The CCTG PA.7 study was a randomized phase II trial comparing chemotherapy with and without dual immune checkpoint inhibition in patients with metastatic pancreatic ductal adenocarcinoma (mPDAC). In follow-up to the published primary results of the trial, the analysis herein focused on long-term survival and exploratory analysis.Plasma sequencing analysis identified concurrent mutations in DNA damage repair genes BRCA1, POLE, ATM and FANCA in 18/173 (10.40%) of patients, and presence of two or more mutations in these genes was associated with overall survival benefit for patients receiving immunotherapy (median overall survival 26.2 vs. 9.7 months; hazard ratio, 0.34; 95% CI, 0.16-0.68; p = 0.001; interaction p = 0.003). Partial response was observed in 7/11 (63.64%) patients with concurrent DNA damage repair gene mutations in the immunotherapy arm. As a prospective study in PDAC identifying a potential biomarker beyond mismatch repair deficiency for benefit from immunotherapy, these data highlight an actionable subgroup of mPDAC.

PURPOSE:Pre-existing and emerging molecular targeted therapies have been identified for the treatment of pancreatic ductal adenocarcinoma (PDAC) and exemplify the need for incorporating tumor sequencing into routine clinical practice. Tumor profiling through circulating tumor DNA (ctDNA) represents an important opportunity in PDAC because of the aggressive nature of the disease. Circulating tumor fraction (CTF) levels are highly variable within and between cancer types and influence the accuracy of plasma-based tumor profiling, and the variability and factors related to CTF levels in PDAC are not well understood. EXPERIMENTAL DESIGN:ctDNA sequencing (PredicineATLAS) and clinical metadata from a cohort of 166 patients with metastatic PDAC (mPDAC) were generated as part of the PA.7 trial (NCT02879318). Patients were stratified into high CTF (>30% CTF; 35/166, 21.08%) and low CTF (131/166, 78.92%) groups for comparative analysis. Matched RNA sequencing data were available for 20 patients. RESULTS:High CTF was associated with lower overall survival (hazard ratio = 1.87; 95% confidence interval, 1.27-2.75; P = 0.0014) as well as clinical presentation that was indicative of higher disease burden, with CTF highest in patients that had liver metastases and distant metastases including bulky lymph nodes versus patients with no liver metastases (P < 0.001). Exploratory gene expression analysis revealed a positive association between CTF and upregulation of cell cycle-related pathways, which included those involving CDK4 (P = 0.0093) and MT2A (P = 0.012), as well as glycolytic (P = 0.0028) and basal-like (P = 0.029) subtyping genes. CONCLUSIONS:These data demonstrate the heterogeneity of CTF and its associated factors in mPDAC, which converge toward an aggressive phenotype from both clinical and molecular standpoints.

