HLA pharmacogenetic analyses in patients treated with pembrolizumab + placebo (mITT population; post-hoc analyses)

Regulatory T cells (Treg) act as a powerful barrier to effective antitumor immunity. Although manipulating Treg is a promising anticancer strategy, doing so while sparing general immune tolerance has been a challenge. Identifying factors specifically expressed in tumor-infiltrating Treg is therefore important for better understanding cancer pathogenesis and identifying novel therapeutic targets that enhance antitumor immunity. We show that T cell Immunoglobulin and Mucin 3 (Tim-3) expression on tumor Treg is required for the function and survival of these cells, in part through Akt and FOXO1 signaling. Deleting Tim-3 in Treg leads to delayed tumor-specific T-cell exhaustion and lower tumor burden, without altering peripheral homeostasis. Similar effects were noted when Tim-3 was only deleted from half of the Treg or when deletion was delayed until after tumor inoculation. Moreover, Treg-specific deletion of Tim-3 cooperated with PD-1 checkpoint blockade to sensitize an immunotherapy-resistant tumor model. In addition, a decrease in Tim-3+ tumor Treg correlated with responsiveness to PD-1/LAG-3 combination checkpoint blockade in a human clinical trial. Overall, our data provide evidence that Tim3-expressing Treg are a promising target to modulate tumor-specific immune responses.

Supplemental Table 4-super_clones_preclone_MPR_vs_NR_post_DEG_sig

Supplemental Figure 3. Boxplot of NF- κB values across the three cluster subtypes. By unsupervised consensus clustering and self-organizing map (SOM) approaches, we identified three distinct biological subtypes within HPV-positive HNSCC: Cl1 (immune-related), Cl2 (epithelial-mesenchymal transition [EMT]-related), and Cl3 (proliferation-related). This classification holds prognostic significance, with Cl1 exhibiting the most favorable survival outcomes, Cl2 associated with the poorest prognosis, and Cl3 displaying an intermediate survival rate [Locati, 2019]. In contrast to recent studies that classified HPV-related HNSCC into immune and keratinocyte subtypes, our analysis confirmed the immune-associated subtype while further resolving the keratinocyte group into two distinct clusters with differing biological profiles and prognostic implications. NF-κB signature is significantly activated in Cl1-immune related cluster compared to the other 2 clusters: A) Metanalyisis-HPV267 cohort (1); B) BD2-HPV286 cohort (2).

Lymph ctDNA outperforms pathology biomarkers.

Supplemental Figure 8A. UMAP dimensional reduction plots with marker gene validating cell type assignments. Marker genes related to epithelial cells, endothelial cells, fibroblasts, mast cells, dendritic cells, and myeloid cells. Supplemental Figure 8B. UMAP dimensional reduction plots with marker gene validating cell type assignments. Marker genes related to natural killer, B, T and cycling cells.

Supplemental Figure 9. Marker genes by cluster using Leiden method.

Orthogonal Validation of Lymph and Plasma MRD Detection.

Supplemental Table 6-Super-Expanders(Pre)MPRvsmR_ovlp_MPRvsNR

Supplemental Table 2. Association of NF-κB high and low groups with clinical parameters in both Metanalysis-HPV267 (1) and BD2-HPV286 (2). P-value assessed by χ2 test, with the exception of age (t-test). Abbreviations: CHT, chemotherapy; RT, radiotherapy.

2514 Background: Immune checkpoint inhibitors (IO) have revolutionized the treatment of advanced solid tumors. Despite their widespread adoption, the optimal duration of IO has not been clearly established, and this remains an unmet clinical need. Methods: In our prospective trial, patients (pts) with advanced solid tumors who achieved at least stable disease (SD) after 1 year of standard-of-care PD-1/PD-L1 inhibitors were randomized 1:1 to either stop or continue IO beyond one year. Pts who stopped IO and then progressed could be retreated with IO at physician discretion. Primary endpoint was cure rate, defined as the proportion of pts without progressive disease, death, or initiation of new treatment (including re-treatment with IO) at five years. Cure rate was calculated using Kaplan-Meier (KM) estimates with 95% confidence intervals (CI) and compared between arms using a multivariable logistic regression model to adjust for pre-specified stratification factors of tumor type and expected efficacy by IO indication, and reported as an odds ratio (OR) with CI and p-value. Secondary endpoints included PFS, OS, and incidence of iRAEs. Median PFS and OS were estimated from KM curves, with hazard ratios (HR) and p-values obtained from stratified Cox proportional hazards models. All analyses were conducted using R version 4.4.2. Results: From 2019 to 2025, leveraging our central hub and community network, a total of 161 pts (Stop = 81, Continue = 80) were enrolled. The trial was closed early due to slow accrual (goal 578 pts). Median age was 71 and the most common tumor types were: NSCLC(39.8%), mismatch repair-deficient tumors(18.6), RCC(11.8%), HNSCC(8.1%), and melanoma (7.5%). There was no difference in baseline characteristics between arms. With a median follow up of 43.2 months (mo), cure rate was 45% (32%-63%)in the stop arm vs 38% (27-54%) in the continue IO arm (OR 2.26, 95% CI [0.62-8.22], p = 0.22). Median PFS was 55.4(27.5-NR) vs. 31.2 mo(20-NR) HR 0.77, 95% CI (0.48-1.24), p = 0.28 and median OS was NR (55.4 vs. NR) vs. 53.5 mo(40.8-NR) HR 0.54, 95%CI (0.3-0.99), p = 0.04 for the stop vs continue IO arms, respectively. In the continue IO arm, the median number of IO doses was 11 (range 0-36) and 52 of these pts discontinued IO with the most common reason being disease progression (42.3%), and 7.7% discontinued for toxicity. In the stop IO arm, 17 pts who progressed were retreated with IO; 20% achieved subsequent partial response and the median PFS was 10.5 mo(6.5-NR) and median OS was 26.7 mo(10.5-NR) after retreatment. Conclusions: Our trial is unique as a prospective randomized evaluation of length of IO treatment in advanced solid tumors. Cure rates and PFS were similar across arms and OS was improved in pts that discontinued IO at 1 year. These findings suggest that stopping IO at 1 year may be as effective as indefinite IO therapy, though validation in a larger, adequately powered trial is needed. Clinical trial information: NCT04157985 .

