Abstract
Objective
To evaluate oncologic outcomes in women with recurrent platinum-sensitive ovarian cancer (OC) receiving antibiotics (ABX) during platinum-based chemotherapy.
Methods
A retrospective, single-institution cohort study was performed in women undergoing platinum chemotherapy for recurrent platinum-sensitive OC from 2009–2017. ABX for >48 hours, including anti-gram-positive antibiotics (G+ ABX), were recorded. The impact of ABX on time to second progression (PFS2), time to platinum resistance, and overall survival (OS) were assessed using univariate and multivariable Cox regression models.
Results
Of 261 women with recurrent platinum-sensitive OC, 80 (30.7%) received ABX during platinum chemotherapy, and 20 (7.7%) received G+ ABX. On univariate analysis for PFS2, there was no difference for ABX versus none (13.1 vs. 12.3 months: hazard ratio [HR]=1.23, 95% confidence interval [CI]=0.93–1.62, p=0.15), but this was decreased for G+ ABX versus none (10.2 vs. 12.3 months: HR=1.71; 95% CI=1.05–2.77; p=0.03). There was no difference in OS for ABX versus none (30.8 vs. 33.5 months: HR=1.01; 95% CI=0.73–1.39; p=0.97), but G+ ABX were associated with decreased OS compared to no ABX (26.4 vs. 33.4 months: HR=2.13; 95% CI=1.28–3.57; p=0.004) and other ABX (26.4 vs. 37.9 months: HR=2.43; 95% CI=1.34–4.41; p=0.003), respectively. On multivariable analysis, no ABX were associated with improved PFS2 (HR=0.54; 95% CI=0.33–0.88; p=0.014) and OS (HR=0.49; 95% CI=0.29–0.81; p=0.006) versus G+ ABX.
Conclusion
This retrospective study of women with recurrent platinum-sensitive OC treatment with G+ ABX during platinum chemotherapy was associated with decreased PFS2 and OS.
Keywords: Ovarian Cancer, Platinum Chemotherapy, Antibiotics, Microbiome
Synopsis
Antibiotics (ABX) have been associated with worse progressive-free survival and overall survival (OS) during platinum chemotherapy for ovarian cancer. This study aimed to assess the impact of ABX in women with recurrent ovarian cancer. Treatment with anti-gram-positive ABX during platinum chemotherapy is associated with reduced time to second progression and OS.
INTRODUCTION
Epithelial ovarian cancer (OC) is the leading cause of gynecologic cancer-related death in the United States [1]. Unfortunately, most women with advanced OC will recur within the first 3 years after diagnosis [2,3,4]. In women with recurrent OC, prognosis, and treatment response depends upon sensitivity to platinum chemotherapy. Unfortunately, women with platinum-resistant OC have fewer therapeutic options, lower treatment efficacy, and a worse prognosis than those with platinum-sensitive disease [5]. Therefore, advancing our understanding of strategies to prevent and combat platinum resistance is essential for transforming the treatment landscape for women with OC.
In recent years, there has been growing evidence supporting the microbiome of the gut and female reproductive tract as a mediator in cancer therapy, including immunotherapy and platinum chemotherapy [6,7,8,9,10,11,12,13,14,15,16,17,18]. Many preclinical studies have utilized antibiotics (ABX) to alter or induce dysbiosis within the gut microbiome and have demonstrated a negative impact on treatment efficacy and oncologic outcomes [6,7,8,9,10,11,12,13,14,15,16,17,18]. We recently assessed the impact of ABX, including those targeting gram-positive bacterial species, upon oncologic outcomes in 424 women with newly diagnosed stage III and IV OC undergoing platinum chemotherapy in the front-line setting [7]. Notably, patients who received ABX during platinum chemotherapy had significantly decreased progression-free survival (PFS) and overall survival (OS). Moreover, ABX targeted against gram-positive bacteria had the most marked adverse impact on PFS and OS, even when controlling for postoperative complications and performance status.
Since platinum chemotherapy remains foundational to treating women with platinum-sensitive OC, resistance mechanisms should be thoroughly considered. Furthermore, strategies to prolong platinum sensitivity in women with recurrent OC are essential to improving outcomes. However, the impact of ABX during platinum chemotherapy in women with platinum-sensitive, recurrent OC is unknown. Therefore, the study objective was to assess the impact of ABX, including those targeting gram-positive bacteria, upon second progression, time to platinum resistance, and OS in women with newly recurrent, platinum-sensitive OC undergoing platinum-based chemotherapy.
MATERIALS AND METHODS
1. Study design
We performed an Institutional Review Board-approved, retrospective cohort study in women treated at the Cleveland Clinic for recurrent, platinum-sensitive OC from January 1st, 2009, to December 31st, 2017. All patients were treated with National Comprehensive Cancer Network preferred platinum-based doublet chemotherapy regimens with or without bevacizumab [19]. For each patient receiving platinum chemotherapy, we recorded antibiotic use (oral and intravenous) during the second-line platinum chemotherapy dates only. This was the first platinum-sensitive recurrence for all patients. Antibiotic treatment was defined as >48 hours of treatment, consistent with prior studies [6,7]. In addition, ABX with primary activity against gram-positive bacteria (vancomycin, linezolid, and daptomycin) were separately noted.
