Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Mini Oral session 2

LBA1 - Tissue-agnostic response predictor for immune checkpoint inhibitor therapy based on MKI67, FOXC1 and PDL1 expression


08 Dec 2022


Mini Oral session 2


Clinical Research;  Cytotoxic Therapy;  Laboratory Diagnostics;  Translational Research;  Immunotherapy

Tumour Site

Breast Cancer;  Head and Neck Cancers


Partha Ray


Annals of Oncology (2022) 16 (suppl_1): 100100-100100. 10.1016/iotech/iotech100100


P.S. Ray, T. Ray, R. Hussa

Author affiliations

  • Onconostic Technologies, Evanston/US


Login to access the resources on OncologyPRO.

Abstract LBA1


Immune checkpoint inhibitors (ICIs) have clinical efficacy in the neoadjuvant setting in multiple solid cancer types. However, suitable tissue-agnostic complementary diagnostics to help guide and tailor treatment recommendations are still lacking. Ki67, a routinely used proliferation marker predicts efficacy of chemotherapeutics but not ICIs. Forkhead Box C1 (FOXC1), a transcriptional driver of cell plasticity/partial EMT/metastasis was recently demonstrated to have potential value in predicting therapeutic efficacy of chemotherapy+immunotherapy in TNBC. PDL1, a marker of immune evasion, is a proven companion diagnostic for ICI therapy in some but not all situations. We sought to evaluate a Ki67+FOXC1+PDL1-expression based response predictor as a possible complementary diagnostic for ICI therapy in the neoadjuvant setting across different cancer types.


Pre-treatment tumor biopsy MKI67, FOXC1 and PDL1 mRNA expression values were retrospectively obtained from patients who had been enrolled and treated in independent clinical trial cohorts (1. I-SPY2 TNBC: AC+taxane+pembrolizumab; 2. I-SPY2 TNBC: AC+olaparib+paclitaxel+durvalumab; 3. HNSCC: Nivolumab/Pembrolizumab). Optimized biomarker cut-off values based on model area-under-curve were leave-one-out cross validated in the first dataset for Predicted Responder (PR) and Predicted Non-responder (NR) prediction. The unmodified strategy was then validated in the other datasets.


Observed response rates were 66% (n=29), 43% (n=21) and 11% (n=102) in the different regimens in datasets 1-3. In the biomarker-defined PR groups (n=22, n=12 and n=38) response rates were 82%, 75% and 21% in datasets 1-3. In the biomarker-defined NR groups (n=7, n=9 and n=64) response rates were 14%, 0% and 4% in datasets 1-3 (OR=27, 2.5-291.2 95%CI, p=0.003; OR=52, 2.3-1141.0 95%CI, p=0.01; OR=5, 1.3-21.9, 95%CI, p=0.008 respectively). Multiple logistic regression models may further improve predictive accuracy.


Complementary diagnostic role of pre-treatment MKI67+FOXC1+PDL1 expression merits prospective clinical trial evaluation in multiple cancer types treated with neoadjuvant ICIs alone or in combination with chemotherapeutics.

Legal entity responsible for the study

Onconostic Technologies, Inc.


Has not received any funding.


P.S. Ray: Non-Financial Interests, Institutional, Advisory Role: Onconostic Technologies: Practicing surgical oncologist and clinical investigator, identified as the presenting author on the submitted abstract. PSR is not an employee of the company (Onconostic Tehcnologies), but is founder and Chairman of the Scientific Advisory Board of the company which is an unpaid position. T. Ray: Financial Interests, Institutional, Project Lead: Onconostic Technologies. R. Hussa: Financial Interests, Institutional, Leadership Role: Onconostic Technologies.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.