Impact of Antibiotic Exposure Before Immune Checkpoint Inhibitor Treatment on Overall Survival in Older Adults With Cancer

A Population-Based Study

Lawson Eng, MD, SM; Rinku Sutradhar, PhD; Yue Niu, MSc; Ning Liu, PhD; Ying Liu, MSc; Yosuf Kaliwal, MSc; Melanie L. Powis, MSc; Geoffrey Liu, MD, MSc; Jeffrey M. Peppercorn, MD, MPH; Philippe L. Bedard, MD; Monika K. Krzyzanowska, MD, MPH

Disclosures

J Clin Oncol. 2023;41(17):3122-3134.

Abstract and Introduction

Abstract

Purpose: Antibiotic exposure before immune checkpoint inhibitor (ICI) treatment can negatively affect outcomes through alteration in the gut microbiome, but large-scale evaluations are lacking. We performed a population-level retrospective cohort study to evaluate the impact of antibiotic exposure before starting ICI on overall survival (OS).

Patient and Methods: Patients with cancer, age 65 years or older, who initiated treatment with ICIs between June 2012 and October 2018 in Ontario, Canada, were identified using systemic therapy administration data. The cohort was deterministically linked to other health care databases to obtain covariates and antibiotic prescription claim data at both 1 year and 60 days before ICI therapy. Multivariable Cox models evaluated the association between exposure and OS.

Results: Among the 2,737 patients with cancer who received ICIs, 59% and 19% of patients received antibiotics 1 year and 60 days before ICI therapy, respectively. Median OS was 306 days. Any antibiotic exposure within 1 year before ICI was associated with worse OS (adjusted hazard ratio [aHR], 1.12; 95% CI, 1.12 to 1.23; P = .03). In antibiotic class analysis, exposure to fluoroquinolones within 1 year (aHR, 1.26; 95% CI, 1.13 to 1.40; P < .001) or 60 days before ICI (aHR, 1.20; 95% CI, 0.99 to 1.45; P = .06) was associated with worse OS, with a dose effect seen on the basis of total weeks of exposure over 1 year (aHR, 1.07 per week; 95% CI, 1.03 to 1.11; P < .001) and 60 days (aHR, 1.12 per week; 95% CI, 1.03 to 1.23; P = .01).

Conclusion: In this population-level study, exposure to antibiotics and specifically fluoroquinolones before ICI therapy was observed to be associated with worse OS among older adults with cancer. Interventions aimed at altering the gut microbiome to boost immunogenicity may help improve outcomes for patients receiving ICIs with prior antibiotic exposure.

Introduction

Immune checkpoint inhibitors (ICIs) are important components in the clinical armamentarium against many solid tumors. In the past 10 years, ICIs have been approved by the US Food and Drug Administration (FDA) in many sites including melanoma,^[1,2] lung,^[3–6] genitourinary,^[7–13] head and neck,^[14] colorectal,^[15] and liver cancers,^[16] reflecting a paradigm shift in cancer therapy toward less cytotoxic chemotherapy.

Biomarkers for ICI response and resistance include PD-L1 expression, tumor mutational burden (TMB),^[17,18] tumor immune infiltrate, and other environmental factors.^[19] One factor that is gaining interest is the gut microbiome; preclinical studies have shown that stool diversity and specific species of bacteria in the gut can have a direct impact on ICI effectiveness and immune-related toxicities.^[20–23] As antibiotic exposure can alter the gut microbiome,^[24] there has been interest in evaluating the impact of antibiotic exposure on ICI outcomes.

Prior studies, including two systematic reviews and meta-analyses, suggest that antibiotic exposure may negatively affect immunotherapy outcomes including overall survival (OS).^[18,25,26] However, many individual studies included in the meta-analyses were small (sample size < 200), single center, and featured heterogeneity in antibiotic exposure period before ICI, ICI classes, and tumor types. Evaluation of antibiotic exposure using a larger population-based cohort across multiple centers may allow for the evaluation of specific subgroups on the basis of ICI type and antibiotic subclass.

Here, we performed a population-level retrospective cohort study to evaluate the impact of antibiotic exposure before starting ICI on OS among patients receiving ICIs. Our specific aims were to (1) evaluate the association of prior antibiotic exposure on OS for patients receiving their first ICI, and (2) identify any specific associations on the basis of ICI drug, tumor type, or antibiotic class.

