Browse Articles
Discover research articles across all indexed journals
Financial toxicity identified by COST–FACIT screening in a community oncology population: A retrospective study.
e23103 Background: Financial toxicity (FT) is common among patients receiving cancer treatment and is associated with adverse clinical and quality of life outcomes. Patients undergoing active cancer therapy at a community hospital who screened positive for FT and met income-based criteria were eligible for financial assistance through an institutional support program. We conducted a retrospective analysis to evaluate demographic, socioeconomic, and clinical factors associated with FT. Methods: We performed a retrospective chart review of 426 adult patients with stage I-IV cancer receiving IV, injectable, or oral systemic therapies dispensed through a specialty pharmacy who completed the COST–FACIT questionnaire. Data collected: demographics, marital and employment status, insurance coverage, tumor characteristics and treatment. Data were de-identified and managed using REDCap. Analyses are descriptive. Results: Median age at first FACIT completion: 68 yrs (range, 30–100). Gender identity: cisgender female in 304 (71.4%), cisgender male in 121 (28.4%), and other in 1 (0.2%). Race and ethnicity: self-reported as white (77.6%), Hispanic or Latino (8.7%), Black or African American (5.6%), Asian (4.9%), and other (3.1%). 59.6% patients were married, with 40.6% employed and 39.4% retired. Insurance coverage included Medicare (n = 163), private/commercial insurance (n = 162), Medicare Advantage (n = 89), uninsured (n = 11), and other (n = 1). Tumor types: breast/gyn (50.7%), gastrointestinal (GI) (17.6%), hematologic (11.7%), thoracic (9.2%), genitourinary (GU) (4.0%) and other (6.8%). At time of first FACIT assessment, patients were receiving IV (56.8%), oral (21.3%), injectable (2.3%), or combination systemic therapy involving more than one treatment modality (19.5%). Mean FACIT score: 22.95 (SD 11.22). Individuals with FACIT scores < 26 were considered to meet criteria for clinically significant FT. 236 (55%) individuals met criteria for FT and were referred to a financial advocate. Tumor types in FT cohort: breast/gyn (46.2%), hematologic (16.9%), GI (14.8%), thoracic (10.2%), GU (3.8%) and other (8%). Treatment modalities in FT cohort: IV (51.6%), oral (21.2%), combination (24.1%), and injectable therapy (3%). Differences in tumor type and treatment modality distributions were observed between the overall and FT cohorts. Conclusions: Clinically significant FT was highly prevalent among patients receiving active cancer treatment, even within a high socioeconomic status region. Differences observed between the overall and FT cohorts underscore the heterogeneity in financial burden across tumor types and treatment modalities. These descriptive findings support routine FT screening and early referral to financial advocacy services. Additional analyses are ongoing and will further evaluate demographic, socioeconomic, and clinical contributors to FT.
Can we be one step ahead?: A systematic review and meta-analysis on the clinical prowess of artificial intelligence (AI) in predicting the malignant potential of pancreatic cysts.
e16008 Background: Population imaging studies show that pancreatic cystic lesions (PCLs)occur in about 2.5% of asymptomatic adults, with prevalence increasing sharply with age to nearly 10% in people over 70. Although most pancreatic cystic lesions are benign, some, such as Intraductal papillary mucinous neoplasms, carry malignant potential, making accurate diagnosis essential to guide appropriate surveillance or surgical management. Predicting malignant transformation is difficult, and the growing number of patients undergoing imaging surveillance places a significant burden on healthcare systems. Recent advances in machine learning and artificial intelligence that integrate multivariate data have demonstrated considerable potential to improve diagnostic precision beyond current clinical guidelines. Methods: The review followed PRISMA guidelines, with extensive searches of PubMed, Google Scholar, and ScienceDirect, followed by systematic qualitative and quantitative analysis of the included studies. Studies evaluating the clinical performance of Artificial Intelligence (AI) in predicting the malignant potential of pancreatic cysts using metrics such as accuracy, Area under the curve (AUC), sensitivity, and specificity were included in the final analysis. The data were analysed using the Meta, Metadata, and Metafor packages in RStudio.The study evaluated pooled AUC, sensitivity, specificity, and accuracy, along with AI model subgroups, to estimate overall clinical performance. Both the common and random effects models were considered within the linear (mixed-effects) model framework for specific statistical analyses. The Higgins’ I^2 statistic was used to assess heterogeneity across studies. Results: Eight studies met the inclusion criteria, comprising a total of 1,070 patients. The pooled AUC for the AI models was 0.93 [0.85; 1.00, 95% CI, p = 0.8832]. The overall pooled accuracy was 0.90 [0.80; 0.95, 95% CI, p < 0.0001]. Among the models, the convolutional neural network (CNN) achieved the highest accuracy at 0.94, whereas the Supervised 862 model demonstrated the lowest accuracy at 0.69. The pooled sensitivity and specificity were 0.92 [0.79; 0.98, 95% CI, p < 0.0001] and 0.93 [0.58; 0.99, 95% CI, p < 0.0001], respectively. Conclusions: The AI models evaluated in this analysis demonstrate strong predictive performance for assessing the malignant potential of pancreatic cysts, as indicated by high performance metrics. Further research is warranted to support broader clinical adoption. Pooled AI diagnostic performance. Outcome/Metric Pooled estimate 95% CI p-value Pooled AUC 0.93 0.85–1.00 0.8832 Overall accuracy (all models pooled) 0.9 0.80–0.95 <0.0001 Pooled sensitivity 0.92 0.79–0.98 <0.0001 Pooled specificity 0.93 0.58–0.99 <0.0001
Direct oral anticoagulants versus low-molecular-weight heparin in patients with liver metastases and cancer-associated thrombosis: A multi-center real-world analysis.