All patients were initially treated at our institution for stage III/IV OC with cytoreductive surgery and chemotherapy with carboplatin and paclitaxel. Following a complete response to platinum chemotherapy, surveillance was conducted according to National Comprehensive Cancer Network guidelines, including surveillance visits and CA-125 levels every three months for the first two years and every three to 6 months for the next 3 years [19]. In addition, computed tomography scans of the abdomen, pelvis, and chest are performed with increasing CA-125 levels or as clinically indicated based on patient symptoms. This analysis included only patients with their first recurrence of platinum-sensitive OC, defined by disease progression at least 6 months from the completion date of front-line platinum chemotherapy.
Patients with OC were extracted from the electronic medical record via International Classification of Diseases, Ninth Revision codes (183.0, 183.2, 183.8, and 183.9). The investigators hand-searched medical records to identify patients with platinum-sensitive, recurrent OC. Only women older than 18 with high-grade histologies were included; borderline/low malignant potential tumors and low-grade serous histology were excluded. Furthermore, those who did not receive surgery and chemotherapy at our institution were excluded.
2. Data collection
Data were extracted by the investigators for patient demographics, including age, body mass index, race, hereditary cancer mutations, and Eastern Co-operative Oncology Group Score (ECOG) at the initiation of chemotherapy for their platinum-sensitive recurrent disease. In addition, data were collected regarding the timing of primary cytoreductive surgery, residual disease, and whether secondary cytoreductive surgery was performed for recurrent disease. Finally, for oncologic characteristics, data were collected regarding the initial stage, histology, time to recurrence, and use of poly(ADP-ribose) polymerase (PARP) inhibitors, bevacizumab, or hyperthermic intraperitoneal chemotherapy (HIPEC). All data were collected with a secure RedCAP database [20].
3. Statistical analysis
For statistical analysis, patients were divided into two groups: those who received ABX versus those who had no ABX. Further, those who received ABX targeting gram-positive bacteria (daptomycin, linezolid, and vancomycin) were compared to patients who did not receive ABX and those who received other ABX. Categorical factors were summarized using frequencies and percentages, while continuous measures summaries were reported as mean and standard deviation or medians and interquartile ranges based on data distribution. Pearson χ2 tests, Fisher’s exact test, 2-sample t-tests, analysis of variance, and Wilcoxon rank sum test were used to evaluate patient and oncologic variables associated with response.
The initial time to the first progression after front-line treatment was calculated from the date of diagnosis to the first recurrence. The primary outcome of time to disease progression after second-line platinum-based chemotherapy was calculated from the date of starting second-line platinum chemotherapy to the date of disease progression. Time to platinum resistance was calculated from the date of starting second-line platinum-based chemotherapy to the date of platinum resistance, defined as progression within six months of platinum chemotherapy. Finally, OS was calculated from the date of starting second-line platinum chemotherapy to the date of death. All survival times considered each month as 30 days and the initial time for survival calculations began at the initiation of second-line platinum chemotherapy. Cox proportional hazards regression right-censored univariate and multivariable models were performed for PFS and OS, log-rank tests, and Wald tests were performed. Cox proportional hazards regression right-censored univariate and multivariable models were performed for time to second progression (PFS2), time to platinum resistance, and OS. Multivariable models were built with ABX type, stage, and residual disease. Stage and residual disease were chosen based on clinical relevance as these factors are known to associate with PFS and OS in advanced and recurrent OC. All p-values are 2-sided, with 0.05 as the level of statistical significance and 95% confidence interval (CI). Statistical analysis was performed using SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA).
RESULTS
1. Patient demographics
In total, 261 women with recurrent, platinum-sensitive OC were included; 80 (30.7%) were treated with ABX, and 20 (7.7%) received anti-gram-positive antibiotics (G+ ABX) (Fig. S1). Indications for ABX include urinary tract infections (n=53, 66.3%), upper respiratory infections/pneumonia (n=12, 15.0%), gastrointestinal infections/abscesses (n=7, 8.7%), dermatologic and/or port infections (n=3, 3.8%), bacteremia and neutropenic fever (n=5, 6.3%). Demographics for patients are demonstrated in Table 1. Patients treated with ABX had significantly lower age (59.2 vs. 63.4 years, p=0.002) and higher body mass index (29.4 vs. 27.2 kg/m2, p=0.019) than no ABX. There were no significant differences in ECOG performance status at chemotherapy (0%–52.5% vs. 55.8%; 1%–37.5% vs. 37.0%, p=0.51) or hereditary genetic mutations including BRCA1 (8.8% vs. 13.8%, p=0.23) and BRCA2 (6.3% vs. 5.5%, p=0.82) for those treated with ABX versus no ABX.