1 2 3 4

Next Section

*Relevance section written by JCO Associate Editor Stephanie B. Wheeler, PhD, MPH.

Table 1. Baseline Characteristics of at the Start of ICI Therapy
Variable	Subgroup	Frequency, No. (%), N = 2,737
Sociodemographic variables
Sex	Male	1,654 (60)
Age at the start of immunotherapy	Years, median (IQR)	73 (69–78)
Time from cancer diagnosis to start of immunotherapy	Days, median (IQR)	357 (147–861)
Income quintile	1	541 (20)
	2	547 (20)
	3	534 (20)
	4	541 (20)
	5	570 (21)
Immigrant status	Immigrant	148 (5)
Rurality	Lives in rural region	391 (14)
BMI	Median (range)	26.1 (15–56)
	Low (< 18.5)	84 (4)
	Normal (18.5–25)	828 (37)
	Overweight (25–30)	806 (23)
	Obese (> 30)	523 (23)
LHIN	1	153 (6)
	2	255 (9)
	3	116 (4)
	4	336 (12)
	5	101 (4)
	6	159 (6)
	7	204 (7)
	8	13 (11)
	9	359 (13)
	10	121 (4)
	11	245 (9)
	12	149 (5)
	13	158 (6)
	14	67 (2)
Johns Hopkins ACG score	0–5	143 (5)
	6–10	1,121 (41)
	11+	1,473 (54)
History of autoimmune disease	Yes	231 (8)
Recent hospitalization	Within 1 year of ICI start	1,296 (47)
Recent hospitalization	Within 60 days of ICI start	544 (20)
Cancer- and treatment-specific information
Facility level of cancer center administering ICI	Level 1/2 cancer care facility	2,118 (77)
Receiving immunotherapy as part of a trial	Yes	85 (3)
Disease site	Lung	1,440 (53)
	Melanoma	926 (34)
	Head and neck	46 (2)
	Bladder	79 (3)
	Renal	201 (7)
	Other	45 (2)
ICI	Atezolizumab	22 (1)
	Avelumab	< 6
	Durvalumab	38 (1)
	Ipilimumab	362 (13)
	Nivolumab + ipilimumab	12 (0.4)
	Nivolumab	1,168 (43)
	Pembrolizumab	1,129 (41)
	Ipilimumab + pembrolizumab	< 6

Abbreviations: ACG, Adjusted Clinical Groups; BMI, body mass index; ICI, immune checkpoint inhibitor; LHIN, local health integration network.

Table 2. Summary of Antibiotic Exposure in Patients Before Starting ICI Therapy
Variable	Subgroup	Frequency, No. (%), n = 2,737
Variable	Subgroup	Any Time 1 Year Before ICI Start	Within 60 Days Before ICI Start
Receiving any antibiotics	Yes	1,627 (59)	532 (19)
Total duration of antibiotics	Weeks, median (IQR)	2 (1–3.4)	1.3 (1–2)
Total doses of antibiotics	Doses, median (IQR)	30 (18–60)	21 (14–36)
Receiving any penicillin	Yes	805 (29)	197 (7)
Total duration of penicillin	Weeks, median (IQR)	1.4 (1–2)	1 (1–1.4)
Total doses of penicillin	Doses, median (IQR)	24 (20–40)	21 (15–30)
Receiving any cephalosporin	Yes	647 (24)	144 (5)
Total duration of cephalosporin	Weeks, median (IQR)	1.4 (1–2)	1 (1–2)
Total doses of cephalosporin	Doses, median (IQR)	28 (20–40)	28 (20–40)
Receiving any fluoroquinolone	Yes	771 (28)	188 (7)
Total duration of fluoroquinolone	Weeks, median (IQR)	1.4 (1–2)	1 (1–1.4)
Total doses of fluoroquinolone	Doses, median (IQR)	14 (7–22)	10 (7–10)

Abbreviation: ICI, immune checkpoint inhibitor.