11179 Background: Patients with liver metastases are at high risk for cancer-associated thrombosis (CAT) and require therapeutic anticoagulation. Although direct oral anticoagulants (DOACs) are increasingly used, their safety relative to low-molecular-weight heparin (LMWH) in this high-risk population, particularly regarding gastrointestinal (GI) bleeding, remains uncertain. Methods: We conducted a retrospective, federated cohort study using the TriNetX Global Collaborative Network, including adult patients with liver metastases and CAT who initiated a DOAC or LMWH within 10 days of CAT diagnosis. The index date was defined as the first qualifying anticoagulant exposure. Propensity score matching was performed to balance measured demographics, comorbidities, and concomitant medications, yielding 19,380 patients per cohort. Outcomes were assessed from 1 to 365 days after the index date and included GI bleeding, thrombotic recurrence by DVT/PE or stroke, intensive care unit (ICU) admission, and all-cause mortality. GI bleeding was identified using diagnosis-based coding algorithms, with exclusion of patients who had GI bleeding prior to anticoagulation initiation. Risk estimates were calculated and time-to-event analyses. Results: GI bleeding occurred more frequently among patients treated with DOACs compared with LMWH (9.7% vs 8.3%; hazard ratio [HR] 1.11, 95% CI 1.02–1.19). Thrombotic recurrence was common in both groups (61.4% with DOACs vs 60.0% with LMWH); however, time-to-event analysis demonstrated a modestly lower hazard of recurrence among DOAC-treated patients (HR 0.94, 95% CI 0.92-0.97). ICU admission was infrequent but occurred more often in the DOAC cohort (0.6% vs 0.3%; HR 1.80, 95% CI 1.33–2.46). DOAC use was associated with significantly lower all-cause mortality compared with LMWH (50.8% vs 55.7%; HR 0.84, 95% CI 0.82–0.86). Conclusions: In patients with liver metastases and CAT, DOAC therapy was associated with a modestly increased risk of GI bleeding but improved overall survival compared with LMWH. Given the small absolute differences in bleeding risk and the potential for residual confounding from unmeasured factors such as cancer burden, performance status, life expectancy, and hospice referral, these findings may support clinical equipoise between anticoagulation strategies. Individualized treatment decisions remain essential in this vulnerable population. Clinical outcomes in propensity-score matched patients with liver metastasis and cancer associated thrombosis. Outcome DOAC (n=19,380) LMWH (n=19,380) HR CI GI bleeding 9.7% 8.3% 1.11 1.02-1.19 Thromboembolic recurrence 61.4% 60.0% 0.94 0.92-0.97 ICU admission 0.6% 0.3% 1.80 1.33-2.46 All-cause mortality 50.8% 55.7% 0.84 0.82-0.86
Liposomal irinotecan plus sintilimab and lenvatinib for second-line advanced HER2-negative gastric/gastro-esophageal junction adenocarcinoma: A phase II trial.
e16035 Background: Effective second-line treatment options are limited for advanced gastric/gastro-esophageal junction (G/GEJ) adenocarcinoma after failure of first-line therapy, especially in immunotherapy-pretreated patients. Immunotherapy combined with anti-angiogenic agents exhibits synergistic antitumor activity, while liposomal irinotecan offers improved efficacy and safety versus conventional irinotecan. This phase II study evaluated the efficacy and safety of a novel combination of liposomal irinotecan, an anti-PD-1 antibody (Sintilimab), and a multi-targeted tyrosine kinase inhibitor (Lenvatinib) in this setting. Methods: This was a prospective, single-arm, open-label, single-center phase II trial. Patients with histologically confirmed advanced HER2-negative G/GEJ adenocarcinoma who progressed on or after first-line therapy were enrolled. The regimen consisted of liposomal irinotecan (70 mg/m² IV, day 1), Sintilimab (200 mg IV, day 1), and Lenvatinib (8 mg or 12 mg orally once daily based on body weight < 60 kg or ≥60 kg) in 3-week cycles for up to 6 cycles. Maintenance therapy with sintilimab plus lenvatinib was optional for responders or stable disease. The primary endpoint was progression-free survival (PFS). Secondary endpoints included objective response rate (ORR), disease control rate (DCR), overall survival (OS), and safety. Results: Between July 2024 and January 2026, 25 patients were enrolled. The median age was 65 years, and 84% were male. Notably, 56.0% of patients received first-line immunotherapy combined with chemotherapy previously, and 64.0% had positive PD-L1 expression. As of the data cutoff (January 8, 2026), 22 patients (88%) remained on treatment. Two patients experienced disease progression (PFS: 6.07 and 6.87 months), and one patient was lost to follow-up. Mature PFS and OS data are pending longer-term follow-up. Among 20 patients with at least one post-baseline tumor assessment (RECIST 1.1), the ORR was 55% (95% CI, 34.2%-74.2%) and the DCR was 100% (95% CI, 83.9%-100.0%). Tumor shrinkage was observed in all evaluable patients. Treatment-related adverse events (TRAEs) of any grade occurred in 88% (22/25) of patients. The most common any-grade TRAEs were diarrhea (28%), fatigue (12%), and vomiting (16%). Grade ≥3 TRAEs were reported in 12% of patients, including diarrhea (4%), leukopenia (4%), thrombocytopenia (4%), and intestinal obstruction (4%). No treatment-related deaths occurred. Conclusions: The combination of liposomal irinotecan, Sintilimab, and Lenvatinib demonstrated promising antitumor activity with a manageable safety profile as a second-line therapy for advanced HER2-negative G/GEJ adenocarcinoma. These encouraging results warrant further validation in larger randomized controlled trials. Clinical trial information: NCT06499610 .
Performance of a tissue-free, molecular residual disease assay in colorectal, breast, and lung cancers from circulating tumor DNA.
3042 Background: The Latitude test is a methylation-based, tissue-free molecular residual disease (MRD) test. In this study, we describe the improved performance of the assay in colorectal cancer (CRC) and its performance in an expanded set of indications including breast and lung cancer. Methods: The methylation-based, tissue-free ctDNA assay uses next-generation sequencing to query differentially methylated genomic regions in patients with CRC, breast, and lung cancer compared to cancer-free individuals in order to classify plasma samples as MRD positive (MRD+) or MRD negative (MRD-). Performance was evaluated in patients with cancer: CRC (N = 214), breast (N = 153), and lung (N = 148) with known ctDNA-positive results from Signatera (tumor-informed ctDNA testing) across a range of variant allele frequency (VAF) ranges, and cancer-free individuals (training set N = 531, validation set N = 200). Performance metrics, such as percent positive agreement (PPA) and specificity, were evaluated using cross validation in the cancer and cancer-free cohorts, respectively. Results: Among 214 CRC patients (VAF range: 0.003%-4.5%), PPA was 95.94% and specificity among all 200 cancer-free individuals was 99.50%. Among 153 breast cancer patients (VAF range: 0.002%-7.8%), PPA was 91.81%, which was maintained across subtypes, including HR+ (94.75%, N = 40), HER2+ (94.45%, N = 20), and triple-negative (87.94%, N = 29). Among the 164 cancer-free females, specificity of the breast tissue-free assay was 98.18%. Among 148 lung cancer patients (VAF range: 0.003%-5.8%), PPA was 92.63%, maintained across subtypes, including non-small cell lung cancer (92.54%, N = 112) and small cell lung cancer (95.23%, N = 25). Among all 200 cancer-free individuals, specificity of the lung tissue-free assay was 99.50%. Conclusions: These data demonstrate a high concordance between the Latitude tissue-free MRD assay and a clinically validated tumor-informed ctDNA assay across CRC, breast, and lung cancers, supporting the use of methylation-based biomarkers for informing prognosis and patient management.
Frailty-comorbidity discordance and non-relapse mortality after allogeneic hematopoietic cell transplantation.