Table 1. Patient characteristics for women with recurrent platinum sensitive advanced epithelial ovarian cancer undergoing platinum chemotherapy with or without concurrent antibiotic therapy.
| Characteristics | No ABX (n=181) | ABX (n=80) | p-value | No ABX (n=181) | G+ ABX (n=20) | Other ABX (n=60) | p-value | ||
|---|---|---|---|---|---|---|---|---|---|
| Demographics | |||||||||
| Age (yr) | 63.4±10.1 | 59.2±10.2 | 0.002* | 63.4±10.1 | 62.1±8.1 | 58.2±10.8 | 0.003* | ||
| BMI (kg/m2) | 27.2±6.2 | 29.4±7.5 | 0.019† | 27.2±6.2 | 31.4±8.9 | 28.8±6.9 | 0.013* | ||
| Race | 0.48∥ | 0.25∥ | |||||||
| Caucasian | 158 (87.3) | 75 (93.8) | 158 (87.3) | 17 (85.0) | 58 (96.7) | ||||
| Black | 17 (9.4) | 4 (5.0) | 17 (9.4) | 3 (15.0) | 1 (1.7) | ||||
| Indian | 3 (1.7) | 0 (0.0) | 3 (1.7) | 0 (0.0) | 0 (0.0) | ||||
| Asian | 3 (1.7) | 1 (1.3) | 3 (1.7) | 0 (0.0) | 1 (1.7) | ||||
| ECOG at chemotherapy | 0.51‡ | 0.21‡ | |||||||
| 0 | 101 (55.8) | 42 (52.5) | 101 (55.8) | 8 (40.0) | 34 (56.7) | ||||
| 1 | 67 (37.0) | 30 (37.5) | 67 (37.0) | 8 (40.0) | 22 (36.7) | ||||
| 2 | 12 (6.6) | 5 (6.3) | 12 (6.6) | 2 (10.0) | 3 (5.0) | ||||
| 3 | 1 (0.6) | 3 (3.8) | 1 (0.6) | 2 (10.0) | 1 (1.7) | ||||
| Genetic mutation | |||||||||
| BRCA1 | 16 (8.8) | 11 (13.8) | 0.23§ | 16 (8.8) | 0 (0.0) | 11 (18.3) | 0.32§ | ||
| BRCA2 | 10 (5.5) | 5 (6.3) | 0.82§ | 10 (5.5) | 1 (5.0) | 4 (6.7) | 0.94§ | ||
| Other (BRIP1, RAD51C/D) | 34 (18.8) | 20 (25.0) | 0.25§ | 34 (18.8) | 2 (10.0) | 18 (30.0) | 0.47∥ | ||
| Medical comorbidities | |||||||||
| HTN | 80 (44.2) | 32 (40.0) | 0.53§ | 80 (44.2) | 8 (40.0) | 24 (40.0) | 0.82§ | ||
| DM | 12 (6.6) | 15 (18.8) | 0.003§ | 12 (6.6) | 3 (15.0) | 12 (20.0) | 0.010§ | ||
| CKD | 7 (3.9) | 5 (6.3) | 0.40§ | 7 (3.9) | 6 (30.0) | 4 (6.7) | 0.67§ | ||
| CAD | 22 (12.2) | 15 (18.8) | 0.16§ | 22 (12.2) | 6 (30.0) | 9 (15.0) | 0.093§ | ||
| VTE | 20 (11.0) | 8 (10.0) | 0.80§ | 20 (11.0) | 2 (10.0) | 6 (10.0) | 0.97§ | ||
| Surgical variables | |||||||||
| Timing of primary CRS | 0.32§ | 0.23§ | |||||||
| Primary CRS | 131 (72.4) | 63 (78.8) | 131 (72.4) | 13 (65.0) | 49 (81.7) | ||||
| Interval CRS | 50 (27.6) | 17 (21.3) | 50 (27.6) | 7 (35.0) | 11 (18.3) | ||||
| Residual disease | 0.33§ | 0.66§ | |||||||
| Optimal (R0) | 71 (39.2) | 39 (48.8) | 71 (39.2) | 10 (50.0) | 29 (48.3) | ||||
| Optimal (<1 cm) | 91 (50.3) | 35 (43.8) | 91 (50.3) | 9 (45.0) | 6 (43.3) | ||||
| Suboptimal | 19 (10.5) | 6 (7.5) | 19 (10.5) | 1 (5.0) | 5 (8.3) | ||||
| Oncologic characteristics | |||||||||
| Stage | 0.85§ | 0.58§ | |||||||
| III | 151 (83.4) | 66 (82.5) | 151 (83.4) | 15 (75.0) | 51 (85.0) | ||||
| IV | 30 (16.6) | 14 (17.5) | 30 (16.6) | 5 (25.0) | 9 (15.0) | ||||
| Serous histology | 174 (96.1) | 77 (96.3) | 0.96§ | 174 (96.1) | 19 (95.0) | 58 (96.7) | 0.70∥ | ||
| IP chemotherapy | 27 (15.0) | 13 (16.5) | 0.77§ | 27 (15.0) | 2 (10.5) | 11 (18.3) | 0.68§ | ||
| Use of bevacizumab | 18 (26.9) | 8 (30.8) | 0.71§ | 18 (26.9) | 7 (38.9) | 1 (12.5) | 0.36§ | ||
| Use of PARP inhibitors | 48 (26.5) | 24 (30.0) | 0.56§ | 48 (26.5) | 6 (30.0) | 18 (30.0) | 0.85§ | ||
| Secondary CRS | 32 (17.7) | 13 (16.3) | 0.78§ | 32 (17.7) | 5 (25.0) | 8 (13.3) | 0.47§ | ||
| Receipt of HIPEC | 21 (11.6) | 6 (7.5) | 0.32§ | 21 (11.6) | 3 (15.0) | 3 (5.0) | 0.27§ | ||
Statistics presented as mean ± standard deviation or number (column %). Bold values denote statistical significance at the p<0.05 level.