Table 3. Impact of Antibiotic Exposure Before Starting Immunotherapy on Overall Survival After Starting Immunotherapy
Variable	Comparison	Univariable Analysis		Multivariable Analysis
Variable	Comparison	HR (95% CI)	P	aHR (95% CI)	P
Any antibiotic exposure
Exposed up to 1 year before ICI	Received v not received	1.19 (1.09 to 1.30)	< .001	1.12 (1.01 to 1.24)	.03
Exposed within 60 days of ICI start	Received v not received	1.17 (1.05 to 1.30)	.004	1.06 (0.94 to 1.20)	.35
Total weeks over 1 year	Per 1-week increase	1.02 (1.01 to 1.03)	.0005	1.01 (1.00 to 1.02)	.10
Total weeks within 60 days of starting ICI	Per 1-week increase	1.05 (1.01 to 1.09)	.007	1.02 (0.98 to 1.06)	.39
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.01	1.00 (1.00 to 1.00)	.64
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (0.99 to 1.00)	.26	1.00 (1.00 to 1.00)	.59
Penicillin exposure
Exposed up to 1 year before ICI	Received v not received	1.14 (1.04 to 1.25)	.007	1.05 (0.95 to 1.17)	.35
Exposed within 60 days of ICI start	Received v not received	1.13 (0.96 to 1.33)	.16	1.01 (0.84 to 1.22)	.89
Total weeks over 1 year	Per 1-week increase	1.01 (0.99 to 1.03)	.28	1.00 (0.98 to 1.02)	1.00
Total weeks within 60 days of starting ICI	Per 1-week increase	1.02 (0.95 to 1.09)	.63	0.97 (0.90 to 1.06)	.53
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.16	1.00 (0.99 to 1.00)	.82
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	1.00	1.00 (0.99 to 1.00)	.27
Cephalosporin exposure
Exposed up to 1 year before ICI	Received v not received	1.10 (0.99 to 1.22)	.07	1.03 (0.92 to 1.16)	.63
Exposed within 60 days of ICI start	Received v not received	1.09 (0.90 to 1.33)	.36	0.97 (0.78 to 1.21)	.77
Total weeks over 1 year	Per 1-week increase	1.00 (0.98 to 1.03)	.74	1.00 (0.97 to 1.02)	.73
Total weeks within 60 days of starting ICI	Per 1-week increase	0.99 (0.88 to 1.12)	.91	0.93 (0.90 to 1.08)	.34
Total doses over 1 year	Per 1-dose increase	1.00 (0.99 to 1.00)	.87	1.00 (1.00 to 1.00)	.70
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.57	1.00 (1.00 to 1.01)	1.00
Fluoroquinolone exposure
Exposed up to 1 year before ICI	Received v not received	1.36 (1.23 to 1.49)	< .001	1.26 (1.13 to 1.40)	< .001
Exposed within 60 days of ICI start	Received v not received	1.33 (1.13 to 1.57)	< .001	1.20 (0.99 to 1.45)	.06
Total weeks over 1 year	Per 1-week increase	1.06 (1.03 to 1.08)	< .001	1.07 (1.03 to 1.11)	.0002
Total weeks within 60 days of starting ICI	Per 1-week increase	1.14 (1.07 to 1.22)	< .001	1.12 (1.03 to 1.23)	.01
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.01)	< .001	1.01 (1.00 to 1.01)	.003
Total doses within 60 days of starting ICI	Per 1-dose increase	1.01 (1.00 to 1.01)	.03	1.01 (1.00 to 1.01)	.20

NOTE. The multivariable analysis results for each of these models used to evaluate the association of each antibiotic variable were adjusted for age, sex, BMI, facility level of cancer center administering treatment, autoimmunity history, and the John Hopkins ACG score and hospitalization within the last year.
Abbreviations: ACG, adjusted clinical groups; aHR, adjusted hazard ratio; ICI, immune checkpoint inhibitor.