Abstract Background: Over two dozen histological types of salivary gland cancers are recognized comprising 3-5% of all head and neck cancers. Response to chemotherapy and actionable target expression in patients with metastatic disease vary by histological type e.g. HER2 and/or androgen receptor are often overexpressed by salivary duct carcinomas and NTRK fusions often present in secretory carcinomas. Overall results of this phase II basket trial have been previously reported (Gupta 2025). Herein we report additional data from the metastatic salivary gland cohort by histological subtype. Methods: IND.228 was a prospective multi-center, non-blinded, open-label phase II basket trial that enrolled 8 cohorts of patients with rare cancers including a salivary gland cohort. Eligible patients had incurable disease with no known life-prolonging treatment options or contraindications to immunotherapy and ECOG performance status 0 or 1. Tumor PD-L1 expression was not required and patients with adenoid cystic carcinoma were excluded. Patients received durvalumab 1500 mg IV plus tremelimumab 75 mg IV q4 weeks for 4 cycles followed by durvalumab q4 weeks until disease progression. The primary outcome was objective response by RECIST 1.1 criteria. Results: 21 patients with incurable salivary gland cancer were enrolled. Of 20 patients evaluable for response there were 6 females and 14 males, median age 61 years (range, 40-72) with a median of 3 metastatic sites (range, 1-5). Histological types were salivary duct carcinoma (8), acinic cell carcinoma (5), adenocarcinoma (3), clear cell carcinoma (2), and others (2). All had diagnosis confirmed by central pathology review. 190 cycles of treatment were given with a median of 7 cycles per patient (range, 1-40). Four patients had partial responses, 8 had stable disease, and 8 had progressive disease as best response. The proportions of response and response plus stable disease by histological type, respectively, were salivary duct carcinoma (2/8, 4/8), acinic cell carcinoma (1/5, 4/5), and adenocarcinoma (1/3, 3/3). 10/13 patients with avaialable tumor grade had high-grade cancers. Progression-free survival ranged from 7.3-59.3+ months in responding patients, and from 3.8-11.4 months in stable disease patients. One patient with salivary duct carcinoma received 40 cycles of treatment and had a response duration of over 5 years. Conclusions: Patients with salivary duct carcinoma, acinic cell carcinoma, and adenocarcinoma appeared to benefit from combined CTLA-4 and PD-L1 immune checkpoint blockade. Immune checkpoint blockade deserves further study and consideration as a treatment option in these histological types of incurable salivary gland cancer. Citation Format: Eric Winquist, John Hilton, Christian Kollmannsberger, Danielle Charpentier, Dorie-Anna Dueck, Hal Hirte, Sebastien Hotte, Rahima Jamal, Raymond Jang, Andrew Maksymiuk, Randeep Sangha, Stephanie Snow, Osama Souied, Jennifer Spratlin, Ralph Wong, Abha Gupta, Thierry Alcindor, Quincy Chu, Derek Jonker, Torsten Nielsen, Ming Tsao, Tricia Cottrell, Joana Sederias, Siwei Zhang, Wei Tu, Janet Dancey. Durvalumab and tremelimumab in patients with metastatic non-adenoid cystic salivary gland cancer treated in the CCTG IND.228 phase II basket trial [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Breaking Barriers in the Fight against Rare Cancers; 2026 Jul 18-20; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2026;86(14_Suppl):Abstract nr A029.

CHIP and adverse events. A, Percentage of patients on each treatment that had an adverse event on therapy, split by cohort and treatment. B, Proportions of patients with adverse events split by CHIP status. Lighter shades indicate no adverse event, and darker shades indicate an adverse event. P values were determined by Fisher exact tests.

BACKGROUND:Preclinical and observational studies suggest that exercise may improve cancer outcomes. However, definitive level 1 evidence is lacking. METHODS:In this phase 3, randomized trial conducted at 55 centers, we assigned patients with resected colon cancer who had completed adjuvant chemotherapy to participate in a structured exercise program (exercise group) or to receive health-education materials alone (health-education group) over a 3-year period. The primary end point was disease-free survival. RESULTS:From 2009 through 2024, a total of 889 patients underwent randomization to the exercise group (445 patients) or the health-education group (444 patients). At a median follow-up of 7.9 years, disease-free survival was significantly longer in the exercise group than in the health-education group (hazard ratio for disease recurrence, new primary cancer, or death, 0.72; 95% confidence interval [CI], 0.55 to 0.94; P = 0.02). The 5-year disease-free survival was 80.3% in the exercise group and 73.9% in the health-education group (difference, 6.4 percentage points; 95% CI, 0.6 to 12.2). Results support longer overall survival in the exercise group than in the health-education group (hazard ratio for death, 0.63; 95% CI, 0.43 to 0.94). The 8-year overall survival was 90.3% in the exercise group and 83.2% in the health-education group (difference, 7.1 percentage points; 95% CI, 1.8 to 12.3). Musculoskeletal adverse events occurred more often in the exercise group than in the health-education group (in 18.5% vs. 11.5% of patients). CONCLUSIONS:A 3-year structured exercise program initiated soon after adjuvant chemotherapy for colon cancer resulted in significantly longer disease-free survival and findings consistent with longer overall survival. (Funded by the Canadian Cancer Society and others; CHALLENGE ClinicalTrials.gov number, NCT00819208.).