6552 Background: Non-relapse mortality (NRM) is a barrier to successful allogeneic hematopoietic cell transplantation (HCT). The HCT-Comorbidity Index (HCT-CI) is used for pre-HCT risk assessment but does not capture physiological reserve. Frailty, assessed by the validated Hematopoietic Cell Transplantation-Frailty Scale (HCT-FS), independently predicts post-HCT outcomes; however, the clinical consequences of frailty-comorbidity discordance remain poorly defined. We hypothesized that such discordance identifies patients whose transplant risk may be misclassified by HCT-CI alone. Methods: We retrospectively analyzed 692 patients undergoing first allogeneic HCT at our center. Frailty was assessed using HCT-FS and categorized as fit, pre-frail, frail. For discordance analyses, HCT-FS was dichotomized as fit vs pre-frail/frail. HCT-CI was dichotomized as low (<3) vs high (≥3). Patients were classified into 4 groups: fit+low HCT-CI (group I), fit+high HCT-CI (group II), pre-frail/frail+low HCT-CI (group III), and pre-frail/frail+high HCT-CI (group IV). Fine-Gray competing-risk models and Cox regression was used for NRM and overall survival (OS), respectively. Results: The median age was 61 years (Q1-Q3: 52-67), and 399 (58%) were male. At baseline, 287 (42%) patients had an HCT-CI ≥3. HCT-FS classified 252 (36%) as fit, 373 (54%) as pre-frail, and 67 (10%) as frail. Overall, 355 (51%) patients demonstrated discordant HCT-FS and HCT-CI scores: 78 (11%) as group II, and 277 (40%) classified as group III. At 2 years, NRM and OS were 11.2% (95% CI, 7-16) and 79.3% (95% CI, 73-84) in fit patients, 17.4% (95% CI, 14-22) and 67.0% (95% CI, 62-72) in pre-frail patients, and 33.5% (95% CI, 22-45) and 54.9% (95% CI, 42-66) in frail patients, respectively (p<0.01 for both). In multivariable analyses, HCT-FS, HCT-CI, and age at HCT were independently associated with NRM and OS. Compared with fit patients with low HCT-CI (group I), pre-frail/frail patients with HCT-CI ≥3 (group IV) experienced the highest risk of NRM (sub-distribution hazard ratio (sHR) 2.7, 95% CI 1.5-5, p<0.01). Pre-frail/frail patients with HCT-CI <3 (group III) also had an increased risk of NRM (sHR 1.9, 95% CI 1.1-3.5, p=0.01). Fit patients with HCT-CI ≥3 (group II) did not have a significant increase in NRM risk (sHR 1.7, 95% CI 0.8-3.7, p=0.15). The corresponding HR for OS were 2.1 (95% CI 1-3; p<0.01) for group IV, 1.7 (95% CI 1-3; p<0.01) for group III, and 1.4 (95% CI 0.8-2.5; p=0.18) for group II, compared with group I. Conclusions: Frailty-comorbidity discordance was observed in 51% of patients and has prognostic implications. Pre-frail and frail patients with low HCT-CI experience higher NRM than would be predicted by HCT-CI alone, suggesting underestimation of transplant risk when frailty is not assessed. Integrating HCT-FS with HCT-CI may improve risk stratification and better identify patients at high risk for NRM.
Sociodemographic and clinical characteristics of women < 50 years and impact on adjuvant chemotherapy decisions across OncotypeDX RS or unknown OncotypeDX categories. A National Cancer Database (NCDB) analysis.
e12547 Background: OncotypeDX recurrence score (RS) provides prognostic and predictive information to guide adjuvant treatment decisions in non-metastatic HR+/HER2-negative breast cancer (BC). However, premenopausal patients (pts) have not been well characterized in the existing literature, particularly with respect to sociodemographic factors. Considering these gaps, we evaluated the impact of sociodemographic and clinical characteristics on chemotherapy (CT) recommendations across RS categories or unknown RS. Methods: This retrospective analysis included NCDB data from 2018-2022 for pts < 50 years old (surrogate for premenoapusal status) with clinically node negative stage I-III HR+/HER2-negative BC who underwent upfront surgery. OncotypeDX RS was coded as ≤15, 16-20, 21-25, ≥26 or unknown. Multivariable binary logistic regressions assessed the associations between sociodemographic as well as clinical characteristics and receipt of adjuvant CT stratified by each RS category. Results: 19,438 pts were included: 7340 had unknown RS, 6040 with low RS (≤15), intermediate RS, including 2908 (RS 16-20) and 1496 (RS 21-25); and 1654 with high RS (≥26). Patients with high (vs low) RS were younger (median age 43 vs 45 years). Racial differences were evident, with more Black pts in the high vs low RS group (15.5% vs 8.2%). Patients with high (vs low) RS had poorly differentiated tumors (42.8% vs 3.5%), and PR negative and low (30% vs 1.9%). Fewer pts had private insurance in the high vs low RS (83.9% vs 84.5%). More patients received CT in the high (vs low) RS group (88.8% vs 4.4%). CT receipt was 16.4% in RS 16-20, and 54.0% in pts with RS 21-25. In the unknown RS, 40.9% of pts received CT. Among the unknown RS cohort, characteristics associated with lower odds of CT receipt were: older age, ductal (vs lobular) histology, moderate/high PR expression (vs none/low), and pts with sentinel LN biopsy (vs axillary LN dissection). In contrast, the following characteristics had higher odds of CT receipt: low/moderate ER (vs high), pT3/T4 (vs pT1/T2) and pN1+ (vs pN0), and moderately or poorly (vs well) differentiated tumors. In pts with low and intermediate (0-25) RS, pts had higher odds of CT receipt if pT3/T4, pN1+ and poorly/moderately differentiated tumors. Conclusions: Over a third of women < 50 years had an unknown OncotypeDX RS. Significant sociodemographic and clinical differences were observed across RS categories, most notably between high and low RS. Patients with a higher RS are associated with distinct sociodemographic characteristics: younger age, Black race and lower proportion had private insurance. Overall, in addition to age, tumor characteristics (size, nodal status, tumor grade) appear to have the strongest association with CT receipt, regardless of RS.
Patient-level clinical response dynamics in advanced melanoma with anzutresgene autoleucel (anzu-cel), a PRAME-directed T-cell receptor (TCR) T-cell therapy.