Table 4. Impact of Antibiotic Exposure Before Starting Immunotherapy on Overall Survival After Starting Immunotherapy by ICI Class
Variable	Comparison	Nivolumab or Pembrolizumab (n = 2,297; anti–PD-1)		Ipilimumab (n = 362; anti–CTLA-4)
Variable	Comparison	aHR (95% CI)	P	aHR (95% CI)	P
Any antibiotic exposure
Exposed up to 1 year before ICI	Received v not received	1.14 (1.02 to 1.27)	.02	1.04 (0.78 to 1.38)	.80
Exposed within 60 days of ICI start	Received v not received	1.06 (0.93 to 1.21)	.35	1.11 (0.74 to 1.67)	.60
Total weeks over 1 year	Per 1-week increase	1.01 (1.00 to 1.02)	.007	0.92 (0.86 to 0.99)	.03
Total weeks within 60 days of starting ICI	Per 1-week increase	1.04 (1.00 to 1.09)	.06	0.90 (0.76 to 1.05)	.18
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.26	1.00 (0.99 to 1.00)	.06
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.96	1.00 (0.99 to 1.00)	.30
Penicillin exposure
Exposed up to 1 year before ICI	Received v not received	1.06 (0.95 to 1.19)	.31	1.01 (0.72 to 1.42)	.96
Exposed within 60 days of ICI start	Received v not received	1.04 (0.85 to 1.27)	.70	0.93 (0.51 to 1.69)	.82
Total weeks over 1 year	Per 1-week increase	1.01 (0.99 to 1.03)	.36	0.92 (0.78 to 1.09)	.31
Total weeks within 60 days of starting ICI	Per 1-week increase	1.02 (0.93 to 1.12)	.74	0.84 (0.61 to 1.17)	.31
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.82	1.00 (0.99 to 1.00)	.31
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.48	0.99 (0.98 to 1.01)	.39
Cephalosporin exposure
Exposed up to 1 year before ICI	Received v not received	1.05 (0.93 to 1.19)	.46	0.94 (0.69 to 1.28)	.69
Exposed within 60 days of ICI start	Received v not received	0.93 (0.73 to 1.19)	.56	1.24 (0.72 to 2.13)	.44
Total weeks over 1 year	Per 1-week increase	1.00 (0.97 to 1.03)	.81	0.96 (0.83 to 1.11)	.58
Total weeks within 60 days of starting ICI	Per 1-week increase	0.91 (0.77 to 1.07)	.24	1.12 (0.77 to 1.63)	.55
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.75	1.00 (0.99 to 1.00)	.71
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (0.99 to 1.01)	.79	1.01 (0.99 to 1.02)	.31
Fluoroquinolone exposure
Exposed up to 1 year before ICI	Received v not received	1.28 (1.15 to 1.44)	< .001	1.00 (0.70 to 1.44)	.98
Exposed within 60 days of ICI start	Received v not received	1.19 (0.98 to 1.44)	.08	1.34 (0.61 to 2.93)	.47
Total weeks over 1 year	Per 1-week increase	1.08 (1.04 to 1.11)	< .001	0.98 (0.82 to 1.18)	.86
Total weeks within 60 days of starting ICI	Per 1-week increase	1.12 (1.02 to 1.23)	.02	1.17 (0.67 to 2.03)	.58
Total doses over 1 year	Per 1-dose increase	1.01 (1.00 to 1.01)	.001	0.99 (0.98 to 1.01)	.42
Total doses within 60 days of starting ICI	Per 1-dose increase	1.01 (1.00 to 1.01)	.23	1.01 (0.97 to 1.05)	.66

NOTE. The multivariable analysis results for each of these models used to evaluate the association of each antibiotic variable were adjusted for age, sex, BMI, facility level of cancer center administering treatment, autoimmunity history, and the John Hopkins ACG score.
Abbreviations: ACG, adjusted clinical groups; aHR, adjusted hazard ratio; ICI, immune checkpoint inhibitor.

Table A1. Summary of Prescription Claims Before Starting Immunotherapy
Antibiotic Name	Any Time 1 Year Before Starting ICI (n = 3,839) No. (%)	Within 60 Days Before Starting ICI (n = 695) No. (%)
Amoxicillin 1 clavulanic acid	519 (14)	106 (15)
Amoxicillin	559 (15)	94 (14)
Ampicillin	< 6	< 6
Cefadroxil	6 (0.2)	< 6
Cefixime	20 (1)	< 6
Cefprozil	52 (1)	11 (2)
Ceftriaxone	< 6	< 6
Cefuroxime	163 (4)	28 (4)
Cephalexin	665 (17)	113 (16)
Ciprofloxacin	518 (13)	79 (11)
Cloxacillin	20 (1)	< 6
Dapsone	< 6	< 6
Fosfomycin	29 (1)	6 (1)
Levofloxacin	350 (9)	73 (11)
Metronidazole	132 (3)	24 (3)
Moxifloxacin	299 (8)	60 (9)
Nitrofurantoin	237 (6)	32 (5)
Norfloxacin	23 (1)	< 6
Penicillin	23 (1)	6 (1)
Sulfamethoxazole + trimethoprim	213 (6)	46 (7)
Trimethoprim	< 6	< 6

NOTE. This table summarizes all the antibiotic prescription claims filled by patients both 1 year and 60 days before starting ICI therapy for our patient cohort.
Abbreviation: ICI, immune checkpoint inhibitor.

