Hyperacute toxicity with combination ipilimumab (ipi) and anti-PD1 immunotherapy.

9545 Background: Combination ipi and nivolumab is approved for advanced melanoma and is in trials across oncology. Toxicity (tox) most often occurs 6-10 wks into treatment. Whether early tox is harder to manage or influences treatment efficacy is unknown. Methods: Consecutive metastatic melanoma patients (pts) who developed hyperacute (HA) tox, defined as grade (G) 2 or higher (2+) tox within 21 d of receiving combination immunotherapy (ipi + PD1), were retrospectively identified from 9 centres. Demographics, disease characteristics, tox and outcome data were examined. Results: 80 pts developed HA tox, at a median (med) 10 d (range 1-21). Pts had med age 55y, 66% were stage IV M1c/d, 49% had elevated LDH, 14% were ECOG 2. 61 (76%) pts were treatment naïve, 9 had received prior BRAF inhibitors, 4 ipi, 2 PD1 antibodies. Most frequent HA tox was colitis (n = 23), rash (n = 17), hepatitis (n = 9), endocrine (n = 9), pneumonitis (n = 6), and neurotoxicity (n = 4). 39% were G2, 54% G3, 8% G4. 48% required treatment beyond oral steroids, including IV steroids (21%), infliximab (18%), and other immunosuppression (9%) including mycophenolate and IVIG. Pts required a med 45 days on > 10mg prednisone equivalent. 83% of HA colitis pts required treatment beyond oral steroids (22% IV steroids, 61% infliximab), with a med 75 d on > 10mg prednisone. 20% pts received further combination therapy, 48% permanently discontinued all immunotherapy. 48% pts developed additional tox (10% had 3+ further tox), and 31% with G3-4 HA tox developed another G3-4 tox without further therapy. The overall response rate (ORR) was 54%, and after a med 11.6 mo follow-up, med PFS was 8.74 mo. Patients with G2 tox had a similar ORR but greater PFS than those with G3-4 tox (65% vs 47%, p = 0.19; 10.2 vs 2.8 mo, p = 0.01). There was no difference in ORR and PFS by type or duration of immunosuppression. Conclusions: Hyperacute toxicities from combination immunotherapy are varied. Many pts require treatment beyond oral steroids and for a prolonged duration, and pts are at risk of further severe tox. Efficacy in such pts appears similar to trial populations, although severe tox may correlate with worse outcomes. The degree and duration of immunosuppression does not appear to influence the efficacy of immunotherapy.