5262 - Immune-related adverse events (irAEs) and outcome in recurrent/metastatic (R/M) Head and Neck Squamous Cell Carcinoma (HNSCC) patients (pts) treated by immune-checkpoints inhibitors (ICI)

Background

ICI is a standard of care in R/M HNSCC pts progressing to platinum compounds. However, few pts respond to ICI and irAEs could limit their use.There are contradictory reports regarding the association of irAEs and outcomes to ICI. The aim of this study is to further evaluate the potential impact of irAEs on outcomes for HNSCC.

Methods

We performed a retrospective observational study based on all R/M HNSCC pts treated by ICI in our center during 2014-19. To limit the bias due to the association between ICI administration duration, obtention of objective response (OR) and occurrence of irAE, a Landmark analysis method with 90 days (d) cut-off was used.

Results

We included 103 R/M HNSCC pts (80% male, 73% smokers). Primary tumor locations were: oropharynx (42%), oral cavity (23%), hypopharynx (18%) and larynx (17%). ICI were administered as monotherapy (43%) or combination (57%; CTLA-4 inhibitor, ICOS agonist). There were 14 responders (R) with median time to obtain OR of 102 d (8 after 90 d). Median ICI duration was 376 d in R pts and 84 d in others. irAEs were observed in 38 pts (37%), with low frequency of grade (G) 3-4 (6%). Occurrence of irAEs within the first 90 d (named early AEs) was observed in 25 pts (24%). 51 pts (50%) remained on ICI beyond 90 d, including all the 14 R: 30 pts (59%) had irAEs and 17 (33%) had early irAEs. In total cohort, OR was more frequent in pts with any irAE (32%) than in others (3%). Results were similar with irAE G2-4: 42% OR vs 5% OR. But, using landmark method, in pts who received ≥ 90 d ICI, this association was no longer observed: 35% of pts with early irAEs had OR vs 24% in other pts (p = 0.37), and 36% vs 25% for early irAEs of G2-4 (p = 0.47). Median [95%CI] PFS and OS were 3.6 months (m) [2.9-4.5] and 10.6m [7.6-13.2] in total cohort; 5.8 m [4.8-7.7] and 19.1 m [12.6-32.5] in ICI ≥90 d. Early irAEs were not associated with PFS or OS in this latter subgroup.

Conclusions

The impact of irAEs on outcomes could be mistaken if not taking into consideration the ICI duration because the likelihood of irAEs and OR increases with treatment duration. Using a Landmark analysis with 90 d cut-off, we did not observe an association between irAEs and outcome in R/M HNSCC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

N. Baste Rotllan: Advisory / Consultancy, Travel / Accommodation / Expenses: Merck Serono; Advisory / Consultancy: MSD; Advisory / Consultancy, Travel / Accommodation / Expenses: Bristol Meyers Squibb; Advisory / Consultancy, Travel / Accommodation / Expenses: Nanobiotix; Honoraria (institution), Travel / Accommodation / Expenses: AstraZeneca; Honoraria (institution): PharmaMar. All other authors have declared no conflicts of interest.