Table A2. Univariate and Multivariate Association Analysis of Covariates on Overall Survival After Starting ICI Therapy
Variable	Comparison	Univariable Analysis		Multivariable Analysis
Variable	Comparison	HR (95% CI)	P	aHR (95% CI)	P
Sex	Female v male	1.02 (0.93 to 1.13)	.68	0.98 (0.88 to 1.08)	.64
Age	Per 1-year increase	1.00 (1.00 to 1.01)	.29	1.00 (0.99 to 1.01)	.67
Time from diagnosis to ICI	Per 1-day increase	1.00 (1.00 to 1.00)	.03
Income quintile	2 v 1	0.96 (0.84 to 1.10)	.57
	3 v 1	0.90 (0.78 to 1.04)	.14
	4 v 1	1.02 (0.89 to 1.16)	.83
	5 v 1	0.93 (0.81 to 1.06)	.26
Immigrant status	Yes v no	0.89 (0.73 to 1.08)	.22
Rurality	Yes v no	1.04 (0.92 to 1.18)	.52
BMI	Low v normal	1.48 (1.16 to 1.89)	.001	1.52 (1.19 to 1.94)	.0008
	Overweight v normal	0.83 (0.74 to 0.92)	<.001	0.85 (0.76 to 0.96)	.006
	Obese v normal	0.73 (0.64 to 0.83)	<.001	0.76 (0.67 to 0.86)	<.001
Recent hospitalization	Within 1 year v none	1.47 (1.35 to 1.60)	<.001	1.38 (1.25 to 1.53)	<.001
John Hopkins ACG	0–5 v 6–10	0.94 (0.76 to 1.15)	.53	0.96 (0.76 to 1.21)	.38
John Hopkins ACG	111 v 6–10	1.36 (1.24 to 1.49)	<.001	1.18 (1.06 to 1.31)	<.001
Autoimmune history	Yes v no	1.20 (1.03 to 1.40)	.02	1.25 (1.06 to 1.47)	.0097
Cancer center facility level	Level 1/2 v level 3/4	0.77 (0.69 to 0.85)	<.001	0.78 (0.70 to 0.88)	<.001

NOTE. This table summarizes the results of our univariate and multivariate Cox proportional hazard models of our covariates on overall survival in our patient cohort. Although age and sex were not significantly associated in univariate analysis, we included them in the final multivariate model as prognostic variables.
Abbreviations: ACG, Adjusted Clinical Groups; aHR, adjusted hazard ratio; BMI, body mass index; HR, hazard ratio; ICI, immune checkpoint inhibitor.

Table A3. Impact of Antibiotic Exposure 1 Year to 60 Days Before Starting Immunotherapy on Overall Survival After Starting Immunotherapy
Variable	Comparison	Univariable Analysis		Multivariable Analysis
Variable	Comparison	HR (95% CI)	P	aHR (95% CI)	P
Any antibiotic exposure
Exposed from 1 year to 60 days before starting ICI	Received v not received	1.16 (1.05 to 1.28)	.003	1.11 (1.00 to 1.24)	.06
Total weeks from 1 year to 60 days before starting ICI	Per 1-week increase	1.04 (1.02 to 1.06)	.001	1.02 (0.99 to 1.05)	.12
Total doses from 1 year to 60 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.006	1.00 (1.00 to 1.00)	.23
Penicillin exposure
Exposed from 1 year to 60 days before starting ICI	Received v not received	1.12 (1.01 to 1.25)	.03	1.05 (0.93 to 1.18)	.41
Total weeks from 1 year to 60 days before starting ICI	Per 1-week increase	1.06 (1.01 to 1.14)	.01	1.03 (0.98 to 1.08)	.32
Total doses from 1 year to 60 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.003	1.00 (1.00 to 1.00)	.17
Cephalosporin exposure
Exposed from 1 year to 60 days before starting ICI	Received v not received	1.10 (0.99 to 1.23)	.09	1.04 (0.92 to 1.18)	.50
Total weeks from 1 year to 60 days before starting ICI	Per 1-week increase	1.02 (0.97 to 1.08)	.47	0.99 (0.93 to 1.06)	.81
Total doses from 1 year to 60 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.97	1.00 (1.00 to 1.00)	.56
Fluoroquinolone exposure
Exposed from 1 year to 60 days before starting ICI	Received v not received	1.34 (1.21 to 1.49)	<.0001	1.25 (1.11 to 1.41)	.0002
Total weeks from 1 year to 60 days before starting ICI	Per 1-week increase	1.04 (1.02 to 1.07)	.003	1.06 (1.01 to 1.11)	.02
Total doses from 1 year to 60 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.01)	.01	1.01 (1.00 to 1.01)	.01