ABX, antibiotics; BMI, body mass index; CAD, coronary artery disease; CKD, chronic kidney disease; CRS, cytoreductive surgery; DM, diabetes mellitus; ECOG, Eastern Co-operative Oncology Group; G+ ABX, anti-gram-positive antibiotics; HIPEC, hyperthermic intraperitoneal chemotherapy; HTN, hypertension; IP, intraperitoneal; PARP, poly(ADP-ribose) polymerase; VTE, venous thromboembolism.
*The t-test, †Satterthwaite t-test, ‡Wilcoxon rank sum test, §Pearson’s χ2 test, ∥Fisher’s exact test.
Most patients who received ABX or no ABX had stage III disease at diagnosis (82.5% vs. 83.4%, p=0.85) and serous histology (96.3% vs. 96.1%, p=0.96), respectively. No significant differences were observed between those who received ABX compared to none for the incidence of primary debulking surgery (78.8% vs. 72.4%, p=0.32) and R0 cytoreduction (48.8% vs. 39.2%, p=0.33) at diagnosis. Further, there was no difference in receipt of bevacizumab (30.8% vs. 26.9%, p=0.71), HIPEC (7.5% vs. 11.6%, p=0.32), PARP inhibitors (30.0% vs. 26.5%, p=0.71) or secondary cytoreduction (16.3% vs. 17.7%, p=0.78) for those treated with ABX versus none. There were no differences in the median number of completed cycles (6.0 vs. 6.0, p=0.22), chemotherapy delays (15.0% vs. 16.5%, p=0.77), or dose reductions (52.2 vs. 61.5%, p=0.42) for those treated with ABX versus none.
2. Oncologic outcomes
The median follow-up duration was 50.5 months for G+ ABX, 55.4 months for other ABX, and 51.2 months for no ABX (p=0.68). A total of 260 patients were included in the progression and survival analyses; one patient did not follow up after chemotherapy at our institution and was excluded from the survival analysis. Notably, there was no difference in the time to first progression for patients treated with ABX versus no ABX (19.6 vs. 19.2 months: hazard ratio [HR]=0.88; 95% CI=0.68–1.15; p=0.36). Further, there were no differences in the time to first progression for patients treated with G+ ABX versus no ABX (23.8 vs. 19.2 months: HR=0.68; 95% CI=0.42–1.08; p=0.10) and other ABX versus no ABX (19.0 vs. 19.2 months: HR=0.98; 95% CI=0.73–1.32; p=0.91). This supports that the populations who received ABX during second-line therapy were biologically similar.
Table 2 displays univariate analysis for oncologic outcomes. On univariate analysis for PFS2, there were no significant differences between patients receiving any ABX versus no ABX (13.1 vs. 12.3 months: HR=1.23; 95% CI=0.93–1.62; p=0.15), other ABX versus no ABX (13.4 vs. 12.3 months: HR=1.11; 95% CI=0.82–1.51; p=0.34), and other ABX versus G+ ABX (13.4 vs. 10.2 months: HR=1.55; 95% CI=0.91–2.64; p=0.10). Those patients who received G+ ABX had a significantly reduced PFS2 compared to no ABX (10.2 vs. 12.3 months: HR=1.71; 95% CI=1.05–2.77; p=0.03) (Fig. 1).
Table 2. Univariate analysis for oncologic outcomes in women with recurrent platinum sensitive epithelial ovarian cancer undergoing platinum chemotherapy.
| Variable | No. | Median months (95% CI) | 2-year event-free rate % (95% CI) | Cox univariate HR (95% CI) | p-value | |
|---|---|---|---|---|---|---|
| Treatment start to second progression | ||||||
| No ABX | 180 | 12.3 (11.2–14.5) | 28.7 (21.9,35.5) | --- | 0.15 | |
| Any ABX | 80 | 13.1 (11.0–15.2) | 14.3 (6.5–22.1) | 1.23 (0.93–1.62) | ||
| No ABX | 180 | 12.3 (11.2–14.5) | 28.7 (21.9,35.5) | --- | 0.03 | |
| G+ ABX | 20 | 10.2 (6.6–16.3) | 5.3 (0.0–15.4) | 1.71 (1.05–2.77) | ||
| No ABX | 180 | 12.3 (11.2–14.5) | 28.7 (21.9,35.5) | --- | 0.49 | |
| Other ABX | 60 | 13.4 (11.7–15.3) | 17.1 (7.5–26.8) | 1.11 (0.82–1.51) | ||
| Other ABX | 60 | 13.4 (11.7–15.3) | 17.1 (7.5,26.8) | --- | 0.10 | |
