Abstract Background: Immune checkpoint inhibitors (ICIs) have transformed contemporary cancer care, yet few patients experience objective responses. There is a need for novel peripheral biomarkers that can not only identify patients likely to respond but also distinguish those with excellent, durable benefit to enable improved patient selection, therapeutic monitoring, and target discovery. Methods: We prospectively collected serial blood samples from patients with advanced/metastatic solid tumors treated at Johns Hopkins on anti-PD(L)1-based therapy from 2021-2024. Patients with complete or durable partial response with progression-free survival (PFS) greater than one year were classified as excellent responders (ERs); other patients were classified as non-ERs. Peripheral blood mononuclear cells were analyzed using CyTOF to quantify 28 immune cell clusters with associated functional markers, and plasma samples were analyzed using a Luminex multiplex assay to quantify concentrations of 39 cytokines at baseline and early-on-treatment (month 1 or 2). Results: Our total cohort included 123 patients, 30 of whom were classified as ERs. The most common tumor types were HCC, head and neck cancer, RCC, and bladder cancer. Our cytokine cohort included 118 patients. At baseline, ERs had significantly elevated Th17-associated cytokines, IL-17F, IL-21, and IL-23 compared to non-ERs (Wilcoxon rank-sum, p0.05). Non-ERs had elevated baseline IL-8 concentrations (p0.05). These results persisted early-on-treatment, with the addition of elevated on-treatment IL-6 being associated with non-ER (p0.05). Patients with greater than median IL-17F concentrations at baseline demonstrated improved PFS (p=0.075), while greater than median IL-6 and IL-8 baseline concentrations demonstrated worse OS (p=0.053 and p0.001). Our CyTOF cohort included 116 patients. As compared to non-ERs, ERs were associated with greater early-treatment expansion of NK cells and double negative (CD4-CD8-) T-cell as well as multiple Th2-central memory clusters (p0.05). Th17 cells were also increased in ERs as compared to non-ERs, although this result did not reach significance (p=0.06). Immune cell functional marker analysis in non-ERs shows that Th17 cells had increased proliferation of exhausted T-cells (Ki67+TIGIT+) from baseline to early-on-treatment (Linear mixed model, p 0.01), which was not present in ERs. Conclusions: In a prospective, pan-tumor cohort, we find that Th17 cytokines, IL-17F, IL-21, and IL-23, as well as Th2-central memory cells are associated with excellent ICI response, in addition to re-demonstrating the negative prognostic implications of Th17 cytokine, IL-6, as well as IL-8. Our study highlights the Th17 pathway as a polarizing determinant of excellent ICI response and represents a comprehensive exploration of other peripheral cytokines and immune cells in mediating durable ICI benefit. Citation Format: Howard L. Li, Soren Charmsaz, Chester Kao, Carlotta Pazzi, Madelena Brancati, James Leatherman, Nicole E. Gross, Royce P. Lee, Xiyu Zhao, Christopher Thoburn, Jeannie Hoffman-Censits, Evan J. Lipson, Yasser Ged, Marina Baretti, G Scott Chandler, Rajat Mohindra, Laura Tang, Sanjay Bansal, Aditi Guha, Elizabeth Jaffee, Daniel J. Zabransky, Won Jin Ho, Mark Yarchoan, Mari Nakazawa. Pan-tumor investigation of peripheral biomarkers of excellent response to anti-PD(L)1-based therapy abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6456.
Building similarity graph...
Analyzing shared references across papers
Loading...
Li et al. (Fri,) studied this question.
www.synapsesocial.com/papers/69d1fd29a79560c99a0a3041 — DOI: https://doi.org/10.1158/1538-7445.am2026-6456
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:
Howard L. Li
Soren Charmsaz
Chester Kao
Cancer Research
Johns Hopkins University
Sidney Kimmel Comprehensive Cancer Center
Roche (Switzerland)
Building similarity graph...
Analyzing shared references across papers
Loading...