9508 Background: Pts with advanced melanoma frequently experience progressive disease (PD) with limited long-term survival following available therapies; there is a need for novel systemic treatments that can induce deep and durable responses. PRAME is a cancer-associated antigen expressed in >50 cancers, including melanoma. Anzu-cel (IMA203) is an autologous PRAME-directed TCR T-cell therapy engineered to recognize intracellular PRAME-derived peptides presented by HLA-A*02:01 and mediates a potent and specific anti-tumor immune response (Wermke et al. ASCO, 2025). A registration-directed phase 3 trial (SUPRAME; NCT06743126) is underway. Methods: IMA203-101 is an ongoing, multicenter, phase 1/2 trial evaluating anzu-cel in pts with advanced PRAME-positive solid tumors. Pts underwent lymphodepleting (LD) chemotherapy followed by a single infusion of anzu-cel +/- low-dose SUBQ IL-2. This analysis focuses on melanoma pts in phase 1b treated at RP2D. Data cutoff was Sep 24, 2025. Exploratory analyses evaluated lesion-level patterns of progression among pts evaluable per RECIST1.1. Results: As of data cutoff, 33 pts with melanoma were evaluable, incl. cutaneous (n=14), uveal (n=16), and other subtypes (n=3). Most frequent AEs were LD-related cytopenias. AEs of special interest included cytokine release syndrome (CRS), immune effector cell–associated neurotoxicity syndrome (ICANS) and hemophagocytic lymphohistiocytosis (HLH). CRS was predominantly low grade (G1: 36%, G2: 46%, G3: 18%) and occurred early after infusion, with incidence and severity decreasing over time. ICANS (G1: 6%, G3: 6%) and HLH (G2: 3%, G3: 3%) were infrequent. Confirmed ORR was 56% (cCR=1, cPR=17) and DCR was 91%. mDOR was 14.6 mos (4.2-38.2+), mPFS was 6.1 mos (1.4-39.6+), and mOS was 16.2 mos (2.4-39.6+). Responses were observed in target and nontarget lesions, with most pts demonstrating initial lesion responses within the first 6 weeks of treatment; median time to BOR of cPR/cCR was 1.4 mo (1.2-2.8). 25 pts were evaluable for patterns of PD. At the time of PD, 68% (17/25) developed new lesions (NL); 1 pt developed new central nervous system (CNS) lesions. PD via growth of existing lesions was observed in 56% (14/25); 24% (6/25) experienced PD by both growth of existing lesions and NLs. No specific pattern of PD was identifiable in those who achieved cPR compared to those who did not. Conclusions: Anzu-cel demonstrated favorable tolerability and induced clinically meaningful and durable responses. Exploratory analyses suggest that the CNS was not a common site for relapse. Future analyses will characterize response dynamics and progression patterns to better define scenarios in which progression following cPR remains clinically manageable, potentially supporting individualized decision-making regarding timing and necessity of subsequent systemic therapy. Clinical trial information: NCT06743126 .
Regional trends in pancreatic cancer incidence, mortality, and disability-adjusted life years in Pakistan, 1990-2023.
e16414 Background: Pancreatic cancer is a highly fatal malignancy with limited therapeutic options due to late-stage diagnosis. In Pakistan, rising metabolic risk factors and disparities in healthcare access may be contributing to an increasing burden, yet long-term provincial and sex-specific trends in incidence, mortality, and disability-adjusted life years (DALYs) remain poorly described. We assessed temporal and geographic patterns to inform targeted pancreatic cancer control strategies. Methods: Using Global Burden of Disease (GBD) 2023 estimates, we analyzed pancreatic cancer incidence, mortality, and DALYs in Pakistan from 1990-2023. Data were stratified by province and sex and expressed as age-standardized rates per 100,000 population. Joinpoint regression was used to estimate annual percent change (APC) and average annual percent change (AAPC) with 95% confidence intervals (CIs). Results: From 1990 to 2023, pancreatic cancer incidence increased across Pakistan, with marked provincial variation. The largest long-term increases occurred in Gilgit-Baltistan (AAPC: 2.07%; 95% CI: 1.75–2.39), Islamabad Capital Territory (AAPC: 1.70%; 95% CI: 1.46–1.94), and Azad Jammu & Kashmir (AAPC: 1.66%; 95% CI: 1.33–2.00), followed by Punjab (AAPC: 1.51%; 95% CI: 1.28–1.74) and Sindh (AAPC: 1.30%; 95% CI: 0.96–1.64). Increases were driven by recent accelerations, particularly during 2018–2023 in Punjab (APC: 3.23%; 95% CI: 2.12–4.35) and 2017–2023 in Sindh (APC: 3.08%; 95% CI: 2.11–4.06). Islamabad Capital Territory showed sustained long-term growth with rapid increases during 2020–2023. Khyber Pakhtunkhwa experienced modest growth following declines during 1995–2000, while Balochistan showed lower long-term increases with sharp surges during 2020–2023. Mortality and DALYs increased nationally, with the steepest long-term rises observed in Gilgit-Baltistan (mortality AAPC: 1.98%; 95% CI: 1.68–2.29; DALYs AAPC: 2.19%; 95% CI: 1.84–2.53) and Islamabad Capital Territory (mortality AAPC: 1.61%; 95% CI: 1.38–1.85; DALYs AAPC: 1.67%; 95% CI: 1.45–1.90). Punjab also experienced sustained increases, with pronounced late accelerations during 2018–2023 (mortality APC: 3.18%; 95% CI: 2.12–4.24; DALYs APC: 3.21%; 95% CI: 2.00–4.44), while Sindh and Azad Jammu & Kashmir showed early declines during 1990–1996 followed by renewed growth after 2017. Across all provinces, males consistently experienced higher incidence, mortality, and DALY rates than females. Conclusions: Pancreatic cancer burden has increased substantially in Pakistan since 1990, with pronounced provincial and sex-based disparities. Males and residents of Islamabad Capital Territory, Punjab, and Sindh experienced the greatest and most sustained increases. Strengthening early detection, equitable access to care, and cancer surveillance is urgently needed, particularly in high-burden regions.
Clinical presentation, management strategies, and recurrence rates in vulvar intraepithelial neoplasia (VIN) globally: A systematic review.
e17649 Background: Vulvar intraepithelial neoplasia (VIN) is a premalignant condition that lacks FDA-approved therapies or consensus management guidelines. This systematic review synthesizes global evidence on the presentation and management of VIN, with particular attention to variation in reporting, treatment strategies, and outcomes across surgical and nonsurgical approaches. Methods: The authors systematically searched global databases for articles related to VIN clinical presentation and management in accordance with PRISMA guidelines. Included studies had to have at least one treatment intervention and report on at least one of the following outcomes: clinical presentation, treatment modalities, recurrence rates, or progression to vulvar carcinoma. Only peer-reviewed articles published in English between 2000-2024 were included. Results: Among 3,650 publications, 58 met inclusion criteria. Studies (66%) were primarily conducted in Europe, with none from Africa. Sixteen studies (27.6%) included patients with HIV. Across studies, mean/median ages were typically in 40s-50s, with studies of immunocompromised cohorts skewing younger (min: 29 years) and those of differentiated VIN (dVIN) skewing older (max: 71 years). Sixteen studies (27.6%) evaluated exclusively usual-type VIN (uVIN) cohorts, while 4 (6.9%) evaluated exclusively dVIN cohorts. Treatment arms with at least one outcome datapoint (response, recurrence, progression) were included in our analysis (n=44 surgical arms, n=31 nonsurgical arms). Among the included studies, 44 evaluated surgical interventions and 31 evaluated nonsurgical therapies. Imiquimod was the most frequently studied nonsurgical treatment (n=17). Among studies reporting response outcomes, complete response rates ranged from approximately 45-100% across surgical modalities and 6-94% across nonsurgical modalities. Recurrence following treatment for VIN was common, with reported rates ranging from approximately 6-70%, most commonly between 20% and 50% for surgical and nonsurgical treatments. Where reported, median time to first recurrence ranged from ~7 to 42 months. At least one progression to vulvar cancer was reported in 15/44 (34.1%) surgical treatment arms and 8/31 (25.8%) nonsurgical arms. Conclusions: Management of VIN is highly variable, with surgical approaches predominating despite heterogeneous outcome reporting. Inconsistent clinical characterization and limited reporting of progression to cancer limit comparative evaluation of treatment strategies. Additionally, populations disproportionately affected by vulvar cancer, including those with HIV and in resource-limited settings, remain underrepresented. Standardized reporting and broader inclusion are needed to advance evidence-based VIN management.