PURPOSE Cetuximab (CET), targeting the epidermal growth factor receptor, is a systemic treatment option for patients with colorectal cancer. One known predictive factor for CET efficacy is the presence of CET-related rash; other putative toxicity factors include fatigue and nausea. Analysis of early CET-associated toxicities may reveal patient subpopulations that clinically benefit from long-term CET treatment. METHODS We analyzed data from CO.20 (ClinicalTrials.gov identifier: NCT00640471 ) trial arms, CET + brivanib alaninate (BRIV) (n = 376) and CET + placebo (n = 374), and CO.17 (ClinicalTrials.gov identifier: NCT00079066 ) trial arms, CET (+best supportive care [BSC]; n = 287) and BSC only (n = 285). Patients were clustered into subpopulations using KmL3D, a machine learning method, to analyze 14 joint longitudinal toxicity trajectories from weeks 0 to 8 of treatment. Landmark survival analyses were performed from 8 weeks after treatment initiation. Regression analyses assessed the relationship between subpopulations and plasma CET concentrations. Three supervised machine learning models were developed to assign patients in the CO.20-CET trial arm into subpopulations, which were then validated using CO.20-CET-BRIV and CO.17-CET trial arm data. RESULTS Joint longitudinal toxicity clustering revealed dichotomous high- and low-toxicity clusters, with all CET-containing arms showing consistent toxicity trajectories and characteristics. High-toxicity clusters were associated with male predilection, fewer metastatic sites, fewer colon-only primaries, and higher body mass indices. In CO.20 trial samples, higher toxicity clusters were associated with improved overall survival and progression-free survival outcomes (adjusted hazard ratios ranging from 2.21 to 4.36) and higher CET concentrations ( P = .003). The random forest predictive model performed the best, with an AUC of 0.981 (0.963-0.999). CONCLUSION We used an innovative machine learning approach to analyze longitudinal joint drug toxicities, demonstrating their role in predicting patient outcomes through a putative pharmacokinetic mechanism.

Prevalence of CHIP variants across cohorts and CHIP association with age. A, Percentage of patients with CHIP variants in each cohort. Insert shows the breakdown by tumor type in the PREDiCT-l cohort. B, Number of patients with a CHIP variant in each gene; DNMT3A, TET2, and ASXL1. C, Protein position of coding variants by gene. Each lollipop represents a patient, multiple dots show patients with the same variant position, and red outlines show variants of high impact (frameshift or stop gain). Orange bars along the length of the genes show functional domains. D, Percentage of patients with CHIP variants by age group. E, Age difference between patients with CHIP (CHIP+) and without CHIP (CHIP−). P value was determined by the Wilcoxon rank sum test.

VAF of all CHIP variants split by cohort

LBA248 Background: Effective treatment options remain an unmet need for patients (pts) with microsatellite stable/mismatch repair proficient (MSS/pMMR) metastatic colorectal cancer (mCRC). Combination therapy with the anti–lymphocyte activation gene (LAG)-3 antibody favezelimab (fave) and the PD-1 inhibitor pembrolizumab (pembro), has shown promising antitumor activity and manageable safety in PD-L1 CPS ≥1 MSS/pMMR mCRC. The phase 3 KEYFORM-007 study (NCT05064059) evaluated the efficacy and safety of co-formulated fave/pembro vs standard-of-care (SOC) in PD-L1–positive MSS/pMMR mCRC. We present results of the pre-specified final analysis of OS. Methods: Eligible pts with PD-L1 CPS ≥1, MSS/pMMR unresectable mCRC (Stage IV per AJCC 8 th edition), who had progressed on or after, or could not tolerate standard treatment were randomized 1:1 to co-formulated fave 800 mg/pembro 200 mg IV Q3W (Arm A) or SOC (regorafenib 160 mg PO Q4W [QD on days 1-21] or TAS-102 35 mg/m 2 PO Q4W [BID on days 1-5 and 8-12]) (Arm B). Randomization was stratified by geographic region, presence or absence of liver metastases, and time from initial diagnosis of metastatic disease to randomization. Treatment continued for up to 35 cycles or until unacceptable toxicity, progression, confirmed CR (Arm A), or withdrawal. The primary endpoint was OS. The data cut-off was August 15, 2024. Secondary endpoints included PFS, ORR, and DOR (central review, RECIST v1.1 [assessed at interim analysis with data cut-off of August 21, 2023]), and safety. Results: At final analysis, 441 pts (63% male; 59% RAS mutant) were randomized (221 fave/pembro; 220 SOC. Median follow-up was 28 mo (range, 21-32). Median OS was not superior with fave/pembro vs SOC (median 7.3 vs 8.5 mo; HR 0.98; 95% CI, 0.80-1.20; P = 0.4183) in pts with MSS/pMMR mCRC. PFS was not superior with fave/pembro vs SOC (median 2.1 vs 2.6 mo; HR 1.34; 95% CI, 1.09-1.64; nominal P = 0.997). Per protocol, PFS was not tested for statistical significance. A confirmed objective response occurred in 15 (14PR; 1CR [6.8%]) vs 2 (2PR; [0.9%]) pts in the fave/pembro and SOC arms, with best response of PD occurring in 143 (65%) vs 94 (43%) pts, respectively. Median DOR was not reached ([NR] range, 1.8 to 16.8+) among the 15 responders in the fave/pembro arm, and was 6.5+ mo and 12.4 mo, for the 2 responders in the SOC arm. At final analysis, treatment-related adverse events (TRAEs) occurred in 145 (66%) vs 167 (80%) pts, respectively (grade ≥3 in 44 [20%] vs 76 [36%] pts). Adverse events of special interest occurred in 84 (38%) vs 13 (6%) pts, respectively. Conclusion: At final analysis, co-formulated fave/pembro did not improve OS vs SOC in pts with PD-L1-positive MSS/pMMR mCRC. The safety profile was manageable with no new safety signals observed. Clinical trial information: NCT05064059 .

