Hamidi, Habib

Abstract Background: Undetectable minimal residual disease (uMRD) at the end of first-line (1L) treatment (tx) with 12-cycles (C) of venetoclax (ven) + obinutuzumab (obin) is associated with prolonged survival for patients (pts) with chronic lymphocytic leukemia (CLL, Al Sawaf et al. Blood 2024). We hypothesized that MRD may be used to guide ven tx duration for CLL pts treated with 1L ven + obin. Methods In this ongoing phase II, multicenter investigator-initiated study, pts receive ven + obin for 1L tx of CLL/small lymphocytic lymphoma (NCT04447768). Eligible pts have a CIRS score of ≤ 6 and require 1L tx per iwCLL 2018 criteria. Pts receive 6 x 28-day C of obin and initiate ven on C1D22. Ven duration is dependent on peripheral blood (PB) MRD status measured by ClonoSeq next generation sequencing. At C7, pts undergo MRD testing. Pts uMRD6 (<10-6 sensitivity) at both C7 and C9 dc ven after 9C and enter treatment free observation (TFO). Pts with detectable MRD6 (dMRD6) at C7 undergo repeat testing at C12; if uMRD5 (<10-5) pts dc ven after 12C. Pts dMRD5 at C12 continue ven for an additional 12C (24C total) prior to TFO. The primary study endpoint is 36-month PFS; the unpromising rate is set at 82% and will be tested by a nonparametric survival estimate (Kaplan-Meier) along with Greenwood's formula. RNA sequencing (RNAseq) was performed pre-tx on patient bone marrow (or PB if inaspirable) samples. Unsupervised clustering using machine learning (ML) based on consensus nonnegative matrix factorization was applied to the most variable genes to identify patient clusters. Clusters were then biologically characterized by gene signature scoring using xCell algorithm and assessed for association with uMRD6 at C7 using a Fisher exact test. Results Baseline characteristics of the 100 pts who initiated tx include: median age 58 (range 34-81); 71% male, 90% White, 3% Black and 2% Asian, 5% unknown; 49% IGHV unmutated, 16% TP53 mutation, 8% del17p (previously reported, Roeker et al. ASH 2024). 3 pts stopped tx prior to C7. 51 pts were uMRD6 at C7 and C9 and completed 9C ven. 46 pts were dMRD6 at C7; 6 pts stopped tx outside of meeting protocol requirements and are included in intention-to-treat (IIT) analyses. 32 pts were uMRD5 at C12 and dc ven after 12C. 8 pts were dMRD5 at C12: 5 continued ven per protocol (3 completed 24C ven, 2 ongoing tx as of 15 May 2025 data cut-off), 2 stopped ven outside of protocol requirements (included in ITT analyses), and 1 pt withdrew consent. At a median follow-up of 35 months, the 36-month PFS rate is 90% (95% CI: 84%, 97%). 36-month OS is 94% (95% CI: 89%, 99%). Landmark PFS analyses from EOT were performed for pts who dc ven after 9C due to uMRD6 (24-month PFS 92%, 95% CI: 85, 100%, n=51) and pts who dc ven after 12C due to uMRD5 (24-month PFS 96%, 95% CI: 89%, 100%, n=32). For pts dc ven after 9C due to uMRD6 (n=51), PB MRD at 12C after EOT was: 59% uMRD6, 12% uMRD5, 4% uMRD4, 8% dMRD4 and 18% unknown. For pts dc ven after 12C due to uMRD5 status (n=32), PB MRD at 12C following EOT was: 9% uMRD6, 31% uMRD5, 9% uMRD4, 25% dMRD4 and 25% unknown. Of the 5 pts with PD, 3 pts have had Richter Transformation to diffuse large B-cell lymphoma (all uMRD6 at C7 / 9). Two pts with PD have not met iwCLL tx criteria and remain in TFO. Five pts have died (suicide, West Nile encephalitis, COVID-19, complications of glioblastoma and RT DLBCL). Five C12 dMRD5 pts who continued ven have reached C18, with only 1 of 5 converting to uMRD5 in PB. However, all 5 maintained uMRD4 (<10-4). Of 3 pts that completed 24C ven (2 ongoing tx), 1 pt had deepening of response to uMRD5 at C24, and 2 pts had rising MRD on ven (dMRD4 at C24). To investigate patient heterogeneity at baseline and its relationship to uMRD, we clustered patient RNAseq samples into two clusters using artificial intelligence/ML based unsupervised clustering. The clusters associated with uMRD6 at C7 (Fisher exact test p=0.038), and patients in the cluster with significantly higher immunosuppressive cells such as T-regulatory cells and M2 macrophages were less likely to be uMRD6 at C7. Conclusions At a median follow-up of 35 months, the estimated 36-month PFS rate is 90%. RNAseq transcriptomic analyses identified that patients with enriched immunosuppressive cells were less likely to be uMRD6 at C7. Additional follow-up is required to confirm the durability of this MRD-guided ven + obin tx strategy.

5510 Background: Transcriptional profiling of ovarian cancers has proven to be complex. As such it has been difficult to validate existing signatures across studies. Cancer Genome Atlas (TCGA) researchers have identified four molecular subtypes of high-grade serous ovarian cancer (HGSOC). However, survival duration did not differ significantly for the TCGA subtypes. Potential limitations of the TCGA data include short clinical follow-up (45% were alive at the time of last follow-up) and the need to unify gene expression measures from multiple platforms. Methods: Clinically annotated stage-II–IV HGSOC samples (n = 175) with >70% tumor cell content were profiled using the Agilent Whole Human Genome 4 x 44K chip. To identify subtypes non-negative matrix factorization (NMF) of mRNA expression was performed using ~2000 genes with the highest variability across patients. In parallel differentially expressed genes were identified using the Rosetta Similarity Search Tool as well as analysis of variance based on genes known to be involved in epithelial to mesenchymal transition. Results: Median follow-up time was 35 months (range, 0–202 months, 12% were alive at the time of last follow-up). NMF clustering confirmed four HGSOC subtypes (immunoreactive, differentiated, proliferative, and mesenchymal) on the basis of gene content in the clusters. Pathway signatures with therapeutic potential were identified for individual or multiple subtypes. Survival differed significantly between the four molecular subgroups in univariate (Hazard Ratio [HR] 2.4, 95% CI 1.5-4.1, p = 0.007) and multivariate (HR 2.3, 95% CI 1.3-4.0, p = 0.003) analyses when accounting for age, stage, grade, and postoperative residual tumor. Using the supervised clustering approach two distinct molecular subtypes of HGSOC based on epithelial and mesenchymal gene expression signatures were identified with significantly different survival outcomes. Conclusions: Here we independently validate and expand upon the molecular TCGA classification of HGSOC. The potential of these two prognostic classifiers may lie in their ability to recognize categories of patients that are more likely to respond to particular therapies.

Abstract Four gene expression subtypes of high-grade serous ovarian cancer (HGSC) have been previously described. In these early studies, a fraction of samples that did not fit well into the four subtype classifications were excluded. Therefore, we sought to systematically determine the concordance of transcriptomic HGSC subtypes across populations without removing any samples. We created a bioinformatics pipeline to independently cluster the five largest mRNA expression datasets using k-means and nonnegative matrix factorization (NMF). We summarized differential expression patterns to compare clusters across studies. While previous studies reported four subtypes, our cross-population comparison does not support four. Because these results contrast with previous reports, we attempted to reproduce analyses performed in those studies. Our results suggest that early results favoring four subtypes may have been driven by the inclusion of serous borderline tumors. In summary, our analysis suggests that either two or three, but not four, gene expression subtypes are most consistent across datasets.