Cisplatin plus paclitaxel chemotherapy with or without bevacizumab in postmenopausal women with previously untreated advanced cervical cancer: a retrospective study

Consecutive postmenopausal women with previously untreated advanced CC who received CPB or CPA between January 2015 and August 2019 and for whom baseline data were available at the present analysis were identified from the First Affiliated Hospital, Sun Yat-sen University. Inclusion criteria were as follows: Chinese population; postmenopausal women with previously untreated histologically proven CC stage IV/B; advanced CC was confirmed by our institutional pathology laboratory and clinical imaging data according to FIGO 2018 cervical cancer staging criteria [12]; measurable disease; and a GOG performance status score of 0 or 1(0: fully active; 1: physically strenuous activities but ambulatory). Key exclusion criteria were as follows: prior use of targeted drug(s); previous chemotherapy, radiotherapy, chemoradiotherapy, or surgery for advanced CC; CPB or CPA discontinuation, regardless of the drug-induced adverse events (AEs); symptomatic brain metastases; cachexia; severe medical illness (such as, uncontrolled infection or hypertension, hypertension with multiple complications, HIV infection); surgical emergency (such as, intestinal obstruction); key organ function failure (such as, lung, brain, kidney, and/or heart); active bleeding conditions or coagulopathy; arterial or venous thrombosis; dementia or psychiatric disorders; and collagenosis.

A retrospective, single-centre study was conducted in which eligible postmenopausal women had underwent intravenous CPB or CPA regimen [1]. The CPB treatment consisted of cisplatin (50 mg/m2 on day 1) plus paclitaxel (175 mg/m2 on day 1) plus BEV (15 mg/kg on day 1). The CPA treatment consisted of cisplatin (50 mg/m2/day) plus paclitaxel (175 mg/m2/day). The regimens for CPB and CPA were repeated every 21 days until disease progression, death, or intolerable toxic effects. The management of hypertension in patients with advanced CC receiving BEV was consistent with previous recommendations reported by Plummer et al. [13].

The primary endpoints were overall survival (OS) and progression-free survival (PFS); secondary endpoints were the frequency and severity of AEs. OS was defined as the date form drug treatment to death from any cause or date last seen. PFS was defined as the date from drug treatment to the date of either disease progression or death from any cause. Postmenopausal women were defined as female patients who had not menstruated for at least a year, excluding menopause due to related diseases (i.e., endometriosis, endocrine disorders). Disease progression was defined according to the Response Evaluation Criteria in Solid Tumours (RECIST, version 1.1) [14]. Computed tomography (CT) was carried out at each follow-up. Safety was assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0 [15]. Persistent disease was defined as not achieving a complete clinical response [8]. Pain was assessed using the Brief Pain Inventory [16]. Details related to pain and hypertension (occurrence, duration, and severity) were registered. Follow-up time was calculated from the date of clinical staging to the date last seen. Follow-up visits occurred 1 and 3 months after starting CPB or CPA. Afterwards, this assessment was conducted every 3 months for 1 year and then yearly thereafter.

Categorical variables were compared using Chi-square tests; continuous variables were compared using Student’s t-test for normally distributed variables and Mann-Whitney U test for non-normally distributed variables. Survivorship curves were completed by means of the Kaplan-Meier methods with a log-rank test. The acquisition of the hazard ratio (HR) and confidence interval (CI) were achieved through a stratified log-rank test. All statistical tests used a 2-tailed of 0.05. The survival curves were made using GraphPad Prism 8.0. To perform other statistical analysis, SPSS 26.0 (IBM, Inc., NY, America) was used.

We identified 350 postmenopausal women with previously untreated advanced CC, of whom 246 postmenopausal women were included. Of these 246 women, 124 received CPB, and 122 received CPA (Fig. 1). Table 1 showed the characteristics of postmenopausal women who underwent CPB versus CPA. The mean age was 62.1[±7.59] years for CPB and 62.2[±6.84] years for CPA. ECOG performance status was 0 in 37.1% of patients and 1 in 62.9% of patients receiving CPB versus 0 in 34.4% of patients and 1 in 65.6% of patients receiving CPA (p = 0.663). Statistically significant differences were not detected with respect to age, body mass index (BMI), haemoglobin < 11 g/dL, histology, IV/B CC, duration of drug, GOG performance status, and number of metastatic sites. The median follow-up for the entire cohort was 24 months (range, 2–32). The median number of drug cycles for individuals undergoing CPB was 8 (range, 1–26), and for those who underwent CPA, the median was 7 (range, 1–28). Of 246, 197 (80.1%) individuals discontinued CPB or CPA, primarily attributed to disease progression (55.8% for CPB vs. 44.2% for CPA, p = 0.001). Even though more patients developed disease progression in the CPB group but there was a significant delay in the time taken for disease to progress in this group which contributed to significantly longer PFS.

Figures 2 and 3 show the KM survival plot for the differences in OS and PFS between the 2 different regimens at the final follow-up, respectively. Throughout the follow-up period, 129 deaths were reported (52% [129/246]; 58 patients for CPB vs. 71 patients for CPA). A borderline noteworthy distinction was detected in the median OS between the two regimens (16.4 months [95% CI, 15.3–17.1] for CPB vs. 12.3 months [95% CI, 10.2–13.5] for CPA). The incorporation of BEV significantly improved the median OS compared with chemotherapy alone (HR 0.69, 95% CI, 0.49–0.99; p = 0.001). There was also a distinct difference in the median PFS between the two regimens (9.2 months [95% CI, 8.3–10.7] for CPB vs. 7.9 months for CPA (95% CI, 6.1–8.6) (HR 0.62, 95% CI, 0.47–0.82; p < 0.001).

Of the 246 patients, 43 (25 [20.2%] receiving CPB vs. 18 [14.8%] receiving CPA; p = 0.265) had not documented persistent disease at the final follow-up. Table 2 demonstrates the frequency of AEs probably associated with CPB or CPA. Compared with CPB, CPA was associated with fewer AEs (9.9% vs. 12.0%, p < 0.001). There was a significant difference between the groups for hypertension grade ≥ 2 (25.0% receiving CPB vs. 4.1% receiving CPA, p < 0.001), but it was generally controllable. Additionally, significant differences were noted between the groups for neutropenia grade ≥ 4 (33.9% receiving CPB vs. 14.8% receiving CPA, p < 0.001) and thrombosis/embolism grade ≥ 3 (8.9% receiving CB vs. 2.5% receiving CA, p = 0.030). There were no significant differences in terms of genitourinary fistula grade ≥ 2, gastrointestinal fistula grade ≥ 2, gastrointestinal bleeding grade ≥ 3, or pain grade ≥ 2.