Solaß, Wiebke

Importance Tumor-infiltrating lymphocytes (TILs) are a provocative biomarker in melanoma, influencing diagnosis, prognosis, and immunotherapy outcomes; however, traditional pathologist-read TIL assessment on hematoxylin and eosin–stained slides is prone to interobserver variability, leading to inconsistent clinical decisions. Therefore, development of newer TIL scoring approaches that produce more reliable and consistent readouts is important. Objective To evaluate the analytical and clinical validity of a machine learning algorithm for TIL quantification in melanoma compared with traditional pathologist-read methods. Design, Setting, and Participants This multioperator, global, multi-institutional prognostic study compared TIL scoring reproducibility between traditional pathologist-read methods and an artificial intelligence (AI)-driven approach. The study was conducted using retrospective cohorts of patients with melanoma between January 2022 and June 2023 across 45 institutions, with tissue evaluated by participants from academic, clinical, and research institutions. Participants were selected to ensure diverse expertise and professional backgrounds. Main Outcomes and Measures Intraclass correlation coefficient (ICC) values were calculated for the manual and AI-assisted arms using log-transformed data. Kendall W values were calculated for Clark scores (brisk = 3, nonbrisk = 2, and sparse = 1). Reliabilities of ICC and W values were classified as moderate (0.40-0.60), good (0.61-0.80), or excellent (>0.80). AI TIL measurements were dichotomized using the 16.6 and median cutoffs. Univariable and multivariable Cox regression analyses assessed the prognostic value of TIL scores adjusted for clinicopathologic variables. Results There were 111 patients with melanoma in the independent testing cohort (median [range] age at diagnosis, 61.0 [25.0-87.0] years; 56 [50.5%] male) who contributed melanoma whole tissue sections. A total of 98 participants evaluated TILs on 60 hematoxylin and eosin–stained melanoma tissue sections. All 40 participants in the manual arm were pathologists, while the AI-assisted arm included 11 pathologists and 47 nonpathologists (scientists). The AI algorithm demonstrated superior reproducibility, with ICCs higher than 0.90 for all machine learning TIL variables, significantly outperforming manual assessments (ICC, 0.61 for AI-derived stromal TILs vs Kendall W , 0.44 for manual Clark TIL scoring). AI-based TIL scores showed prognostic associations with patient outcomes (n = 111) using the median cutoff approach with a hazard ratio (HR) of 0.45 (95% CI, 0.26-0.80; P = .005), and using the cutoff of 16.6, with an HR of 0.56 (95% CI, 0.32-0.98; P = .04). Conclusions and Relevance In this prognostic study of TIL quantification in melanoma, the AI algorithm demonstrated superior reproducibility and prognostic associations compared with traditional methods. Although the retrospective nature of the cohorts limits demonstration of clinical utility, the publicly available dataset and open-source AI tool offer a foundation for future validation and integration into melanoma management.

125 Background: Pressurized intraperitoneal aerosolized chemotherapy (PIPAC) is being evaluated as a novel minimally invasive palliative treatment of peritoneal metastases (PM). Prior studies have established the feasibility and safety of repeated PIPAC treatments in gastrointestinal and gynecologic cancers. The goal of the present phase 1 trial was to establish the safety and feasibility of PIPAC oxaliplatin in a highly chemotherapy refractory colorectal and appendiceal cancer patient population. Methods: Patients with biopsy-proven peritoneal metastases from colorectal or appendiceal cancer underwent up to three PIPAC treatments using oxaliplatin (92 mg/m2) with a six-week interval at two academic centers. Patients with bowel obstruction, extra-peritoneal metastases, or poor performance status (ECOG>2) were excluded. PIPAC was nebulized over 5 min with a 30 min aerosol dwell time. Apart from the first PIPAC cycle, the patients also received a sensitizing dose of 5FU/LV (400mg/m2) within 24 hours of the procedure. Primary end point was safety as assessed by dose limiting toxicities within 6 weeks of the first PIPAC. Secondary endpoints included safety with the addition of 5FU/LV, efficacy, surgical morbidity, technical failure rate, progression-free and overall survival, pharmacokinetics (PK), and quality of life assessment. Results: A total of 8 patients were included: 5 colorectal; and 3 appendiceal. Median number of prior chemotherapy cycles was 2 (Interquartile range – IQR; 1.5-3.5). All patients were refractory to systemic oxaliplatin-based chemotherapy. Median time from diagnosis to PIPAC was 16 months (IQR; 5.6, 17.5) and Peritoneal Carcinomatosis Index was 29 (IQR; 20.5, 31.5). Five (62.5%) patients completed 3 PIPAC cycles while in 3 (37.5%) patients PIPAC was discontinued due to disease progression within the peritoneal cavity. No surgical complication or dose limiting toxicity was observed. Only one patient developed grade 3 treatment-related toxicity after first PIPAC (anemia), and another patient after second PIPAC (abdominal pain and anemia). At the completion of PIPAC treatment 5 patients had stable disease and 3 had disease progression. Pharmacokinetic, histologic response and preliminary survival data will be presented at the meeting. Conclusions: PIPAC with oxaliplatin is safe and feasible in a highly chemotherapy refractory cohort of appendiceal and colorectal carcinomatosis patients with or without sensitizing 5-FU/ LV. Clinical trial information: NCT04329494.