Abstract 2139: [18F]-FMISO-PET imaging to characterize the impact of obesity on the hypoxic tumor microenvironment during immunotherapy.

Abstract Introduction: 40% of patients with triple-negative breast cancer (TNBC) also have obesity (BMI>30 kg/m2). Clinical studies have demonstrated that obesity worsens cancer outcomes, urging investigation of novel therapeutic targets within the obese tumor microenvironment (TME). Preclinical studies have shown that obesity increases tumor hypoxia, diminishing immunotherapy (IMT) response via reduced IMT distribution and pro-tumoral immune shifts. [18F]-fluoromisonidazole (FMISO) positron emission tomography (PET) can non-invasively quantify hypoxia, allowing for patient stratification by hypoxia levels to predict IMT response and inform the use of secondary hypoxia-activated prodrugs, like evofosfamide (EVO). Methods: C57/Bl6J mice were fed a high-fat diet (HFD, n=68) or low-fat diet (LFD, n=19) for 13 weeks prior to E0771 implantation in the 3rd mammary fat pad on day -12. To explore the impact of obesity on baseline tumor hypoxia, a cohort of LFD and HFD were imaged with [18F]-FMISO PET. A second HFD cohort was split into controls (saline, n=19) or IMT (anti-CTLA-4 and anti-PD-1 every 3 days, n=49). Mice were imaged with [18F]-FMISO PET on days 0, 5, and 12, and signal was quantified via standard uptake value (SUV) metrics. On day 6, IMT mice were randomized to continue with IMT (n=30) or also receive EVO (n=19). Based on ROC analysis, a threshold of day 5 tumor to background ratio (TBR) of the muscle of 1.3 was used to categorize tumors as hypoxic/normoxic. A correlation evaluated the predictive ability of day 5 TBR on long-term response (defined by RECIST criteria). A subset of treatment groups was euthanized at imaging timepoints for flow cytometry of immune populations. A two-way ANOVA was used for statistical analysis. Results: HFD mice demonstrated significantly increased tumor hypoxia (SUVmean) compared to LFD mice on day 0 (p<0.0001) without tumor volume differences. In HFD, IMT resulted in a 25% response rate by day 21. Day 5 TBR was inversely correlated with long term survival for IMT-treated mice (p=0.003, r2=0.5). When stratifying tumors as normoxic or hypoxic, adding EVO to IMT demonstrated a 21% reduction of tumor hypoxia (SUVmax) within hypoxic tumors compared to IMT alone (p=0.008). Interestingly, combination therapy led to a 330% and 173% increase in M2-like macrophages (p=0.06) and regulatory T cells (p=0.06) compared to IMT alone and did not reveal improvements in long-term tumor volume. Conclusions: Our study supports ongoing literature demonstrating increased hypoxia in tumors in a HFD environment and that [18F]-FMISO stratification predicted response to immunotherapy in a TNBC model. Previous studies showed that lean tumor models have additive response when adding EVO to IMT in hypoxic tumors; however, our study demonstrated that EVO reduces hypoxia in hypoxic, obese tumors, but this did not salvage IMT response, likely due to increased tumorigenic immune changes. Citation Format: Corinne Crawford, Shannon Elizabeth Lynch, Jonathan Moye, Chloe T. DeMellier, Patrick N. Song, Kelsey M. O'Brien, Benjamin P. Lee, Benjamin Larimer, Anna G. Sorace. [18F]-FMISO-PET imaging to characterize the impact of obesity on the hypoxic tumor microenvironment during immunotherapy [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 2139.