| G+ ABX | 20 | 10.2 (6.6–16.3) | 5.3 (0.0–15.4) | 1.55 (0.91–2.64) | ||
| Time to platinum resistance | ||||||
| No ABX | 180 | 14.9 (11.7–18.7) | 36.2 (29.0–43.4) | --- | 0.69 | |
| Any ABX | 60 | 14.9 (11.9–17.5) | 34.8 (24.2–45.4) | 0.94 (0.70–1.27) | ||
| No ABX | 180 | 14.9 (11.7–18.7) | 36.2 (29.0–43.4) | --- | 0.09 | |
| G+ ABX | 20 | 10.2 (6.6–17.5) | 26.6 (6.6–46.5) | 1.52 (0.93–2.50) | ||
| No ABX | 180 | 14.9 (11.7–18.7) | 36.2 (29.0–43.4) | --- | 0.25 | |
| Other ABX | 60 | 16.0 (12.4–23.8) | 37.3 (24.9,49.8) | 0.82 (0.59–1.15) | ||
| Other ABX | 60 | 16.0 (12.4–23.8) | 37.3 (24.9–49.8) | --- | 0.04 | |
| G+ ABX | 20 | 10.2 (6.6–17.5) | 26.6 (6.6–46.5) | 1.76 (1.02–3.07) | ||
| Treatment start to death | ||||||
| No ABX | 180 | 33.5 (28.8–36.6) | 69.1 (62.2–76.0) | --- | 0.97 | |
| Any ABX | 80 | 30.8 (25.7–40.6) | 64.8 (54.1–75.5) | 1.01 (0.73–1.39) | ||
| No ABX | 180 | 33.5 (28.8–36.6) | 69.1 (62.2–76.0) | --- | 0.004 | |
| G+ ABX | 20 | 26.4 (12.9–28.2) | 56.1 (33.3–79.0) | 2.13 (1.28–3.57) | ||
| No ABX | 180 | 33.5 (28.8–36.6) | 69.1 (62.2–76.0) | --- | 0.26 | |
| Other ABX | 60 | 37.9 (25.7–49.9) | 67.5 (55.4–79.6) | 0.81 (0.56–1.17) | ||
| Other ABX | 60 | 37.9 (25.7–49.9) | 67.5 (55.4–79.6) | --- | 0.003 | |
| G+ ABX | 20 | 26.4 (12.9–28.2) | 56.1 (33.3–79.0) | 2.43 (1.34–4.41) | ||
Bold values denote statistical significance at the p<0.05 level.
ABX, antibiotics; CI, confidence interval; G+ ABX, anti-gram positive antibiotics; HR, hazard ratio; ---, reference level.
Fig. 1. PFS2 in women with recurrent platinum sensitive epithelial ovarian cancer undergoing platinum chemotherapy treated with or without concurrent ABX therapy.
ABX, antibiotics; G+ ABX, anti-gram-positive antibiotics; PFS2, time to second progression.
Univariate analysis for the time from treatment start to platinum resistance demonstrated no difference for any ABX vs. no ABX (14.9 vs. 14.9 months: HR=0.94; 95% CI=0.70–1.27; p=0.69), G+ ABX versus no ABX (10.2 vs. 14.9 months: HR=1.52; 95% CI=0.93–2.50; p=0.09) and other ABX versus no ABX (16.0 vs. 14.9 months: HR=0.82; 95% CI=0.59–1.15; p=0.25). However, G+ ABX were associated with significantly shorter time to platinum resistance compared to other ABX (10.2 vs. 16.0 months: HR=1.76; 95% CI=1.02–3.07; p=0.04).
There was no difference in OS for ABX versus none (30.8 vs. 33.5 months: HR=1.01; 95% CI=0.73–1.39; p=0.97) and other ABX versus none (37.9 vs. 33.5 months: HR=0.81; 95% CI=0.56–1.17; p=0.26). Notably, G+ ABX had significantly decreased OS compared to no ABX (26.4 vs. 33.4 months: HR=2.13; 95% CI=1.28–3.57; p=0.004) and other ABX (26.4 vs. 37.9 months: HR=2.43; 95% CI=1.34–4.41; p=0.003), respectively (Fig. 2).
Fig. 2. OS in women with recurrent platinum sensitive epithelial ovarian cancer undergoing platinum chemotherapy treated with or without concurrent ABX therapy.
ABX, antibiotics; G+ ABX, anti-gram-positive antibiotics; OS, overall survival.
3. Multivariable analysis
Results of multivariable analysis controlling for stage and residual disease at initial surgery are displayed in Table 3. Receipt of no ABX (HR=0.54; 95% CI=0.33–0.88; p=0.014) or other ABX (HR=0.65; 95% CI=0.38–1.10; p=0.011) were associated with improved PFS2 compared to G+ ABX. Furthermore, no ABX (HR=0.55; 95% CI=0.32–0.95; p=0.033) or other ABX (HR=0.65; 95% CI=0.40–1.07; p=0.087) were associated with significantly improved time to platinum resistance compared to G+ ABX. Receipt of no ABX (HR=0.49; 95% CI=0.29–0.81; p=0.006) or other ABX (HR=0.40; 95% CI=0.22–0.72; p=0.002) was associated with improved OS, compared to G+ ABX.