20 Background: In the NEO trial (CCTG CO.28, NCT03259035) patients with node negative (N0) T1-T3 rectal cancer were treated with neoadjuvant CAPOX/FOLFOX and transanal excision surgery (TES) with the goal of organ preservation. Patients without documented response following neoadjuvant chemotherapy were recommended total mesorectal excision (TME). We present a post-hoc assessment of ctDNA detection in this early-stage population and correlate kinetics with response and outcomes. Methods: Fifty-eight patients enrolled in CO.28. Whole blood was collected in EDTA tubes and processed for future ctDNA analysis at up to six timepoints: pre-chemotherapy, after neoadjuvant chemotherapy (and pre-TES), yearly during surveillance for 3 years and upon progression. A total of 195 samples were analyzed with the Guardant Reveal™ assay: a tissue-free epigenomic assay leveraging >20,000 epigenomic regions for ctDNA detection. Results: Of 48 available pre-treatment samples, ctDNA was detected in 22 (46%). Sensitivity by T-stage was 14% (1/7) for T1, 53% (17/32) for T2, 44% (4/9) for T3 cancers. ctDNA detection was lower following neoadjuvant chemotherapy (4/46, 9%, p<0.001 vs pre-chemotherapy). Of 41 patients with paired pre- and post-chemo samples, 49% (20/41) had undetectable ctDNA at both timepoints, 44% (18/41) had reduced tumour fraction following chemotherapy, the majority of which completely cleared ctDNA (94%; 17/18) and 7% (3/41) had an increased tumour fraction following chemotherapy (67%, 2/3 changing from negative to positive), all three of whom failed to respond to neoadjuvant chemotherapy and had TME recommended (100%). In contrast, of those that cleared ctDNA following chemotherapy, 35% had TME recommended (6/17, p=0.074). Of five recurrences (3 distant, 2 local), all had negative or reduced ctDNA tumour fraction during neoadjuvant therapy before surgical intervention. Among patients with local relapse who were managed with TES, 1/2 (50%) had detectable ctDNA at the time of local progression. No samples were available for those with distant recurrences (3/5) at time of relapse. Conclusions: The Guardant Reveal tissue-free assay was able to detect ctDNA even in this extremely early-stage cohort and identified cancers with inadequate response to therapy and in whom TME was recommended. A tissue-free approach may support timeliness of ctDNA results that would support ctDNA as an additional decision tool with endoscopic and MRI assessments to increase physician and patient comfort with organ preservation. Clinical trial information: 03259035.

CHIP and treatment outcomes. Kaplan–Meier curves showing the difference in outcome between patients with (CHIP+) and without CHIP (CHIP−) and outcomes on different therapies. PFS is shown on the left two columns for all three cohorts, and OS is on the right two columns for CO.26 (top row) and PA.7 (middle row). Hazard ratios displayed are derived from univariable Cox proportional hazards models, and interaction P values are from multivariable models (see “Materials and Methods”).