Precision or perplexity: Assessment of discordance among different PD-L1 assays in the same specimen in advanced non–small cell lung cancer in a community setting.
e14579 Background: PD-L1 expression guides immunotherapy selection in NSCLC, yet multiple assays are used interchangeably despite known analytic variability. Real-world concordance across PD-L1 assays in community oncology practice is poorly characterized. Methods: We retrospectively analyzed PDL-1 expression data among 40 advanced NSCLC patients diagnosed in 2024 at a community oncology practice. PD-L1 testing was performed on each specimen using 22C3 (Pembrolizumab), 28-8 (Nivolumab), and SP142 assays (Atezolizumab). Positivity patterns (1% or higher) and inter-assay concordance were evaluated. Results: Among 40 Metastatic NSCLC patients, 40 underwent PD-L1 testing. 35 patients had all three assays performed on the same specimen (n = 35) PD-L1 positivity rates differed substantially: Complete concordance across all three assays occurred in only 11 PDL-1 positive patients (31%), while 10 (29%) were PDL-1 negative across all assays. Clinically significant discordance was frequent: 12 patients (34%) were positive by 22C3 and 28-8 but negative by SP142 Additional discordance included isolated positivity by 28-8 (n = 1) or SP142 (n = 1). In total PDL-1 assay discordance was seen in 14/35 patients (40%). Conclusions: Substantial inter-assay variability in PD-L1 expression in the same tissue specimen was observed in this community NSCLC cohort. However, due to concerns regarding costs and tissue exhaustion, additional and ongoing real-world data needs to be explored to determine the clinical utility of multiple versus a single umbrella PDL-1 assay in this setting.
SATMM: A systems genetic network analysis–based risk stratification tool validated across the disease trajectory of relapsed/refractory multiple myeloma.
7559 Background: Precision medicine in multiple myeloma (MM) requires risk stratification tools that account for the patient-specific impact of cytogenetic abnormalities on disease progression. We developed SATMM (SYGNAL Analytics Tests for MM) based on each patient’s chromosomal abnormality subtype using the previously developed mmSYGNAL network. While originally validated in newly diagnosed (ND) cohorts, this study provides updated validation of SATMM in the relapsed/refractory (RR) setting. Methods: SATMM performance was evaluated in a new RR cohort from the Moffitt Cancer Center, stratified into Early RR (ERR; 1–3 prior events, n=258) and Late RR (LRR; > 3 prior events, n=251). Predictive accuracy was assessed via ROC analysis (AUC) and Kaplan-Meier (KM) survival curves to determine Median Progression-Free Survival (mPFS) and Cox Hazard Ratios (HR). SATMM was benchmarked against the established GEP70 and SKY92 gene panels. Results: As detailed in the accompanying table, SATMM demonstrated robust and superior predictive performance compared to existing panels throughout the RR disease trajectory: Prognostic Accuracy: SATMM achieved superior overall accuracy in both ERR (0.61) and LRR (0.55) groups compared to GEP70 (0.52 and 0.37) and SKY92 (0.54 and 0.43). Early-Stage Performance: In the ERR group, SATMM provided the most significant clinical separation with a Δ mPFS of 10.7 months (KM p =1.60e-05), outperforming GEP70 (8.9 months, KM p =0.00053) and SKY92 (4.4 months, non-significant). Late-Stage Performance: In the heavily pre-treated LRR group, SATMM maintained significant stratification (KM p =0.0068, HR p =0.0078) with an AUC of 0.62, GEP70 showed lesser performance (KM p =0.021, HR p =0.021) while SKY92 failed to provide accurate risk prognosis (AUC=0.43). High-Risk Patient Detection: SATMM identified a larger proportion of high-risk patients (ERR n=93 sen = 0.47; LRR n=117, sen = 0.51) with higher sensitivity (sen) compared to GEP70 (ERR n=25, sen =0.13; LRR n=46, sen = 0.20) and SKY92 (ERR n=59, sen = 0.27; LRR n=111, sen = 0.41). Conclusions: SATMM is a patient-specific prognostic test that effectively stratifies risk based on the unique patient genomic profile compared to current gene panels. These data confirm that SATMM not only accurately predicts progression in ND patients but maintains its high performance across the entire disease trajectory, offering superior accuracy and sensitivity for high-risk detection for RR patients. Method Disease status AUC Δ Median PFS KM pvalue High risk n Cox HR Cox HR pvalue Accuracy Sensitivity SATMM ERR 0.65 10.7 1.6E-05 93 0.512 1.3E-04 0.609 0.47 GEP70 ERR 0.64 8.9 0.00053 25 0.412 7.9E-04 0.523 0.129 SKY92 ERR 0.58 4.4 0.37 59 0.825 0.324 0.539 0.273 SATMM LRR 0.62 3 0.0068 117 0.679 0.0078 0.546 0.508 GEP70 LRR 0.6 3.8 0.021 46 0.661 0.0212 0.367 0.204 SKY92 LRR 0.43 0.74 0.6 111 1.11 0.494 0.426 0.414
Colorectal cancer detection with a multiomic liquid biopsy: Comparing and combining blood- and stool-based tests.
e15641 Background: Colorectal cancer (CRC) remains one of the leading causes of cancer morbidity and mortality worldwide. However, early detection and timely intervention are strongly associated with improved survival and quality of life. The Dxcover Liquid Biopsy Platform is a rapid, multiomic blood-based assay that interrogates samples using infrared spectroscopy to generate a distinctive spectral signature reflective of the global biomolecular composition of the analysed sample. This technology has previously been evaluated as a standalone test for CRC detection, but it also has potential for integration with complementary data sources, including biomarkers and clinical risk factors. Methods: In this study, samples from 1377 patients were collected across sites in the USA (n = 989) and UK (n = 388). Blood was obtained from patients either prior to scheduled colonoscopy or before surgical resection and any anti-cancer therapies. Streck plasma samples were analyzed by the Dxcover Liquid Biopsy Platform to generate spectral data. Carcinoembryonic Antigen (CEA) values were determined for all samples. Fecal hemoglobin levels from fecal immunochemical testing (FIT) were also obtained for the UK samples. Machine learning algorithms were developed to compare test performance and assess combinations. Results: Initially, machine learning models were developed for the spectral dataset alone. The area under the curve (AUC) was 0.95, test sensitivity and specificity were 90% and the model reported consistent detection rates across CRC stages. There was limited diagnostic utility reported for CEA alone (37% sensitivity with 80% specificity). There was no improvement to the spectral model with the inclusion of CEA. For the UK cohort with FIT results, the FIT only model (AUC = 0.83) was enhanced by the addition of spectral data, with the combined model (spectra+FIT) reporting an AUC of 0.90. Conclusions: Integration of this liquid biopsy with orthogonal tests, including FIT or blood-based biomarkers, may enhance early-stage CRC detection. Limitations of existing screening programs underscore the need for novel technologies to support earlier diagnosis and improved outcomes.