Table 3. Multivariable analysis for oncologic outcomes in women with recurrent platinum sensitive epithelial ovarian cancer undergoing platinum chemotherapy.
| Variable | Cox multivariable hazard ratio (95% CI) | Cox multivariable Wald p-value (n=20) | ||
|---|---|---|---|---|
| Treatment start to second progression | ||||
| ABX type | ||||
| G+ ABX | --- | --- | ||
| No ABX | 0.54 (0.33–0.88) | 0.014 | ||
| Other ABX | 0.65 (0.38–1.10) | 0.011 | ||
| Stage | ||||
| III | --- | --- | ||
| IV | 1.38 (0.97–1.97) | 0.07 | ||
| Residual disease after surgery | ||||
| Optimal (R0) | --- | --- | ||
| Optimal (Other) | 1.14 (1.06–1.87) | 0.018 | ||
| Suboptimal | 1.48 (0.93–2.34) | 0.094 | ||
| Treatment start to platinum resistance | ||||
| ABX type | ||||
| G+ ABX | --- | --- | ||
| No ABX | 0.65 (0.40–1.07) | 0.087 | ||
| Other ABX | 0.55 (0.32–0.95) | 0.033 | ||
| Stage | ||||
| III | --- | --- | ||
| IV | 0.98 (0.68–1.42) | 0.93 | ||
| Residual disease after surgery | ||||
| Optimal (R0) | --- | --- | ||
| Optimal (Other) | 1.27 (0.94–1.70) | 0.12 | ||
| Suboptimal | 1.39 (0.88–2.20) | 0.16 | ||
| Treatment start to death | ||||
| ABX type | ||||
| G+ ABX | --- | --- | ||
| No ABX | 0.49 (0.29–0.81) | 0.006 | ||
| Other ABX | 0.40 (0.22–0.72) | 0.002 | ||
| Stage | ||||
| III | --- | --- | ||
| IV | 1.08 (0.72–1.64) | 0.70 | ||
| Residual disease after surgery | ||||
| Optimal (R0) | --- | --- | ||
| Optimal (Other) | 1.23 (0.89–1.69) | 0.22 | ||
| Suboptimal | 1.11 (0.67–1.84) | 0.68 | ||
Bold values denote statistical significance at the p<0.05 level.
ABX, antibiotics; CI, confidence interval; G+ ABX, anti-gram positive antibiotics; HR, hazard ratio; ---, reference level.
DISCUSSION
In this large retrospective cohort study of women with recurrent, platinum-sensitive OC undergoing platinum chemotherapy, we demonstrate that treatment with G+ ABX is associated with reduced PFS2 and OS. Further, compared to other ABX, G+ ABX are associated with a shorter time to platinum resistance. Finally, the negative effect of G+ ABX is independent of stage and residual disease at initial surgery, thus controlling for patients at a higher baseline risk of poorer oncologic outcomes.
Burgeoning evidence suggests that the gut microbiome and the female reproductive tract influence carcinogenesis and response to cancer therapy, including immunotherapy and platinum chemotherapy [6,7,8,9,10,11,12,13,14,15,16,17,18]. In preclinical mouse models of non-gynecologic cancers, ABX were used to alter the gut microbiota, and cisplatin or oxaliplatin was administered. Notably, antibiotic treatment was associated with a reduced response rate and OS compared to non-antibiotic controls. The presence of Lactobacillus fermetum, a gram-positive species, was associated with improved response to treatment through the activation of tumor-infiltrating myeloid cells [10]. Subsequently, Pflug and colleagues [9] assessed the impact of G+ ABX during chemotherapy with cisplatin in patients with recurrent lymphoma and leukemia. Notably, G+ ABX were associated with significantly decreased response, PFS, and OS [9].
In pre-clinical studies led by our team, the gut microbiome was disrupted via antibiotic therapy in mouse models of OC, and then treated with platinum-based chemotherapy was initiated [21]. Notably, mice treated with ABX had significant changes to the gut microbiome and significantly accelerated tumor growth and worse OS. Notably, cecal microbiota transplantation from control mice, not treated with ABX, was able to overcome this negative antibiotic effect and restore treatment response [22,23]. We additionally evaluated the impact of ABX upon oncologic outcomes in 424 women undergoing platinum and taxane chemotherapy for newly diagnosed stage III and IV OC after upfront or interval cytoreductive surgery [7]. ABX during chemotherapy were associated with a significantly decreased PFS and OS, most pronounced in women who received G+ ABX. Notably, this effect was independent of performance status and Clavien-Dindo post-operative complications on multivariable analysis. Despite this, a significant criticism of our previous study was that even though ABX administered only during chemotherapy were included in the analysis, the significance of necessary peri-operative ABX treatment, post-operative infectious morbidity, and the associated unmeasured frailty was unclear [7]. Building on this prior data, the present study was designed to assess further the impact of ABX upon platinum chemotherapy response in a homogenous cohort of women with recurrent, platinum-sensitive OC in whom only a minority were treated with secondary cytoreductive surgery. Here, we identified that G+ ABX specifically were associated with shorter time to platinum resistance, decreased PFS, and OS. While the cohorts were overall well matched for demographics, those who were treated with ABX had a higher incidence of diabetes mellitus and increased body mass index (BMI), which are important confounders. Therefore, we are unable to elucidate whether it is antibiotic specifically, or a baseline increased medical frailty and subsequent susceptibility to infections that contribute to the worse survival outcomes.
Studies have demonstrated that a cervicovaginal microbiota with a lower abundance of gram-positive Lactobacillus species is associated with a higher risk of OC. For example, in a case-control study of women with a BRCA1 mutation, the presence of a community type O cervicovaginal microbiota, where lactobacilli accounted for less than 50% of species, was significantly higher in women with OC (odds ratio=2.80), compared to age-matched healthy controls [22]. Further, in a study by Jacobson, only 24% of women with OC had a Lactobacillus-dominant cervicovaginal microbiota, much lower than studies of similarly aged women without OC, where Lactobacillus communities are dominant in approximately one-half of women [22,23]. This data supporting an inverse relationship between a Lactobacillus species dominant vaginal microbiota and OC risk is undoubtedly intriguing in the context of our studies’ results and warrants further prospective investigation.