LBA3510 Background: Multiple observational studies have reported that post-diagnosis physical activity (PA) is associated with reduced recurrence rates in early-stage colon cancer but epidemiologic data is limited by confounding and reporting bias. CCTG CO.21 was designed to test the hypothesis that a meaningful increase in recreational PA after adjuvant therapy is achievable and will improve DFS in stage 3 or high-risk stage 2 colon cancer. Methods: CCTG CO.21 enrolled patients at 55 sites in 6 countries. Patients with resected stage 3 or high-risk stage 2 colon cancer who had received adjuvant chemotherapy were randomized to a structured exercise program (SEP) or health education materials (HEM). HEM participants received education materials promoting PA and healthy nutrition in addition to standard surveillance. SEP participants worked with a PA consultant who delivered an exercise intervention using behavior change methodology over 3 years. The SEP goal was to increase recreational PA by at least 10 MET-hours/week from baseline during the first 6 months and sustain this for 3 years. Participants chose the type, frequency, intensity and duration of aerobic exercise. The primary endpoint is DFS compared by a stratified log-rank test performed on an intention-to-treat basis. Secondary endpoints include overall survival (OS) and patient-reported outcomes (SF-36 physical function scale was primary PRO). Results: Between 2009 and 2024,889 participants were randomized to SEP (n=445) or HEM (n=444); 51% female, median age 61 years, 90% stage 3 disease. Compared to HEM, SEP resulted in statistically significant improvements in recreational PA, predicted VO2max, and 6-minute walk distance, all maintained over the 3-year intervention period. With a median follow-up of 7.9 years, 224 DFS events (93 in SEP and 131 in HEM) and 107 deaths (41 in SEP and 66 in HEM) were observed. 5-year DFS was 80% in SEP and 74% in HEM (HR 0.72; 95% CI 0.55-0.94; p=0.017). 8-year OS was 90% in SEP and 83% in HEM (HR=0.63; 95% CI=0.43-0.94; p=0.022). SF-36 physical function was substantially improved with SEP at 6 months (mean change scores 7.42 vs 1.10, p<0.001) and was sustained to 24 months. In the safety analysis, 19% (79/428) of patients on SEP reported any grade of musculoskeletal adverse event (MSK AE) over the course of the study, compared to 12% (50/433) on HEM. 10% (8/79) of MSK AE on SEP were considered to be related to participation in the PA program. Conclusions: Inpatients with stage 3 and high-risk stage 2 colon cancer, a 3-year structured exercise program initiated shortly after completion of adjuvant chemotherapy improves DFS, OS, patient-reported physical functioning, and health-related fitness. Health systems should incorporate structured exercise programs as standard of care for this patient population. Clinical trial information: NCT00819208 .

Background Dual inhibition of cytotoxic T-lymphocyte associated protein 4 (CTLA-4) and programmed death ligand 1 (PD-L 1) has been shown to be an effective treatment strategy in many cancers. We sought to determine the objective response rate of combination durvalumab (D) plus tremelimumab (TM) in parallel cohorts of patients with carefully selected rare cancer types in which these agents had not previously been evaluated in phase II trials and for which there was clinical or biological rationale for dual immune checkpoint inhibitor therapy to be active. Methods We designed a multi-centre, non-blinded, open-label phase II basket trial with each of the following 8 rare cancers considered a separate phase II trial: salivary carcinoma, carcinoma of unknown primary (CUP) with tumour infiltrating lymphocytes and/or expressing PD-L1, mucosal melanoma, acral melanoma, osteosarcoma, undifferentiated pleomorphic sarcoma, clear cell carcinoma of the ovary (CCCO) or squamous cell carcinoma of the anal canal (SCCA). The primary objective was to evaluate the response rate of the combination of D and TM, and the secondary objectives were to evaluate the tolerability and safety of D and TM combination. Eligible patients had advanced, metastatic or recurrent, or unresectable cancer with no known life-prolonging treatment option, age >= 16 years, ECOG performance status 0 or 1. Patients received D (1500 mg IV) + TM (75 mg IV) on Day 1 q4 weeks for 4 cycles followed by D q4 weeks until disease progression. This trial is registered with ClinicalTrials.gov, NCT02879162. Findings From December 14th, 2016, to August 14, 2019, 140 patients enrolled into seven cohorts. The rare melanoma cohorts were closed due to lack of accrual. Of the 140 patients enrolled, 138 were eligible, 138 were evaluable for toxicity and 128 (91%) were evaluable for response. Durable responses were noted in all cohorts except for osteosarcoma. The overall response rate for eligible patients was 16% (95% CI: 10-23%). The response rates in each cancer cohort were undifferentiated pleomorphic sarcoma 15% (n = 3/20; 95% CI 3-38%), salivary carcinoma 20% (n = 4/20; 95% CI: 6-44%), CUP 17% (n = 3/18; 95% CI 4-41%), SCCA 10% (n = 2/20; 95% CI 12-32%) and CCCO 21% (n = 8/39; 95% CI 9-37%). Grade 3/4 adverse events were rare, where 4 patients experienced grade 4 related events and39 patients experienced grade 3 events. Interpretation Durvalumab + tremelimumab treatment resulted in meaningful responses in salivary carcinoma and CCCO and deserves further exploration in front-line studies.