NOTE. This sensitivity analysis excluded patients exposed to antibiotics (or specific antibiotic classes) within 60 days of starting immunotherapy to evaluate the association of antibiotic exposure between 1 year to 60 days before starting immunotherapy on overall survival. The multivariable analysis results for each of these models used to evaluate the association of each antibiotic variable were adjusted for age, sex, BMI, facility level of cancer center administering treatment, autoimmunity history, and the John Hopkins ACG score and hospitalization within the last year.
Abbreviations: ACG, adjusted clinical groups; aHR, adjusted hazard ratio; BMI, body mass index; HR, hazard ratio; ICI, immune checkpoint inhibitor.

Table A4. Impact of Antibiotic Exposure Within 30 Days Before Starting Immunotherapy on Overall Survival After Starting Immunotherapy
Variable	Comparison	Univariable Analysis		Multivariable Analysis
Variable	Comparison	HR (95% CI)	P	aHR (95% CI)	P
Any antibiotic exposure
Exposed within 30 days before starting ICI	Received v not received	1.19 (1.04 to 1.36)	.009	1.12 (0.97 to 1.31)	.13
Total weeks within 30 days before starting ICI	Per 1-week increase	1.07 (1.01 to 1.13)	.01	1.04 (0.97 to 1.10)	.28
Total doses within 30 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.10	1.00 (1.00 to 1.01)	.41
Penicillin exposure
Exposed within 30 days before starting ICI	Received v not received	1.09 (0.86 to 1.38)	.49	0.96 (0.73 to 1.25)	.74
Total weeks within 30 days before starting ICI	Per 1-week increase	1.10 (0.98 to 1.23)	.11	0.99 (0.84 to 1.15)	.84
Total doses within 30 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.01)	.34	1.00 (0.99 to 1.01)	.93
Cephalosporin exposure
Exposed within 30 days before starting ICI	Received v not received	1.11 (0.88 to 1.40)	.39	1.00 (0.77 to 1.31)	.98
Total weeks within 30 days before starting ICI	Per 1-week increase	1.05 (0.89 to 1.23)	.60	0.98 (0.81 to 1.19)	.84
Total doses within 30 days before starting ICI	Per 1-dose increase	1.00 (1.00 to 1.01)	.17	1.00 (1.00 to 1.01)	.27
Fluoroquinolone exposure
Exposed within 30 days before starting ICI	Received v not received	1.35 (1.09 to 1.67)	.007	1.31 (1.03 to 1.67)	.03
Total weeks within 30 days before starting ICI	Per 1-week increase	1.11 (1.00 to 1.24)	.06	1.09 (0.96 to 1.23)	.17
Total doses within 30 days before starting ICI	Per 1-dose increase	1.01 (1.00 to 1.01)	.24	1.00 (0.99 to 1.02)	.39