Supplemental Table 16-NR_super_expanders_degs_Pre-existing_vs_New

Supplemental Table 2-super_clones_allclone_MPR_vs_mR_post_DEG_sig

Lymph ctDNA outperforms pathology biomarkers.

Supplemental Figure 2. UMAP dimensional reduction plot with malignancy probabilities from scMalignantFinder. Low probability predictions (<0.5) were imputed with to a value of 0.5 for visual clarity.

Curative immune checkpoint blockade (ICB)-based multimodal therapy is now widely used across oncology; yet drivers of efficacy and resistance in most cancer types, including squamous cell carcinoma of the head and neck (SCCHN), are not well understood. To address this issue, we comprehensively characterized the tumor genome, microenvironment, microbiome, and TCR repertoire in a phase III international randomized trial to identify factors that shape outcomes to avelumab plus chemoradiotherapy versus chemoradiotherapy alone. Patients receiving avelumab whose tumors contained distinct immunologic and genetic features had superior outcomes compared to patients in the control arm that did not receive immunotherapy. In contrast, avelumab-treated patients with increased myeloid/neutrophil signatures had poorer outcomes compared to placebo. Strikingly, these tumors possessed telltale intratumoral bacteria, elevated tumor-associated neutrophils, high systemic neutrophil-to-lymphocyte ratio, suppressed levels of adaptive immunity, and were the least likely to respond to therapy. We use these findings to define tumor ecosystem classes that associate with response and resistance to combination chemoimmunotherapy. Importantly, our data demonstrate for the first time, the effects of intratumoral bacteria on ICB response within the context of a randomized trial. These discoveries enhance our understanding of combination immunotherapy response, provide a highly useful multi-omic resource, and identify unanticipated interactions between ICB and CRT that may guide future therapeutic strategies. Tyler Alban, Nadeem Riaz, Robert Haddad, Michelle Saul, Vladamir Makarov, Yingjie Zhu, Ezra Cohen, Robert Ferris, Peter Chang, Jin-Ching Lin, Amanda Pyrri, Prerana Parthasarathy, Ardijana Novaj, Mruniya Gawali, Jennifer Ko, Phineas Hamilton, Natalie Silver, Ivan Juric, Daniel Chawla, Ana Gradissimo, Daniel McGrail, Craig Davis, Nancy Lee, Timothy Chan. Tumor Ecosystem and Microbiome Features Associated with Efficacy from Avelumab-based Multimodal Therapy in a Phase III Randomized Trial [abstract]. In: Proceedings of the AACR Immuno-Oncology Conference (AACR IO): Discovery and Innovation in Cancer Immunology: Revolutionizing Treatment through Immunotherapy; 2026 Feb 18-21; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Immunol Res 2026;14(2 Suppl):Abstract nr C064.

Supplemental Figure 6. Survival Plot. Prognostic Value of NanoString NAC Classifier Measurements. In Figure 4, PFS was shown for (E) all patients and (F) de-escalation candidates in UNC and E1308 cohorts (non-smoker,

BACKGROUND:Head and neck cancer (HNC) treatment often involves multimodal regimens, with 30-40% of patients experiencing significant toxicity or treatment intolerance. Existing risk models do not fully capture the complex interplay of sarcopenia, frailty, and treatment factors, limiting their utility for surgical decision-making. METHODS:We developed and validated the modified Head and Neck Surgery Risk Index (mHNSRI), a tool incorporating multimodal domains of sarcopenia (cervical paraspinal skeletal muscle index, SMI), frailty (modified Frailty Index, Clinical Risk Analysis Index) and treatment factors, using an ambispective cohort of patients with operable HPV-negative stage II-IV HNC. Treatment intolerance was defined as a severe toxicity or therapy non-completion. Logistic regression and recursive partitioning analysis were used to identify predictors variables for treatment intolerance. The mHNSRI's performance was compared to traditional preoperative risk indices using internal and external validation cohorts. RESULTS:In 568 patients, including 343 in the derivation and 225 in the validation cohort, the mean age was 65 with 228 (40%) patients being frail and 259 (46%) sarcopenic. Treatment intolerance occurred in 270 (48%) patients. Independent predictors of treatment intolerance included low cervical paraspinal SMI, elevated Risk Analysis Index (RAI), reduced high-to-low attenuation SMI ratio and obesity. Sarcopenia and frailty had a synergistic effect on treatment intolerance. Adding cervical paraspinal SMI and high-to-low attenuation SMI ratio improved model performance (Δc-statistic 0.069, P = 0.05). The mHNSRI had an AUC of 0.71 (95% CI:0.61-0.80) in external validation cohort and demonstrated superior predictive ability compared to the RAI (Δc 0.14, modified Frailty Index-5 (Δc 0.13), ASA (Δc 0.2) and Charlson Comorbidity Index (Δc 0.14, all P < 0.05) CONCLUSION: The mHNSRI offers a personalized approach to preoperative risk stratification in HNC surgery, integrating sarcopenia and frailty metrics to better predict treatment intolerance.

Supplemental Table 4. Association of NF-κB high and low groups with clinical parameters in combined UNC &E1308 cohorts (1) as pictured in Figure 4D-E. P-value assessed by χ2 test, except for age (assessed with t-test).