Abstract OT1-20-02: Intralesional injection of anti-PD-1 (pembrolizumab) and OX40L/IL-23/IL-36 mRNAs (mRNA-2752) results in regression of DCIS characterized by lymphocytic infiltrates

Abstract Background: High-grade DCIS with immune infiltrates may represent lesions that are able to be kept in check by the immune system, but still are at risk for progression to invasive disease. We and others have demonstrated that the presence of T cells, and in particular the spatial proximity of CD8+/PD-1+ T cells and PD-L1+ cells predicts response to chemotherapy and PD-1 inhibitors in the setting of invasive triple negative breast cancer. DCIS with high-risk features (e.g. large, palpable, high grade, HR-, or HER2+) often have T cell infiltrates. We hypothesized that it might be possible to potentiate the immune response in high-risk DCIS with immune checkpoint blockade. Since mortality for DCIS is extremely low, we proposed an intralesional pembrolizumab (pembro) treatment approach to avoid systemic adverse effects. In a phase 1 dose escalation study of single agent pembro we found 2 doses of 8 mg, administered intralesionally 2-3 wks apart, was tolerable, induced immunological changes within the DCIS lesions, but had little clinical benefit. Herein, we expanded the number of injections to 4 and also tested a combination with mRNA-2752 (Moderna), an mRNA-based therapeutic encoding a T cell co-stimulator OX40L, and pro-inflammatory cytokines IL-23 and IL-36 gamma, to determine if we could find a dose that was both tolerable and elicited anti-tumor responses. Methods Women eligible for this study had DCIS with at least 2 of the following high-risk features: age< 45; high grade, extensive comedonecrosis; palpable mass; hormone receptor negative [HR-]; HER2+; size >5 cm, or microinvasion. A dose expansion cohort was performed using 4 doses of 8 mg pembro, 2-3 wks apart (n=5). We then combined pembro with mRNA-2752 (n=8 cases), initially at 8 and 4 mg, respectively. Dose reductions were implemented based on AEs. Patients received an MRI before and after 2 injections, spaced 2-3 wks apart. A total of 4 injections were allowed. Core biopsy or surgery was conducted after the last MRI. Results We observed an increase in T cell density in the dose expansion cohort (pembro only), except when there was a paucity of T cells in the pre-treatment biopsy. However, only 1 patient in this cohort demonstrated a reduction in lesion size (clinically and on MRI). As of 22SEP22, 8 patients have received the combination of pembro and mRNA-2752; 7 are evaluable. Two patients with minimal T cell infiltrates at baseline (both HR+) failed to respond, based on imaging or pathology. 4/5 patients with moderate to high T cell infiltrates at baseline had partial (2) or complete responses (2) based on imaging (surgery pending), one of whom had absence of DCIS on post-therapy core biopsy and a clear MRI 4 months later. Correlative studies (immune cell densities, spatial proximities) will be presented at the meeting along with complete data (including post-treatment pathology) on the first 8 patients. Side effects were independent of response and included Gr1/2 fever, myalgias, and fatigue starting within 12 hours of injection (3-4 days), enlargement of regional nodes by day 2, and Gr 1/2 erythema and induration of the breast starting at 4 days and lasting 4-20 days. Earlier onset and persistence of symptoms with subsequent exposure required dose reductions for the majority of patients to maintain tolerability (avoid fevers over 39.4°C, intense erythema and swelling of the breast lasting > 4 days). No AEs were higher than grade 2. Conclusion: The combination of intralesional pembro and mRNA-2752 demonstrated modulation of the tumor immune microenvironment and robust antitumor activity in high-risk DCIS (typically HR- or HER2+) with existing T cell infiltrates. This is an ongoing study; the optimal Phase 2 dose for the combination will be based on the totality of evolving safety and translational data. Citation Format: Laura J. Esserman, Alexa Glencer, Kirithiga Ramalingam, Christopher J. Schwartz, Alexander Borowsky, Gillian L. Hirst, Rachel Woody, Nicole Schindler, Nola M. Hylton, Michael Campbell. Intralesional injection of anti-PD-1 (pembrolizumab) and OX40L/IL-23/IL-36 mRNAs (mRNA-2752) results in regression of DCIS characterized by lymphocytic infiltrates [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr OT1-20-02.