Progression free survival for patients on PREDiCT-l, split by tumor type. Therapy received is indicated in each title above each Kaplan Meier curve.

4011 Background: Modified FOLFIRINOX (mFFX) is increasingly used in the perioperative setting in r-PDAC and patients (pts) would benefit from a biomarker approach. GATA6 expression enriches for the classical RNA subtype, associated with improved OS in advanced PDAC. Low expression identifies the basal subtype which may predict mFFX resistance. NeoPancONE is a single arm Phase II multicentre study evaluating clinical outcomes and investigating GATA6 as a biomarker of response to perioperative mFFX in r-PDAC. Methods: Pts were enrolled following central radiology review (CRR) and underwent an EUS FNB for GATA6 in-situ hybridization (ISH). Six cycles of mFFX were planned pre and postoperatively. The primary endpoint was 1 yr event-free survival (EFS) according to GATA6 ISH (high vs low). Secondary endpoints include OS, RECIST response, SAEs, R0 resection rates and RNA subtyping by PurIST. Statistical assumptions used a ratio of 3:1 GATA6 high:low, with a 1 yr EFS of 65% for high and 34% for low (HR 2.5, 80% power, 2 sided alpha 0.05). KM method and log-rank test were used. Results: Between Sep-2020– Sep 2023, 146 pts were screened and 84 enrolled (58%) at 8 Canadian centres. CRR deemed 39 (27%) ineligible. Clinical data are summarized (Table). GATA6 ISH was analysed in 74 (88%); 62 (84%) were high, 16% low. At a median follow up of 24.5 mos, the med EFS and OS in the ITT were 16.1 mos (95 CI; 13-21) and 34.2 mos (95 CI; 28-NE). Med OS in the 73 pts who underwent surgery was 35.6 mos (95 CI 33-NE). The 1 yr EFS was 71% in GATA6 high vs 58% in GATA6 low p= 0.53. 1 yr OS was 87% in high vs 75% for low p= 0.29. The proportion progressing within 6 mos of enrollment in the GATA6 low group was significantly higher (42% vs 12% p=0.02). PuriST subtyping was reported in 49 (67%) resections; 14% basal, 86% classical. The 1 yr EFS was 79% in classical vs 43% in the basal subtype p=0.1. The 1 yr OS was 95% in classical vs 57% in basal p=0.034. Conclusions: This is one of the first trials in r-PDAC to identify potential biomarkers to predict perioperative mFFX response. GATA6 by ISH can be assessed on baseline tissue. GATA6 high is a prognostic biomarker, although NS, trends towards improved EFS and an encouraging OS. Disease progression within 6 months of enrollment occurs in nearly 50% of patients with low GATA6 expression. Neoadjuvant mFFX should not be the standard of care in these patients. Basal/Classical subtyping had stronger prognostic value than GATA6 and should be considered at baseline EUS FNB for future perioperative strategies in r-PDAC studies. Clinical trial information: NCT04472910 . Characteristic n=84 Age med. (range) yrs 64 (44, 83) Baseline EUS FNB tissue n (%) 83 (98) Pre-op completed 6 cycles n (%) 62 (74) Pre-op RECIST CR/PR/SD/PD/NE % 1/18/64/11/6 Surgery Completed Y/N n (%) / R0 / R1 n (%) 73 (87) / 11(13) / 62 (85) / 11 (15) Adjuvant Chemotherapy Y / N n (%) 63 (86) / 10 (14) mFFX associated SAE ≥G3 n (%) 13 (15) Pre-op mFFX related deaths 3 (4)

