Abstract B23: Autoimmune hypophysitis is a marker of favorable outcome during treatment of melanoma with ipilimumab

Abstract

Abstract Introduction: Autoimmune hypophysitis (AH) is an immune-related adverse effect (ir-AE) of the human IgG1 monoclonal antibody ipilimumab. Evidence suggests AH is mediated by complement deposition in adenohypophyseal cells expressing cytotoxic T lymphocyte antigen-4 (CTLA-4). Despite increasing incidence, AH remains poorly characterized. We conducted a retrospective cohort study of melanoma subjects treated with ipilimumab, with the primary objective to characterize AH and examine its association with overall survival (OS). Methods: Inclusion criteria: stage 3/4 cutaneous, ocular, or mucosal melanoma at a single institution, treated with ipilimumab between 2005 and 2013. AH was defined as primary hypopituitarism clinically diagnosed by endocrinologist or cutaneous oncologist. Baseline BMI and laboratory values were collected within 30 days prior to first dose. Overall survival (OS) was defined as time from first dose to death of any cause. Time-to-progression (TTP) was defined as time from first dose to any objective progression event, and censored if >4 months (mo) elapsed between evaluations to ensure homogenous follow-up time. Statistical tests were non-parametric and 2-sided. Results: Of 295 patients screened, 269 were eligible (32 AH+, 237 AH-). Demographics at first dose: median age 61.6, 35% female, 99.3% white, 1.9% Hispanic, mean body mass index (BMI) 28 kg/m2, mean lactate dehydrogenase (LDH) 644 IU/L, mean albumin 4.0 mg/dL, mean absolute lymphocyte count 1.39 x 103/μL. For TNM stage, 65% M1c, 14% M1b, 6% M1a, 15% M0. At time of first dose, 29.5% were resected no evidence-of-disease (NED). Of 132 (49%) with known allelic status, 38 (29%) were BRAF mutant. Median time-to-AH was 2.8 mo (range 0.9 – 18.7). Initial dose of 10 mg/kg was associated with increased AH risk (RR 1.2, p=0.002). Presenting symptoms included fatigue (58%), headache (33%), nausea (24%). Lab findings at diagnosis (median, range): ACTH 7.0 (1-26), cortisol 2.2 (0-29.7), TSH 0.34 (0-3.9), T4 0.7 (0.3 – 3.6), FSH 7.0 (1.1 – 41), LH 0.7 (0.2 – 20.5). 17 of 30 evaluable subjects had abnormal pituitary identified on MRI. All were administered exogenous corticosteroids and 72% required exogenous thyroid supplementation. 13% eventually recovered endogenous pituitary function. There was no association with prior endocrine conditions or other ir-AEs. For evaluable subjects, AH also seemed to have favorable objective response rate, although this was not statistically significant (p=0.20). Median follow-up duration was 43.7 mo, median OS was 20.3 mo, and median TTP was 5.2 mo. AH was associated with superior OS (HR 0.4, 95% CI 0.3 – 0.7, p=0.001) and TTP (HR 0.4, 95%CI 0.3- 0.8, p=0.0002). Using Landmark analysis to avoid bias from immortal person-time, after 4 cycles of ipilimumab, AH remained associated with OS (HR 0.5, p=0.04). Using the alternative Crowley method, similar results were seen for both OS (HR 0.5, p=0.002) and TTP (HR 0.6, p=0.005). AH was then examined as a time-dependent covariate in Cox regression. On univariate analysis, LDH, M-stage, resected NED, albumin, age, absolute lymphocyte count, and AH were significantly associated with OS (p <0.01 for all). Similar findings were found for TTP. After including a propensity score regression term to control for all known confounders (LDH, albumin, BMI, lymphocyte count, M-stage, NED, smoking, race, age, gender), AH remained an independent predictor for improved OS (HR 0.49;95%CI 0.24-0.98; p=0.04). Conclusion: Hypophysitis appears to predict benefit from ipilimumab treatment, possibly since it may serve as a marker of cellular immune activation. Independent validation of these results in external cohorts is required. Citation Format: Jennifer M. Eatrides, Jeffrey Weber, Kathleen Egan, Marjorie Acierno, Michael Schell, Howard Lillienfeld, Ben Creelan. Autoimmune hypophysitis is a marker of favorable outcome during treatment of melanoma with ipilimumab. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Melanoma: From Biology to Therapy; Sep 20-23, 2014; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(14 Suppl):Abstract nr B23.