It is essential to consider the limitations of our study when interpreting the results, including the retrospective cohort design without case-control matching, the small sample size of patients treated with G+ ABX, with data collected from a single tertiary healthcare institution. Further, our study lacks a corresponding analysis of the patient’s stool samples to determine how ABX impact the microbiota. Therefore, while our results are consistent with other prospective studies in non-gynecologic cancers and our retrospective data in women with newly diagnosed OC, from this data alone, we are unable to determine the causation of whether it is G+ ABX use specifically or the physiologic stress of infectious sequelae or unmeasured medical frailty, during platinum chemotherapy that led to worsened oncologic outcomes. Further, from the available data, we are unable to draw conclusions about the duration of antibiotic therapy and outcomes, as well as performing correlations between time on chemotherapy and associated outcomes. There are important differences between the cohorts who received ABX compared to those who did not, including BMI and diagnoses of diabetes. Lastly, despite seeing a statistically significant effect in this population, our study is limited by the small sample size of those who received G+ ABX, who accounted for only 7.7% of the study population. Therefore, replicating these findings in larger, more heterogeneous cohorts is essential.
Despite these limitations, our findings contribute essential information to the literature regarding G+ ABX’ impact on platinum chemotherapy response and resistance in women with recurrent, platinum-sensitive OC. This data builds on our prior study of women with newly diagnosed OC undergoing chemotherapy. We demonstrated that ABX, most notably G+ ABX, were associated with decreased PFS and OS in the front-line setting. Notably, all three cohorts had comparable baseline time to first progression before second-line platinum chemotherapy, arguing for biologic and clinical equivalence before antibiotic exposure in the second-line setting. Furthermore, there were no significant differences in various factors associated with prognosis, including ECOG performance status, hereditary genetic mutations, the timing of initial surgery and residual disease, stage, secondary cytoreductive surgery, and bevacizumab, PARP inhibitors, and HIPEC.
Given that platinum-based chemotherapy is standard-of-care for treating women with newly diagnosed and recurrent platinum-sensitive OC, interventions associated with the development of resistance should be evaluated. In addition, understanding how G+ ABX, and potentially the gut and female reproductive tract microbiotas, impact platinum chemotherapy response and resistance in women with OC may provide an actionable therapeutic strategy in the future. Lastly, clinicians should be aware of the negative association of G+ ABX during platinum chemotherapy and follow national and institutional guidelines for managing infections.
In this single-institution retrospective cohort study of women with recurrent, platinum-sensitive OC treated with platinum chemotherapy, we demonstrate that G+ ABX are associated with decreased disease progression, shorter time to platinum resistance, and worsened OS. Thus, advancing our understanding of how G+ ABX and potentially microbiome dysbiosis impacts platinum chemotherapy response and resistance holds promise for improving outcomes in women with OC.
Footnotes
Funding: This study was supported in part by K12 Grant (5K12CA076917-22) for Roberto Vargas, MD.
Conflict of Interest: No potential conflict of interest relevant to this article was reported.
- Conceptualization: C.L.M., R.O., R.P.G., M.C.M., V.R.
- Data curation: C.L.M., C.J., M.C.M., V.R.
- Funding acquisition: V.R.
- Investigation: C.L.M., M.C.M., V.R.
- Methodology: C.L.M., M.C.M., V.R.
- Project administration: C.L.M.
- Supervision: M.C.M., V.R.
- Writing - original draft: C.L.M., R.O., R.P.G., M.C.M., V.R.
- Writing - review & editing: C.L.M., C.J., Y.M., R.O., R.P.G., M.C.M., V.R.
SUPPLEMENTARY MATERIAL
Study patient inclusion and exclusion criteria.
References
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70:7–30. doi: 10.3322/caac.21590. [DOI] [PubMed] [Google Scholar]
- 2.Heintz AP, Odicino F, Maisonneuve P, Quinn MA, Benedet JL, Creasman WT, et al. Carcinoma of the ovary. FIGO 26th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet. 2006;95(Suppl 1):S161–S192. doi: 10.1016/S0020-7292(06)60033-7. [DOI] [PubMed] [Google Scholar]
- 3.Coleridge SL, Bryant A, Lyons TJ, Goodall RJ, Kehoe S, Morrison J. Chemotherapy versus surgery for initial treatment in advanced ovarian epithelial cancer. Cochrane Database Syst Rev. 2019;2019:CD005343. doi: 10.1002/14651858.CD005343.pub5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kehoe S, Hook J, Nankivell M, Jayson GC, Kitchener H, Lopes T, et al. Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial. Lancet. 2015;386:249–257. doi: 10.1016/S0140-6736(14)62223-6. [DOI] [PubMed] [Google Scholar]
- 5.Hanker LC, Loibl S, Burchardi N, Pfisterer J, Meier W, Pujade-Lauraine E, et al. The impact of second to sixth line therapy on survival of relapsed ovarian cancer after primary taxane/platinum-based therapy. Ann Oncol. 2012;23:2605–2612. doi: 10.1093/annonc/mds203. [DOI] [PubMed] [Google Scholar]
- 6.Chambers LM, Michener CM, Rose PG, Reizes O, Yao M, Vargas R. Impact of antibiotic treatment on immunotherapy response in women with recurrent gynecologic cancer. Gynecol Oncol. 2021;161:211–220. doi: 10.1016/j.ygyno.2021.01.015. [DOI] [PubMed] [Google Scholar]
- 7.Chambers LM, Kuznicki M, Yao M, Chichura A, Gruner M, Reizes O, et al. Impact of antibiotic treatment during platinum chemotherapy on survival and recurrence in women with advanced epithelial ovarian cancer. Gynecol Oncol. 2020;159:699–705. doi: 10.1016/j.ygyno.2020.09.010. [DOI] [PubMed] [Google Scholar]
- 8.Chambers LM, Bussies P, Vargas R, Esakov E, Tewari S, Reizes O, et al. The microbiome and gynecologic cancer: current evidence and future opportunities. Curr Oncol Rep. 2021;23:92. doi: 10.1007/s11912-021-01079-x. [DOI] [PubMed] [Google Scholar]
- 9.Pflug N, Kluth S, Vehreschild JJ, Bahlo J, Tacke D, Biehl L, et al. Efficacy of antineoplastic treatment is associated with the use of antibiotics that modulate intestinal microbiota. OncoImmunology. 2016;5:e1150399. doi: 10.1080/2162402X.2016.1150399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Iida N, Dzutsev A, Stewart CA, Smith L, Bouladoux N, Weingarten RA, et al. Commensal bacteria control cancer response to therapy by modulating the tumor microenvironment. Science. 2013;342:967–970. doi: 10.1126/science.1240527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Routy B, Le Chatelier E, Derosa L, Duong CP, Alou MT, Daillère R, et al. Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors. Science. 2018;359:91–97. doi: 10.1126/science.aan3706. [DOI] [PubMed] [Google Scholar]
- 12.Viaud S, Saccheri F, Mignot G, Yamazaki T, Daillère R, Hannani D, et al. The intestinal microbiota modulates the anticancer immune effects of cyclophosphamide. Science. 2013;342:971–976. doi: 10.1126/science.1240537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pinato DJ, Howlett S, Ottaviani D, Urus H, Patel A, Mineo T, et al. Association of prior antibiotic treatment with survival and response to immune checkpoint inhibitor therapy in patients with cancer. JAMA Oncol. 2019;5:1774–1778. doi: 10.1001/jamaoncol.2019.2785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gopalakrishnan V, Spencer CN, Nezi L, Reuben A, Andrews MC, Karpinets TV, et al. Gut microbiome modulates response to anti-PD-1 immunotherapy in melanoma patients. Science. 2018;359:97–103. doi: 10.1126/science.aan4236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Abdel-Rahman O, Ghosh S, Walker J. Outcomes of metastatic colorectal cancer patients in relationship to prior and concurrent antibiotics use; individual patient data analysis of three clinical trials. Clin Transl Oncol. 2020;22:1651–1656. doi: 10.1007/s12094-020-02301-1. [DOI] [PubMed] [Google Scholar]
- 16.Wilson BE, Routy B, Nagrial A, Chin VT. The effect of antibiotics on clinical outcomes in immune-checkpoint blockade: a systematic review and meta-analysis of observational studies. Cancer Immunol Immunother. 2020;69:343–354. doi: 10.1007/s00262-019-02453-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vich Vila A, Collij V, Sanna S, Sinha T, Imhann F, Bourgonje AR, et al. Impact of commonly used drugs on the composition and metabolic function of the gut microbiota. Nat Commun. 2020;11:362. doi: 10.1038/s41467-019-14177-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yoon MY, Yoon SS. Disruption of the gut ecosystem by antibiotics. Yonsei Med J. 2018;59:4–12. doi: 10.3349/ymj.2018.59.1.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Armstrong DK, Alvarez RD, Bakkum-Gamez JN, Barroilhet L, Behbakht K, Berchuck A, et al. Ovarian Cancer, version 2.2020, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2021;19:191–226. doi: 10.6004/jnccn.2021.0007. [DOI] [PubMed] [Google Scholar]
- 20.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chambers LM, Esakov Rhoades EL, Bharti R, Braley C, Tewari S, Trestan L, et al. Disruption of the gut microbiota confers cisplatin resistance in epithelial ovarian cancer. Cancer Res. 2022;82:4654–4669. doi: 10.1158/0008-5472.CAN-22-0455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nené NR, Reisel D, Leimbach A, Franchi D, Jones A, Evans I, et al. Association between the cervicovaginal microbiome, BRCA1 mutation status, and risk of ovarian cancer: a case-control study. Lancet Oncol. 2019;20:1171–1182. doi: 10.1016/S1470-2045(19)30340-7. [DOI] [PubMed] [Google Scholar]
- 23.Jacobson D, Moore K, Gunderson C, Rowland M, Austin R, Honap TP, et al. Shifts in gut and vaginal microbiomes are associated with cancer recurrence time in women with ovarian cancer. PeerJ. 2021;9:e11574. doi: 10.7717/peerj.11574. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Study patient inclusion and exclusion criteria.