Cell-free DNA (cfDNA) sequencing has demonstrated great potential for early cancer detection. However, most large-scale studies have focused only on either targeted methylation sites or whole-genome sequencing, limiting comprehensive analysis that integrates both epigenetic and genetic signatures. In this study, we present a platform that enables simultaneous analysis of whole-genome methylation, copy number, and fragmentomic patterns of cfDNA in a single assay. Using a total of 950 plasma (361 healthy and 589 cancer) and 240 tissue samples, we demonstrate that a multifeature cancer signature ensemble (CSE) classifier integrating all features outperforms single-feature classifiers. At 95.2% specificity, the cancer detection sensitivity with methylation, copy number, and fragmentomic models was 77.2%, 61.4%, and 60.5%, respectively, but sensitivity was significantly increased to 88.9% with the CSE classifier (p value < 0.0001). For tissue of origin, the CSE classifier enhanced the accuracy beyond the methylation classifier, from 74.3% to 76.4%. Overall, this work proves the utility of a signature ensemble integrating epigenetic and genetic information for accurate cancer detection.

Purpose: Clinical implications of neoadjuvant immunotherapy in patients with locally advanced but resectable head and neck squamous cell carcinoma (HNSCC) remain largely unexplored. Patients and Methods: Patients with resectable HNSCC were randomized to receive a single dose of preoperative durvalumab (D) with or without tremelimumab (T) before resection, followed by postoperative (chemo)radiotherapy based on multidisciplinary discretion and 1-year D treatment. Artificial intelligence (AI)–powered spatial distribution analysis of tumor-infiltrating lymphocytes and high-dimensional profiling of circulating immune cells tracked dynamic intratumoral and systemic immune responses. Results: Of the 48 patients enrolled (D, 24 patients; D+T, 24 patients), 45 underwent surgical resection per protocol (D, 21 patients; D+T, 24 patients). D±T had a favorable safety profile and did not delay surgery. Distant recurrence-free survival (DRFS) was significantly better in patients treated with D+T than in those treated with D monotherapy. AI-powered whole-slide image analysis demonstrated that D+T significantly reshaped the tumor microenvironment toward immune-inflamed phenotypes, in contrast with the D monotherapy or cytotoxic chemotherapy. High-dimensional profiling of circulating immune cells revealed a significant expansion of T-cell subsets characterized by proliferation and activation in response to D+T therapy, which was rare following D monotherapy. Importantly, expansion of specific clusters in CD8+ T cells and non-regulatory CD4+ T cells with activation and exhaustion programs was associated with prolonged DRFS in patients treated with D+T. Conclusions: Preoperative D±T is feasible and may benefit patients with resectable HNSCC. Distinct changes in the tumor microenvironment and circulating immune cells were induced by each treatment regimen, warranting further investigation.

Tumor-infiltrating lymphocytes (TIL) have been suggested as an important prognostic marker in colorectal cancer, but assessment usually requires additional tissue processing and interpretational efforts. The aim of this study is to assess the clinical significance of artificial intelligence (AI)-powered spatial TIL analysis using only a hematoxylin and eosin (H&E)-stained whole-slide image (WSI) for the prediction of prognosis in stage II–III colon cancer treated with surgery and adjuvant therapy. In this retrospective study, we used Lunit SCOPE IO, an AI-powered H&E WSI analyzer, to assess intratumoral TIL (iTIL) and tumor-related stromal TIL (sTIL) densities from WSIs of 289 patients. The patients with confirmed recurrences had significantly lower sTIL densities (mean sTIL density 630.2/mm² in cases with confirmed recurrence vs. 1021.3/mm² in no recurrence, p < 0.001). Additionally, significantly higher recurrence rates were observed in patients having sTIL or iTIL in the lower quartile groups. Risk groups defined as high-risk (both iTIL and sTIL in the lowest quartile groups), low-risk (sTIL higher than the median), or intermediate-risk (not high- or low-risk) were predictive of recurrence and were independently associated with clinical outcomes after adjusting for other clinical factors. AI-powered TIL analysis can provide prognostic information in stage II/III colon cancer in a practical manner.

Purpose: Recently, anti-programmed cell death-1/anti-programmed cell death ligand-1 (anti-PD1/L1) immunotherapy has been demonstrated for its efficacy when combined with cytotoxic chemotherapy in randomized phase 3 trials for advanced biliary tract cancer (BTC). However, no biomarker predictive of benefit has been established for anti-PD1/L1 in BTC. Here, we evaluated tumor-infiltrating lymphocytes (TIL) using artificial intelligence-powered immune phenotype (AI-IP) analysis in advanced BTC treated with anti-PD1. Experimental Design: Pretreatment hematoxylin and eosin (H&E)–stained whole-slide images from 339 patients with advanced BTC who received anti-PD1 as second-line treatment or beyond, were employed for AI-IP analysis and correlative analysis between AI-IP and efficacy outcomes with anti-PD1. Next, data and images of the BTC cohort from The Cancer Genome Atlas (TCGA) were additionally analyzed to evaluate the transcriptomic and mutational characteristics of various AI-IP in BTC. Results: Overall, AI-IP were classified as inflamed [high intratumoral TIL (iTIL)] in 40 patients (11.8%), immune-excluded (low iTIL and high stromal TIL) in 167 patients (49.3%), and immune-desert (low TIL overall) in 132 patients (38.9%). The inflamed IP group showed a substantially higher overall response rate compared with the noninflamed IP groups (27.5% vs. 7.7%, P < 0.001). Median overall survival and progression-free survival were significantly longer in the inflamed IP group than in the noninflamed IP group (OS, 12.6 vs. 5.1 months; P = 0.002; PFS, 4.5 vs. 1.9 months; P < 0.001). In the TCGA cohort analysis, the inflamed IP showed increased cytolytic activity scores and IFNγ signature compared with the noninflamed IP. Conclusions: AI-IP based on spatial TIL analysis was effective in predicting the efficacy outcomes in patients with BTC treated with anti-PD1 therapy. Further validation is necessary in the context of anti-PD1/L1 plus gemcitabine–cisplatin.