Abstract Clonal hematopoiesis of indeterminate potential (CHIP) is the clonal expansion of hematopoietic stem cells from somatic mutations. It is a common incidental finding in cell-free DNA (cfDNA). We investigated the incidence of CHIP in cfDNA from patients with solid tumors and explored its association with treatment outcomes and adverse events. We reviewed cfDNA results from a local prospective solid tumor cohort (PREDiCT-l) and two randomized trials: Canadian Cancer Trials Group CO.26 [durvalumab + tremelimumab (D + T) or best supportive care in metastatic colorectal cancer] and Canadian Cancer Trials Group PA.7 (gemcitabine and nab-paclitaxel ± D + T in metastatic pancreatic adenocarcinoma). CHIP+ was defined as any mutation in DNMT3A, TET2, or ASXL1 with a variant allele frequency ≥2%. Presumed germline variants (variant allele frequency >40%) were removed. The first line of treatment after cfDNA was reviewed for grade ≥3 and dose-limiting toxicities. The prevalence of CHIP in the 465 included patients was 10% to 30%, and it was more common as age increased (P = 0.003). DNMT3A was the gene most frequently mutated in all cohorts. Patients with CHIP in PA.7 treated with immunotherapy showed an improved progression-free survival versus CHIP− [HR = 0.55 (0.28–1.07); P = 0.079, P-interaction = 0.098 (multivariable)]. However, patients with CHIP treated with chemotherapy in PREDiCT-l showed a trend toward worse progression-free survival [HR = 1.82 (0.98–3.38); P = 0.059]. There was no difference in adverse event rates between CHIP ± groups for those treated with chemotherapy or immunotherapy. CHIP is common in patients with solid tumors. Although not appearing to affect rates of adverse events, CHIP may affect outcomes from immunotherapy or chemotherapy. Significance: Liquid biopsy is increasingly being used in oncology for tumor molecular characterization. CHIP is a common incidental finding in cfDNA, and its prevalence increases with age. This study builds on growing evidence of common CHIP variants in patients with solid tumors. The results suggest a possible clinical impact of CHIP on treatment outcomes from immunotherapy or chemotherapy. This may have implications for treatment selection for patients with solid tumors.

4178 Background: CCTG PA.7 (NCT02879318) was a randomized phase II trial comparing gemcitabine (G) and nab-paclitaxel (N) with and without dual immune checkpoint inhibition with durvalumab (D) and tremelimumab (T) as 1st-line therapy in pts with mPDAC. Matched plasma and tissue-based sequencing was performed for exploratory correlative biomarker analysis. Methods: Pts received G+N+D+T (n = 11 run-in, 119 randomized, 2:1 randomization) or G+N (n = 61). Long-term trial analysis was performed with a median follow-up time of 81.6 months. Correlative analysis was performed for pts with baseline ctDNA sequencing using a 600-gene PredicineATLAS panel (n = 173), with a subset having matched archival tissue available for whole-genome sequencing (WGS; n = 46). Cox-based elastic net regression models were used to identify and rank combinations of mutations by their ability to predict survival hazard. Results: Long-term follow up analysis demonstrated no significant difference in median overall survival (mOS) between pts randomized to G+N+D+T vs G+N (9.8 vs 8.8 months; hazard ratio (HR) = 0.88; p = 0.46). Median progression-free survival (mPFS) was also not significantly different between treatment arms (5.5 vs 5.4 months, respectively; HR = 0.95, p = 0.77). Landmark analysis demonstrated 4-year survivorship of 5.4% in pts treated with G+N+D+T arm compared to 1.6% with G+N (p = 0.07). Two or more ctDNA-based mutations (somatic and germline considered separately) in DNA damage repair (DDR) genes BRCA1 , POLE , ATM or FANCA was present in 18/173 pts (10.4%) and was associated with improved OS with G+N+D+T vs G+N (mOS 26.2 months vs. 7.1 months; HR = 0.22 [0.07-0.7]; p = 0.0041, p-interaction = 0.012) as well as PFS (mPFS 14.6 vs. 4.6; HR = 0.17 [0.05-0.6]; p = 0.0020, p-interaction = 0.0070). In pts treated with G+N+D+T, partial response (PR) was seen in 63.6% of pts with ≥2 DDR gene mutations compared to 26.9% in other pts (p = 0.033), and this effect was not observed with G+N (p = 0.18). The DDR gene biomarker was validated in 5/6 (83%) biomarker-positive samples using archival tissue WGS. Conclusions: The presence of ≥2 DDR gene mutations was strongly associated with benefit from the combination of chemotherapy with dual immune checkpoint inhibitor therapy, and pts with this signature had prolonged mOS of over 2 years. This represents the first prospective study in PDAC to define a predictive biomarker beyond mismatch repair deficiency for benefit from immune checkpoint therapy. Given the long-term survival noted in this subgroup, assessment of DDR gene mutations could be considered as part of routine standard of care testing for mPDAC pts. Clinical trial information: NCT02879318 .

All CHIP variants identified in the study