Table A5. Impact of Antibiotic Exposure Before Starting Immunotherapy on Overall Survival After Starting Immunotherapy by Disease Site
Variable	Comparison	Lung (n = 1,140)		Melanoma (n = 926)		Renal (n = 201)
Variable	Comparison	aHR (95%CI)	P	aHR (95%CI)	P	aHR (95%CI)	P
Any antibiotic exposure
Exposed up to 1 year before ICI	Received v not received	1.07 (0.93 to 1.23)	.33	1.06 (0.89 to 1.27)	.49	0.91 (0.65 to 1.44)	.88
Exposed within 60 days of ICI start	Received v not received	0.97 (0.83 to 1.14)	.74	1.06 (0.85 to 1.34)	.60	1.11 (0.66 to 1.86)	.70
Total weeks over 1 year	Per 1-week increase	1.01 (1.00 to 1.02)	.02	0.97 (0.94.1.01)	.11	1.02 (0.94 to 1.10)	.67
Total weeks within 60 days of starting ICI	Per 1-week increase	1.04 (1.00 to 1.10)	.08	0.94 (0.84 to 1.04)	.21	0.96 (0.76 to 1.20)	.70
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.40	1.00 (1.00 to 1.00)	.06	1.00 (1.00 to 1.00)	.77
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (1.00 to 1.00)	.51	1.00 (0.99 to 1.00)	.06	1.00 (0.98 to 1.01)	.53
Penicillin exposure
Exposed up to 1 year before ICI	Received v not received	1.01 (0.88 to 1.16)	.86	1.01 (0.83 to 1.24)	.90	0.75 (0.47 to 1.18)	.22
Exposed within 60 days of ICI start	Received v not received	0.93 (0.73 to 1.17)	.52	0.95 (0.66 to 1.38)	.79	0.95 (0.40 to 2.26)	.90
Total weeks over 1 year	Per 1-week increase	1.00 (0.98 to 1.03)	.97	0.96 (0.89 to 1.04)	.30	1.07 (0.93 to 1.23)	.34
Total weeks within 60 days of starting ICI	Per 1-week increase	0.97 (0.87 to 1.08)	.58	0.89 (0.73 to 1.09)	.26	0.83 (0.38 to 1.82)	.64
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.51	1.00 (1.00 to 1.00)	.31	1.00 (1.00 to 1.01)	.34
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (0.99 to 1.00)	.08	0.99 (0.99 to 1.00)	.20	0.99 (0.95 to 1.03)	.53
Cephalosporin exposure
Exposed up to 1 year before ICI	Received v not received	1.13 (0.97 to 1.32)	.11	0.89 (0.74 to 1.09)	.26	1.00 (0.60 to 1.64)	.98
Exposed within 60 days of ICI start	Received v not received	0.98 (0.72 to 1.32)	.88	0.84 (0.57 to 1.22)	.35	1.10 (0.43 to 2.76)	.84
Total weeks over 1 year	Per 1-week increase	1.01 (0.99 to 1.03)	.51	0.94 (0.87 to 1.02)	.13	0.87 (0.70 to 1.08)	.20
Total weeks within 60 days of starting ICI	Per 1-week increase	1.02 (0.83 to 1.24)	.87	0.82 (0.63 to 1.06)	.13	0.88 (0.54 to 1.43)	.61
Total doses over 1 year	Per 1-dose increase	1.00 (1.00 to 1.00)	.60	0.98 (0.99 to 1.00)	.10	0.99 (0.99 to 1.00)	.20
Total doses within 60 days of starting ICI	Per 1-dose increase	1.00 (0.99 to 1.01)	.78	0.99 (0.98 to 1.00)	.16	0.99 (0.97 to 1.01)	.51
Fluoroquinolone exposure
Exposed up to 1 year before ICI	Received v not received	1.22 (1.06 to 1.39)	.005	1.15 (0.92 to 1.44)	.22	1.08 (0.67 to 1.75)	.76
Exposed within 60 days of ICI start	Received v not received	1.05 (0.84 to 1.32)	.65	1.66 (1.12 to 2.47)	.01	0.97 (0.41 to 2.33)	.95
Total weeks over 1 year	Per 1-week increase	1.08 (1.03 to 1.12)	.004	1.01 (0.93 to 1.10)	.82	1.05 (0.94 to 1.18)	.39
Total weeks within 60 days of starting ICI	Per 1-week increase	1.20 (1.04 to 1.38)	.01	1.44 (1.10 to 1.90)	.009	0.90 (0.62 to 1.32)	.60
Total doses over 1 year	Per 1-dose increase	1.01 (1.00 to 1.01)	.004	1.00 (0.99 to 1.01)	.88	1.00 (0.99 to 1.01)	.56
Total doses within 60 days of starting ICI	Per 1-dose increase	1.01 (1.00 to 1.03)	.15	1.02 (1.00 to 1.05)	.04	0.99 (0.97 to 1.02)	.59

NOTE. The multivariable analysis results for each of these models used to evaluate the association of each antibiotic variable were adjusted for age, sex, BMI, facility level of cancer centre administering treatment, autoimmunity history, and the John Hopkins ACG score.
Abbreviations: ACG, adjusted clinical groups; aHR, adjusted hazard ratio; BMI, body mass index; HR, hazard ratio; ICI, immune checkpoint inhibitor.