Analysis of eligibility and reporting standards for older patients with hormone receptor–positive, HER2-negative metastatic breast cancer (HR+ MBC) in pivotal phase III clinical trials, 2015-2025.
e13087 Background: Despite the high burden of breast cancer among older adults, there remains underrepresentation of older patients in clinical trials and inconsistent reporting of data specific to older patients or geriatric-specific endpoints from pivotal trials. We sought to examine eligibility criteria, representation of older adults, reporting of age-stratified efficacy and safety, and inclusion of geriatric-specific variables from phase III randomized trials leading to FDA approval of new drugs or drug combinations for HR+ MBC over the past decade. Methods: We conducted a search to identify new FDA approvals for treatment of HR+ MBC between 2015 and 2025 and retrieved the corresponding phase III, registration trials. Data were extracted from publications of the primary trial results Extracted variables included the number and percentage of patients aged ≥65 years, which were compared with age-specific incidence estimates from SEER to calculate the Enrollment-to-Incidence Ratio (EIR). Additional trial characteristics extracted included the presence of explicit patient upper age limits, inclusion/exclusion criteria as reported in the trial protocol of primary manuscripts, use of geriatric assessments (any frailty index applied within the trial) and descriptive statistics were performed. Results: We identified 14 new FDA-approvals for HR+ MBC and identified the corresponding 14 pivotal trials (N = 7,848 patients) evaluating CDK4/6 inhibitors, PI3K pathway inhibitors, and other novel agents. Four trials (29%) imposed explicit upper age limits ( > 59, > 86, > 89, > 92 years). None of the trials (0%) performed geriatric assessment or frailty assessment. Nine trials explicitly reported the number of patients aged ≥65 years; among these trials (pooled N = 5,449), the sample-size–weighted mean proportion of patients aged ≥65 years was 31.7% (95% CI, 30.5%–32.9%). This is substantially below the 54.9% disease incidence in this age group (EIR: 0.58 [95% CI: 0.56-0.6]). In reporting progression-free and overall survival results, 42.8% (6/14) and 29% (4/14), stratified by age, respectively. None of the trials reported treatment dose modification or discontinuation rates, safety or quality of life stratified by age. Conclusions: Over the past decade, reports from the pivotal trials leading to FDA approvals for drugs to treat HR+ MBC inconsistently report information pertinent to treating older patients, including age-specific efficacy and safety information. None of the primary reports incorporated measures of frailty or geriatric-specific endpoints. These deficiencies limit the ability to translate efficacy and safety data to real-world older populations and highlight the need for more inclusive trial design, age-stratified reporting, and inclusion of variables specific to this population.
Real-world experience with lifleucel, tumor selection, and dead-in therapy from a single-institution series.
9536 Background: Lifileucel is the first approved cellular therapy for persons with metastatic melanoma after upfront treatment with immune checkpoint inhibitor(s) (ICIs) and BRAF-targeted therapy (if applicable). There is not an established consortium to centralize data surrounding procurement and outcomes. Until then we report a single institution series of lifileucel used in the real-world in patients (pts) with previously treated metastatic melanoma. We evaluate whether there are differences in outcomes based on tumor procurement sites, lead-in therapy, and presence of CNS metastases prior to being treated with lifileucel. In addition, for those with progressive disease (PD) will describe the pattern of progression after lifileucel. Methods: The University of Colorado and its affiliated hospital UCHealth became an authorized treatment center for TIL in November 2024. Between November 2024 and January 2026, tumor tissue procurement was completed on 27 patients diagnosed with a variety of melanoma subtypes, stages and received bridging therapy. 24 pts have been treated with lifileucel. All pts received prior combination ICIs and, if BRAF-V600mut, most received prior treatment with BRAF/MEK inhibitors. Results: 41.7% of pts treated with lifileucel have had at least partial response (PR) as best overall response (BOR). 12% of pts have had a complete response (CR) as BOR. 16.7% had an initial PR but then had PD within 3 mos. 37.5% of pts had PD as BOR. In regard to tumor procurement site, tumor sites were subcutaneous/superficial soft tissue, lymph node, or visceral locations. The response rate from subcutaneous tumors was 16.7%, the response rate from lymph node was 33.3% and the response rate from visceral sites was 42.3%. The response rate from a single tumor used for procurement was 27%. The response rate from combining multiple tumors from different sites was 46%. The most recent therapy prior to lifileucel included: 1) ipilimumab+nivolumab, (5 pts) with 40% response rate, 2) relatlimab+nivolumab, (8 pts) with 75% response rate, 3) anti-PD-1 monotherapy, (3 pts) with 66% response rate, 4) BRAF/MEK inhibitor, (5 pts) with no response to lifileucel. 10 pts had CNS lesions treated prior to lifileucel, 6 pts had CNS lesions treated after lifileucel infusion, 5 were treated with BRAF/MEK prior to receiving lifileucel. 4 pts had PR and 2 had CR of metastatic sites present prior to lifileucel but had PD within 3 mos to new visceral sites. Conclusions: From our single-institution series, using lymph node and visceral tumors for lifileucel generation is associated with higher overall response rate. Combining more than one tumor for lifileucel generation is also associated with higher overall response rate. There were 6 patients that had CNS progression after lifileucel but the majority had used BRAF/MEK prior to receiving lifileucel which raises question of what effect this has on lifileucel efficacy.
Real-world analysis of epigenomic molecular tumor-type prediction for biliary tract cancer in CUP.
4161 Background: Biliary cancers (BTC) are often diagnosed at advanced stages and can present as metastatic liver tumors, making determination of the primary site difficult, leading to frequent cancer of unknown primary (CUP) designations. Conventional histopathology or imaging may misclassify tumor origin, yet accurate distinction is essential because therapies differ for BTC and CUP. A novel liquid biopsy (LB) molecular tumor type (MTT) classifier using epigenomic signatures predicts tumor origin across 14 solid tumors with graded confidence. This study evaluated concordance, genomic features, and real-world outcomes of high-confidence MTT-BTC predictions in patients (pts) with BTC or CUP using the Infinity AI, a clinical-genomics and epigenomics database. Methods: Guardant360 LB results showing high-confidence (>80%) MTT predictions for BTC in pts with BTC or CUP listed on the test requisition form (TRF) were identified. Infinity AI links de-identified clinical, genomic, epigenomic, and administrative claims The LB assay interrogated >700 genes and thousands of methylation regions. First-line (1L) therapies were categorized as chemotherapy (chemo), immunotherapy (IO), or chemo-IO. Real-world time to treatment discontinuation (rwTTD) and time to next treatment (rwTTNT) were compared between TRF-BTC and TRF-CUP cohorts using log-rank tests. Results: Among pts with high-confidence MTT-BTC predictions (TRF-BTC n=410; TRF-CUP n=56), MTT accuracy for TRF-BTC was 51.7%. Female proportion was 58% (TRF-BTC) and 64% (TRF-CUP), respectively. Median ages for TRF-BTC and TRF-CUP pts were 68 years, respectively. The most frequent oncogenic mutations in TRF-BTC were TP53 (34), ARID1A (8), KRAS (7), and IDH1 (6); none had FGFR2 . In TRF-CUP, common mutations included TP53 (3), ARID1A (2), and IDH2 (2). For TRF-BTC pts, rwTTD was longer with 1L IO+chemo vs IO or chemo alone (7.9 vs 4.3 vs 3.4 mo; p<0.0001). TRF-CUP pts showed similar, but non-significant trends. rwTTNT was longer for TRF-BTC pts receiving IO (p=0.00026) and trended higher for TRF-CUP (p=0.069). Conclusions: Guardant360 LB epigenomic–based MTT classifier consistently predicted biliary origin in pts labeled as BTC and in a subset labeled as CUP, with broadly concordant genomic features and real-world treatment patterns. CUP pts with high-confidence MTT-BTC predictions demonstrated outcomes that approximated those of TRF-BTC, supporting biologic similarity and suggesting that some CUP cases may represent occult BTC. These findings support consideration of MTT as a tool to refine uncertain diagnoses, guide BTC-aligned therapy selection (including targetable alterations), and improve clinical trial matching. Larger cohorts with longer follow-up and prospective validation are needed to define the impact of MTT-guided management on patient outcomes.
Image-only and multimodal AI digital pathology biomarkers to demonstrate risk stratification across standard prostate cancer management strategies.
5023 Background: Reliable risk stratification across standard treatment pathways is essential for the clinical adoption of precision medicine biomarkers in localized prostate cancer. We previously developed and validated a multimodal artificial intelligence (MMAI) model that integrates digitized hematoxylin and eosin (H&E) prostate biopsy slides with clinical variables to prognosticate distant metastasis (DM) and prostate cancer specific mortality (PCSM). To evaluate the independent prognostic signal derived from routine histopathology alone without clinical variables, we built an Image-only artificial intelligence model. In this study, we validate the prognostic ability of the new Image-only and existing MMAI models in a multi-institutional cohort of nonmetastatic prostate cancer patients receiving guideline-concordant standard-of-care treatments. Methods: The Image-only model was trained to estimate the risk of 10-year distant metastasis (DM) without clinical inputs. Both the Image-only and MMAI models were evaluated in a cohort of men from 3 institutions with non-metastatic prostate cancer treated with active surveillance (AS), radical prostatectomy (RP), or radiation therapy (RT). The primary endpoint was 10-year risk of DM, with prostate cancer–specific mortality (PCSM) as a secondary endpoint. Prognostic associations were evaluated within each treatment subgroup using Fine–Gray models. Results: Among 886 patients with Image-only scores and 911 with MMAI scores, approximately 36% were managed with AS, 41% with RP, and 23% with RT. The Image-only model continuous raw scores were consistently associated with DM risk across treatment modalities, including AS (subdistribution hazard ratio [sHR] 2.38, p < 0.001), RP (1.99, p < 0.001), and RT (sHR 2.84, p < 0.001). Similarly, the Image-only model calibrated scores showed significant prognostic association with DM across all treatment groups. The MMAI showed consistent prognostic performance for DM using both calibrated and raw scores (raw scores: AS sHR 2.87, p < 0.001; RP sHR 2.12, p < 0.001; RT sHR 2.73, p < 0.001). Both Image-only and MMAI scores were significantly associated with PCSM despite low event rates. Conclusions: Both Image-only and MMAI biomarkers demonstrate consistent prognostic performance across standard prostate cancer management strategies, including AS, RP, and RT, supporting their utility for risk stratification regardless of ultimate treatment selection. These findings highlight the robustness of image-derived prognostic information and demonstrate that routinely available H&E pathology alone captures clinically meaningful risk information that generalizes across treatment contexts. Together, these results support the use of AI-based digital pathology biomarkers for prognostication in localized prostate cancer.
Neighborhood disadvantage and correlation with differences in somatic mutations in a real-world breast cancer cohort.
577 Background: African compared to European genetic ancestry is associated with lower prevalence of clinically actionable alterations despite comparable overall prevalence of driver alterations by ancestry group. This likely contributes to breast cancer (BC) disparity. A limitation of prior ancestry studies is lack of neighborhood-level disadvantage (ND) data, a contributor to BC disparity. The objective of this study was to investigate associations between ND and somatic mutations in a real-world BC cohort. Methods: Somatic mutations from primary and metastatic BC samples and genetic ancestry data were identified from MSK IMPACT (FDA-authorized sequencing panel). Samples sequenced between 2015-2025 were annotated and linked with census-tract level socioeconomic status (Yost Index 1-100; higher reflects increasing ND). Multivariate logistic regression analyzed associations between ND and somatic mutations, adjusting for covariates (FDR p<0.05). Results: 6703 samples (62% primary, 38% metastatic) were analyzed. Mean age was 60 (SD 13). Mean Yost was 26 (SD 24). 67% had ER+/HER2- disease, 9.7% had ER+/HER2+, 5.1% had ER-/HER2+, and 19% had ER-/HER2-. Adjusting for age, BMI, ancestry, and subtype, patients from ND had lower odds of CDH1 (OR 0.99, 95% CI 0.98- 0.99, p<0.001), BRCA1 (OR 0.96, 95% CI 0.94-0.99, p=0.021), and AKT1 (OR 0.98, 95% CI 0.97- 0.99, p=0.006) somatic mutations in metastatic samples. Conclusions: We identify novel associations between ND and somatic mutations. Notably, patients with ND had lower odds of actionable BRCA1 and AKT1 mutations. This highlights the importance of ensuring patients of diverse ancestry and ND have equitable inclusion in clinical sequencing workflows to improve identification of actionable mutations in all populations. Genetic ancestry (GA) and somatic mutational frequency by Yost. TotalN=6703 1 Yost 1-20(Most Advantaged)N=3601 1 Yost 21-40N=1539 1 Yost 41-60N=826 1 Yost 61-80N=477 1 Yost 81-100(Most Disadvantaged) N=260 1 p-value 2 European GA* 0.75(0.36) 0.83(0.30) 0.72(0.38) 0.64(0.39) 0.58(0.39) 0.42(0.35) <0.001 African GA* 0.12(0.27) 0.04(0.16) 0.15(0.30) 0.21(0.34) 0.25(0.36) 0.44(0.38) <0.001 East Asian GA* 0.05(0.21) 0.05(0.21) 0.06(0.22) 0.06(0.21) 0.07(0.23) 0.04(0.17) <0.001 South Asian GA* 0.06(0.16) 0.06(0.17) 0.05(0.15) 0.06(0.17) 0.04(0.12) 0.03(0.08) <0.001 Native American GA* 0.03(0.10) 0.01(0.07) 0.03(0.10) 0.04(0.12) 0.06(0.16) 0.08(0.17) <0.001 Mutational Frequency, metastases N=2536 1 N=1403 1 N=584 1 N=311 1 N=165 1 N=73 1 p-value 2 ^ CDH1 341(13%) 220(16%) 66(11%) 26(8.4%) 18(11%) 11(15%) <0.001 AKT1 127(5.0%) 74(5.3%) 34(5.8%) 12(3.9%) 4(2.4%) 3(4.1%) 0.006 BRCA1 21(0.8%) 13(0.9%) 7(1.2%) 1(0.3%) 0(0%) 0(0%) 0.021 *Calculated using >3,000 common SNP markers. ^MVA, control for age, BMI, ancestry, subtype. 1 Mean (SD), n (%). 2 Pearson’s Chi 2 test; Kruskal-Wallis rank sum test; Fisher’s exact test.
