2180P - Detrimental effect of an early exposure to antibiotics on the outcomes of immunotherapy in a multi-tumor cohort of patients

Background

The modulation of microbiome by antibiotics (Abs) seems to interfere with the efficacy of immune checkpoint inhibitors (ICI). Previous data suggest that this effect is stronger when the exposure to Abs occurs around the beginning of ICI, although this has not been evaluated in large multi-tumor cohorts.

Methods

Retrospective study of patients (pts) with advanced solid tumors who were treated with ICI alone (2016-2022) at a tertiary university hospital. For each pt, the exposure to Abs was registered during several periods of time: from 60 days before to 42 days after 1^st cycle (C1) (-60D-D42), from D42 to 6 months (D42-6m) and after 6 months from C1 (>6m). We evaluated its association with the objective response rate (ORR), the disease control rate (DCR), progression-free survival (PFS) and overall survival (OS).

Results

475 pts were evaluated. The most frequent histologic subtypes were NSCLC (33.9%), urothelial (17.3%), renal (13.7%), melanoma (11.2%) and head and neck (11.0%). 84.8% of pts received anti-PD1/PDL1, 13.3% a combination of anti-PD1 plus anti-CTLA4, and 1.9% anti-CTLA4 monotherapy (42.1% in 1^st line, 43.1% in 2^nd line, 14.8% in further lines). Performance status (PS) prior to C1 was 0-1 in 84.6%. Median age was 67.5 years. 66.5% of pts received Abs during treatment with ICI (25.7%: 1 cycle; 16.0%: 2 cycles; 11.0%: 3 cycles; 13.8% ≥4 cycles). The most frequently used Abs were beta-lactams (53.8%), quinolones (35.9%), carbapenems (10.9%) and macrolides (10.9%). The early exposure to Abs (-60D-D42) was associated with a lower ORR (27.4% vs 39.4%; p<0.01), a lower DCR (37.3% vs 57.4%; p<0.001), lower PFS (16.8 m vs 27.8 m; HR 0.66 [95% CI 0.53-0.82; p<0.001]) and lower OS (2.5 m vs 6.6 m; HR 0.68 [95% CI 0.55-0.85; p=0.001]). The negative impact of Abs was confirmed by a multivariate analysis including PS, type of tumor, disease burden, type of ICI and line of treatment (p=0.01). This effect was not significant for pts receiving Abs in D42-6m and >6m.

Conclusions

Our results confirm the detrimental impact of the exposure to Abs on the outcomes of immunotherapy. This effect seems time-dependent, prevailing in pts who receive Abs around the initiation of ICI.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

P. Garrido Lopez: Financial Interests, Institutional, Advisory Board: AstraZeneca, Bristol, Janssen, Takeda, Lilly, Gilead, Pfizer, MSD, Roche, Novartis, Rovi. P. Gajate Borau: Financial Interests, Institutional, Advisory Board: Astellas, Bristol Myers, Ipsen, Merck, Ipsen, Pfizer, Roche. All other authors have declared no conflicts of interest.