Authors and Disclosures

Lawson Eng, MD, SM^1,2, Rinku Sutradhar, PhD^3,4, Yue Niu, MSc³, Ning Liu, PhD³, Ying Liu, MSc³, Yosuf Kaliwal, MSc³, Melanie L. Powis, MSc¹, Geoffrey Liu, MD, MSc^1,2, Jeffrey M. Peppercorn, MD, MPH⁵, Philippe L. Bedard, MD^1,2 and Monika K. Krzyzanowska, MD, MPH^1–3

¹Division of Medical Oncology and Hematology, Department of Medicine, Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada
²Divison of Medical Oncology, Department of Medicine, University of Toronto, Toronto, Ontario, Canada
³Cancer Research Program, Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
⁴Division of Biostatistics, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
⁵Division of Hematology/Oncology, Massachusetts General Hospital and Harvard Medical School, Boston, MA

Corresponding Author
Lawson Eng, MD, SM, Division of Medical Oncology and Hematology, Department of Medicine, Princess Margaret Cancer Centre, University Health Network, 610 University Ave, Toronto, ON M5G 2M9, Canada; e-mail: Lawson.eng@utoronto.ca.

Authors' Disclosures of Potential Conflicts of Interest
Impact of Antibiotic Exposure Before Immune Checkpoint Inhibitor Treatment on Overall Survival in Older Adults With Cancer: A Population-Based Study
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I 5 Immediate Family Member, Inst 5 My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Geoffrey Liu
Honoraria: Pfizer, novartis, Merck, AstraZeneca, Takeda, AbbVie, Bayer, Bristol Myers Squibb, Roche Canada, Jazz Pharmaceuticals, Amgen
Consulting or Advisory Role: Pfizer, Novartis, AstraZeneca/MedImmune, Takeda, Roche Canada
Speakers' Bureau: AstraZeneca, Takeda, Pfizer
Research Funding: Roche (Inst), AstraZeneca/MedImmune (Inst), Takeda, Boehringer Ingelheim

Jeffrey M. Peppercorn
Employment: GlaxoSmithKline
Stock and Other Ownership Interests: GlaxoSmithKline
Consulting or Advisory Role: Abbott Laboratories
Research Funding: Outcomes4Me

Philippe L. Bedard
Consulting or Advisory Role: BMS (Inst), Pfizer (Inst), Seattle Genetics, Lilly, Amgen, Merck, Gilead Sciences
Research Funding: Bristol Myers Squibb (Inst), Sanofi (Inst), AstraZeneca (Inst), Genentech/Roche (Inst), GlaxoSmithKline (Inst), Novartis (Inst), Nektar (Inst), Merck (Inst), Seattle Genetics (Inst), Immunomedics (Inst), Lilly (Inst), Amgen (Inst), Bicara Therapeutics (Inst), Zymeworks (Inst)

Monika K. Krzyzanowska
Consulting or Advisory Role: Eisai, Lilly, Ipsen, Bayer
Research Funding: Eisai (Inst), Exelixis (Inst), Lilly (Inst)

No other potential conflicts of interest were reported.

Author Contributions
Conception and design: Lawson Eng, Melanie L. Powis, Philippe L. Bedard, Monika K. Krzyzanowska
Financial support: Lawson Eng, Monika K. Krzyzanowska
Administrative support: Monika K. Krzyzanowska
Provision of study materials or patients: Ying Liu, Geoffrey Liu
Collection and assembly of data: Lawson Eng
Data analysis and interpretation: Lawson Eng, Rinku Sutradhar, Yue Niu, Ning Liu, Ying Liu, Yosuf Kaliwal, Geoffrey Liu, Jeffrey M. Peppercorn, Philippe L. Bedard, Monika K. Krzyzanowska
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors