analysis based on treatment history and longer follow up is presented. After 19.4 (range: 0-27.7) months of median follow up, daratumumab plus bortezomib and dexamethasone prolonged progression-free survival (median: 16.7 versus 7.1 months; hazard ratio, 0.31; 95% confidence interval, 0.24-0.39; P<0.0001) and improved the overall response rate (83.8% versus 63.2%; P<0.0001) compared with bortezomib and dexamethasone alone. The progression-free survival benefit of daratumumab plus bortezomib and dexamethasone was most apparent in patients with 1 prior line of therapy (median: not reached versus 7.9 months; hazard ratio, 0.19; 95% con fidence interval, 0.12-0.29; P<0.0001). Daratumumab plus bortezomib and dexamethasone was also superior to bortezomib and dexamethasone alone in subgroups based on prior treatment exposure (bortezomib, thalidomide, or lenalidomide), lenalidomide-refractory status, time since last therapy (≤12, >12, ≤6, or >6 months), or cytogenetic risk. Minimal residual disease-negative rates were >2.5-fold higher with daratumumab across subgroups. The safety profile of daratumumab plus bortezomib and dexamethasone remained consistent with longer follow up. Daratumumab plus bortezomib and dexamethasone demonstrated significant clinical activity across clinically relevant subgroups and provided the greatest benefit to patients treated at first relapse. Trial registration: clinicaltrials.gov identifier: 02136134.

Improving accuracy of GPT-3/4 results on biomedical data using a retrieval-augmented language model.

Large language models (LLMs) have made a significant impact on the fields of general artificial intelligence. General purpose LLMs exhibit strong logic and reasoning skills and general world knowledge but can sometimes generate misleading results when prompted on specific subject areas. LLMs trained with domain-specific knowledge can reduce the generation of misleading information (i.e. hallucinations) and enhance the precision of LLMs in specialized contexts. Training new LLMs on specific corpora however can be resource intensive. Here we explored the use of a retrieval-augmented generation (RAG) model which we tested on literature specific to a biomedical research area. OpenAI's GPT-3.5, GPT-4, Microsoft's Prometheus, and a custom RAG model were used to answer 19 questions pertaining to diffuse large B-cell lymphoma (DLBCL) disease biology and treatment. Eight independent reviewers assessed LLM responses based on accuracy, relevance, and readability, rating responses on a 3-point scale for each category. These scores were then used to compare LLM performance. The performance of the LLMs varied across scoring categories. On accuracy and relevance, the RAG model outperformed other models with higher scores on average and the most top scores across questions. GPT-4 was more comparable to the RAG model on relevance versus accuracy. By the same measures, GPT-4 and GPT-3.5 had the highest scores for readability of answers when compared to the other LLMs. GPT-4 and 3.5 also had more answers with hallucinations than the other LLMs, due to non-existent references and inaccurate responses to clinical questions. Our findings suggest that an oncology research-focused RAG model may outperform general-purpose LLMs in accuracy and relevance when answering subject-related questions. This framework can be tailored to Q&A in other subject areas. Further research will help understand the impact of LLM architectures, RAG methodologies, and prompting techniques in answering questions across different subject areas.

Epcoritamab in relapsed/refractory large B-cell lymphoma: 2-year follow-up from the pivotal EPCORE NHL-1 trial.

Primary results (median follow-up, 10.7 months) from the pivotal EPCORE® NHL-1 study in relapsed or refractory (R/R) large B-cell lymphoma (LBCL) demonstrated deep, durable responses with epcoritamab, a CD3xCD20 bispecific antibody, when used as monotherapy. We report long-term efficacy and safety results in patients with LBCL (N = 157; 25.1-month median follow-up). As of April 21, 2023, overall response rate was 63.1% and complete response (CR) rate was 40.1%. Estimated 24-month progression-free survival (PFS) and overall survival (OS) rates were 27.8% and 44.6%, respectively. An estimated 64.2% of complete responders remained in CR at 24 months. Estimated 24-month PFS and OS rates among complete responders were 65.1% and 78.2%, respectively. Of 119 minimal residual disease (MRD)-evaluable patients, 45.4% had MRD negativity, which correlated with longer PFS and OS. CR rates were generally consistent across predefined subgroups: 36% prior chimeric antigen receptor (CAR) T-cell therapy, 32% primary refractory disease, and 37% International Prognostic Index ≥3. The most common treatment-emergent adverse events were cytokine release syndrome (51.0%), pyrexia (24.8%), fatigue (24.2%), and neutropenia (23.6%). These results underscore the long-term benefit of epcoritamab for treating R/R LBCL with deep responses across subgroups, including patients with hard-to-treat disease and expected poor prognosis (ClinicalTrials.gov Registration: NCT03625037).

Inherited determinants of early recurrent somatic mutations in prostate cancer.

Prostate cancer is a highly heritable molecularly and clinically heterogeneous disease. To discover germline events involved in prostate cancer predisposition, we develop a computational approach to nominate heritable facilitators of somatic genomic events in the context of the androgen receptor signaling. Here, we use a ranking score and benign prostate transcriptomes to identify a non-coding polymorphic regulatory element at 7p14.3 that associates with DNA repair and hormone-regulated transcript levels and with an early recurrent prostate cancer-specific somatic mutation in the Speckle-Type POZ protein (SPOP) gene. The locus shows allele-specific activity that is concomitantly modulated by androgen receptor and by CCAAT/enhancer-binding protein (C/EBP) beta (CEBPB). Deletion of this locus via CRISPR-Cas9 leads to deregulation of the genes predicted to interact with the 7p14.3 locus by Hi-C chromosome conformation capture data. This study suggests that a polymorphism at 7p14.3 may predispose to SPOP mutant prostate cancer subclass through a hormone-dependent DNA damage response.Prostate cancer is a heterogeneous disease, and many cases show somatic mutations of SPOP. Here, the authors show that a non-coding polymorphic regulatory element at 7p14.3 may predispose to SPOP mutant prostate cancer subclass through a hormone dependent DNA damage response.

A DNA Hypomethylation Signature Predicts Antitumor Activity of LSD1 Inhibitors in SCLC.

Epigenetic dysregulation has emerged as an important mechanism in cancer. Alterations in epigenetic machinery have become a major focus for targeted therapies. The current report describes the discovery and biological activity of a cyclopropylamine containing inhibitor of Lysine Demethylase 1 (LSD1), GSK2879552. This small molecule is a potent, selective, orally bioavailable, mechanism-based irreversible inactivator of LSD1. A proliferation screen of cell lines representing a number of tumor types indicated that small cell lung carcinoma (SCLC) is sensitive to LSD1 inhibition. The subset of SCLC lines and primary samples that undergo growth inhibition in response to GSK2879552 exhibit DNA hypomethylation of a signature set of probes, suggesting this may be used as a predictive biomarker of activity.