A phase II combination study of galunisertib and capecitabine in patients with chemotherapy-resistant colorectal cancer with peritoneal metastases.
3571 Background: Transforming growth factor β (TGF-β) plays a dynamic role in cancer biology. It acts as a tumor suppressor in early stages of tumor development. Certain tumor types, such as colorectal cancer (CRC), exploit this signaling route to induce pathways that promote tumor growth, enhancing the tumor's aggressiveness. CRC of the mesenchymal subtype exhibits prominent TGF-β activation, contributing to drug resistance, including to 5-fluorouracil (5-FU)-based chemotherapy. This subtype is found in most CRC patients with peritoneal metastases. Therefore, combining TGF-β inhibition with chemotherapy could be a beneficial approach to overcome resistance. Methods: This two-center, open-label, non-randomized proof-of-principle phase II study with a safety lead-in involved combining TGF-β inhibitor galunisertib (150 mg twice daily) with capecitabine (1000 mg/m 2 twice daily) for 14 days in a 28-day cycle. Adult patients with histologically or cytologically proven CRC and confirmed peritoneal metastases were eligible for inclusion if they were in a good clinical condition, had exhausted all their treatment options, and had no gastrointestinal impaired function. In the safety lead-in phase, six patients were evaluated for safety and tolerability, as assessed by the incidence of dose limiting toxicities (DLTs) and adverse events. Phase II followed a Simon two stage design with progression to the second stage when ≥2 objective tumor responses out of 15 patients were reached. Secondary objectives included pharmacokinetics (PK) and pharmacodynamics (e.g. TGF-β expression, immunohistochemical staining for phosphorylated SMAD2/3). Results: As of January 23 2026, inclusion was closed, after 22 patients were recruited. 54% of patients were female, with a median age of 60 years (38-77). No DLTs were observed in 21 DLT-evaluable patients. Only grade 1/2 treatment-related adverse events occurred. The most common events were nausea (23%), hypophosphatemia (18%), and fatigue (18%). No objective responses were observed. Four of 20 patients evaluable for efficacy showed stable disease with a median duration of 4 months. PK analysis of patients in the safety lead-in showed high interpatient variability for galunisertib, capecitabine and 5-FU (CV% AUC (0-12) : between 54%-70%). Galunisertib did not show drug accumulation over time (AUC (0-inf) on cycle 1 day 1 and AUC (0-12) on cycle 1 day 14 were 7.6 and 6.6 µg*h/mL, respectively). The PK of capecitabine and 5-FU were similar to those described in the literature. Pharmacodynamic analyses are ongoing. Conclusions: The combination of galunisertib and capecitabine did not result in objective responses or long term stable disease in chemotherapy-resistant CRC patients with peritoneal metastases. The treatment was well tolerated and PK results for both compounds were similar to those described in the literature. Clinical trial information: NCT05700656 .
Development of multiplex digital PCR–based <i>ESR1</i> and <i>PIK3CA</i> mutation assays for therapeutic decision-making in HR+/HER2− breast cancer.
e13008 Background: Mutations in ESR1 and PIK3CA are clinically important molecular alterations associated with treatment response and therapeutic decision-making in HR+/HER2− breast cancer. Accurate detection of these mutations is essential for guiding targeted therapy and supporting companion diagnostic (CDx) strategies. While next-generation sequencing (NGS) is widely used, its routine clinical implementation may be limited by cost, long turnaround time (TAT), and workflow complexity. Multiplex digital PCR (dPCR) provides a simplified and cost-efficient alternative, enabling highly sensitive and specific mutation detection with low DNA input. Based on these advantages, we developed multiplex dPCR–based assays for rapid detection of ESR1 and PIK3CA mutations. Methods: The DigiPlex ESR1 Mutation Test and DigiPlex PIK3CA Mutation Test were developed as multiplex dPCR assays targeting clinically actionable mutations in HR+/HER2− breast cancer. The ESR1 assay detects key hotspot mutations associated with acquired endocrine resistance, while the PIK3CA assay detects 25 mutations across exons 1, 4, 7, 9, and 20. Both assays were optimized for use on the Roche Digital LightCycler Analyzer with multi-channel fluorescence–based detection and low DNA input. Analytical performance was evaluated using reference materials and contrived samples with predefined mutant allele frequencies, including DNA derived from FFPE tissue and plasma, to assess sensitivity, specificity, limit of detection (LoD), and precision. Results: The DigiPlex ESR1 and PIK3CA Mutation Tests demonstrated robust multiplex detection performance on a dPCR platform. Reliable mutation detection was achieved using low DNA input (5–10 ng) from both FFPE tissue and plasma-derived DNA. All targeted mutations were consistently detected at the evaluated LoD, demonstrating high analytical sensitivity. No false-positive or false-negative results were observed, indicating high analytical specificity. Clear fluorescence signal separation and reproducible quantitative performance were maintained across independent runs on the Digital LightCycler Analyzer. Conclusions: Multiplex digital PCR–based DigiPlex assays for ESR1 and PIK3CA mutation detection were successfully developed and analytically validated on the Roche Digital LightCycler Analyzer. These assays combine high analytical sensitivity and specificity with low DNA input and a streamlined workflow, representing a practical alternative to sequencing-based approaches for molecular testing in HR+/HER2− breast cancer. Accurate detection of ESR1 and PIK3CA mutations using multiplex dPCR may support individualized molecular diagnosis and CDx-driven therapeutic strategies. Clinical validation studies are ongoing.