1158P - Association of corticosteroid (CS) exposure with treatment failure in patients (pts) with advanced melanoma treated with immune checkpoint inhibitors (ICIs)

Background

Around 25% of pts with advanced melanoma lose established responses to ICIs. Recent studies suggest that systemic CS, predominantly used for managing immune-related adverse events (IRAEs), could be a cause of acquired resistance to ICIs, with a detrimental effect on survival. The impact of systemic CS on pts with initial response to ICIs remains unknown.

Methods

This is a retrospective study of 113 pts with advanced melanoma treated with ipilimumab and nivolumab (I/N) at The Royal Marsden Hospital between 2013 and 2021 who achieved initial disease stability or disease response. Pts were divided into cohort A (sustained response to I/N, N = 72) and cohort B (loss of response to I/N, N = 41). The daily CS exposure was evaluated from cycle 1 of I/N (prednisolone equivalent doses). STATA was used for statistical analysis.

Results

Most pts had cutaneous melanoma (73.6% in cohort A vs 75.6% in cohort B), stage IV disease (91.7% vs 95.1%) and were treatment-naïve (86.1% vs 85.4%). More pts in cohort B had a BRAF mutation (46.3% vs 38.9% in cohort A), high LDH (29.3% vs 19.4%), brain (24.4% vs 20.8%) and liver metastases (26.8% vs 18.1%) at baseline. The median number of I/N cycles was 4 in cohort A and 3 in cohort B. Up to 27 pts in cohort A (38%) and 24 in cohort B (59%) discontinued treatment during the induction phase due to toxicity. All pts in our study were given CS for IRAEs management. Up to 97.6% of pts in cohort B received CS, compared to 69.4% in cohort A. Pts who were treated with CS had a significantly higher rate of treatment failure (44.4% vs 4.4%, p < 0.001). The average cumulative CS dose per patient in cohort A was 4830 mg compared to 4201 mg in cohort B (p = 0.897). The average daily CS dose in cohort A was 32 mg/day compared to 30 mg/day in cohort B (p = 0.889). Pts unexposed to CS have a trend for longer overall survival (NR vs NR, HR 0.25 [0.03-1.92]).

Conclusions

Our study suggests a correlation between systemic CS use and higher rates of treatment failure, as well as a trend for shorter survival in pts with advanced melanoma treated with I/N. The dose of CS was similar in both cohorts. However, as this is a single centre retrospective experience, these findings should be confirmed in larger studies.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

H.M.L. Le: Financial Interests, Personal, Sponsor/Funding, Meeting attendance: Pfizer, Novartis. J. Larkin: Financial Interests, Personal, Invited Speaker: BMS, Pfizer, Roche, Pierre Fabre, AstraZeneca, Novartis, EUSA Pharma, MSD, Merck, GSK, Ipsen, Aptitude, Eisai, Calithera, Ultimovacs, Seagen, Goldman Sachs, eCancer, Inselgruppe, Agence Unik; Financial Interests, Personal, Other, Consultancy: Incyte, iOnctura, Apple Tree, Merck, BMS, Eisai, Debiopharm; Financial Interests, Personal, Other, Honorarium: touchIME, touchEXPERTS, VJOncology, RGCP, Cambridge Healthcare Research, Royal College of Physicians; Financial Interests, Institutional, Funding: BMS, MSD, Novartis, Pfizer, Achilles, Roche, Nektar, Covance, Immunocore, Pharmacyclics, Aveo. All other authors have declared no conflicts of interest.