Metronomic chemotherapy using oral cyclophosphamide and bevacizumab for recurrent cervical cancer: A multi-institutional retrospective study

Highlights • No standard chemotherapy exists for recurrent cervical cancer patients who respond poorly to platinum administration.• Metronomic chemotherapy is a low-toxicity treatment and can maintain quality of life.• Low-dose oral cyclophosphamide and bevacizumab (CPA-BEV) chemotherapy shows promising results in recurrent cervical cancer.


Introduction
The standard first-line chemotherapy for recurrent cervical cancer is a combination of platinum and paclitaxel, or a triple regimen of these reagents combined with bevacizumab (Tewari et al., 2017). However, if this first-line treatment fails, there is no standard treatment for secondline chemotherapy. In particular, patients who relapse within a short period after treatment with platinum-based agents do not respond well to re-administration of platinum-based agents, with relapse occurring within 12 months according to Moore's criteria (Tewari et al., 2015) and less than 7 months according to Takekuma et al. (2017). The objective of second-line chemotherapy is to palliate symptoms and improve quality of life. In the GOG240 study, only 20% of patients received posttreatment chemotherapy, and the prognosis after relapse in posttreatment patients was not reported.
Metronomic chemotherapy is a low-toxicity treatment that is expected to inhibit angiogenesis and normalize vascular architecture through frequent small-dose administration of cytotoxic anticancer agents (Mpekris et al., 2017). Bevacizumab (an anti-vascular endothelial growth factor antibody), which has similar effects, has shown high synergy when combined with weekly paclitaxel in platinum-resistant recurrent ovarian cancer (Pujade-Lauraine et al., 2014). Weekly paclitaxel administration is also included in metronomic chemotherapy. Combination chemotherapy of low-dose oral cyclophosphamide with intravenous bevacizumab (CPA-BEV) has been reported in several breast and ovarian cancers with minimal side effects (Simsek et al., 2019). In this study, we report a multicenter retrospective review of CPA-BEV treatment in patients with recurrent cervical cancer who had a platinum-resistant recurrence or a history of platinum anaphylaxis. We previously reported 4 cases with cervical cancer who had CPA-BEV, (Isono-Nakata et al., 2018) two of which were included in this report. Among other two excluded cases, one case was treated after platinum sensitive recurrence and another case was treated with CPA-BEV in two lines in between.

Materials and methods
Recurrent cervical cancer patients were enrolled in the study if their disease course progressed within 6 months after the last administration of platinum or if they had a history of platinum anaphylaxis, and subsequently received CPA-BEV between December 2016 and December 2021. The CPA-BEV treatment regimen consisted of a daily dose of 50 mg cyclophosphamide administered orally, concurrent with 15 mg/kg doses of bevacizumab administered intravenously every 3 weeks. Adverse events were evaluated according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (CTCAE), version 5.0. (United States Department of Health and Human Services, 2009) Efficacy was evaluated according to the Response Evaluation Criteria In Solid Tumors (RECIST), ver. 1.1. (Eisenhauer et al., 2009). Progressionfree survival (PFS) was defined as the time from the date of the first administration of oral cyclophosphamide to the date of objectively determined disease progression. In principle, computed tomography or magnetic resonance imaging was performed when progression was expected, such as worsening of symptoms or increase in tumor markers, or as a routine examination every 2-6 months. In addition, obvious worsening of symptoms was also considered as progression. Overall survival (OS) was defined as the time from the date of the first administration of oral cyclophosphamide to death. Statistical analyses were performed on the observed distributions of PFS and OS using the Kaplan-Meier method.Statistical analyses were conducted using XLSTAT (Addinsoft, Paris, France).

Eight and 3 patients in Hyogo Hyogo Medical University and Kansai
Rosai Hosipital were treated by multiple physicians and enrolled in this study, respectively. The median age was 50 years (range 29-76 years). The types of cervical cancer pathology represented in the study were squamous cell carcinoma in seven patients, adenocarcinoma in three, and large cell neuroendocrine carcinoma in one. Six patients received CPA-BEV because of disease progression during prior platinum-based chemotherapy, four because of disease progression within 6 months of the last platinum administration, and one because of platinum anaphylaxis. Eight patients had received concurrent chemoradiotherapy (CCRT) as prior therapy, and five patients had received two or more chemotherapy regimens (excluding CCRT). Only one patient had received bevacizumab as prior therapy (Case 4 in Table 1). If a patient was determined to have progressive disease based on RECIST criteria, the CPA-BEV regimen was continued if there was no worsening of symptoms, side effects were minimal, and the attending physician judged that the patient's progress was slow. Adverse events included grade 3 neutropenia in one patient, grade 2 proteinuria in one, grade 1 anorexia in one, and grade 1 gastrointestinal hemorrhage in one (Table 1). No non-hematologic toxicities of grade 2 or higher were observed. The median duration of chemotherapy was 2.8 months (range 0.2-30.6 months); one patient had a complete response but no patient had a partial response. Median PFS was 2.8 months (95% CI: 2.1-10.7 months) and median OS was 13.6 months (95% CI: 8.4-33.7 months). (Fig. 1) The PFS rate at 6 months was 27% (95% CI: 1.0-54%), and the OS rate at 18 months was 36% (95% CI: 7.9-64.8%).

Discussion
The median PFS and OS were 2.8 months and 13.6 months, respectively, of recurrent cervical cancer patients who received CPA-BEV as their second-line chemotherapy. The PFS rate at 6 months was 27% (95% CI: 1.0-54%). These outcomes compare favorably to the results of previous second-line chemotherapy studies. First, in a GOG phase II study of single agent bevacizumab for patients with recurrent cervical cancer, the primary endpoint measures of PFS for at least 6 months, median PFS, and median OS were 23.9% (90% CI, 14%-37%), 3.4 months (95% CI: 2.5-4.5 months), and 7.3 months (95% CI: 6.1-10.4 months), respectively (Monk et al., 2009). Secondly, the immune checkpoint inhibitor cemiplimab and an investigator's choice of a single cytotoxic anticancer agent were compared for second-line treatment in the GOG-3016 randomized trial (Tewari et al., 2022). The results for cemiplimab showed a median PFS of 2.8 months (95 %CI, 2.6-3.9 months) and a median OS of 12 months (95% CI: 10.3-13.5 months), but those of the cytotoxic anticancer agent showed a PFS of 2.9 months (95% CI: 2.7-3.4 months) and an OS of 8.5 months (95% CI: 7.5-9.6 months). In the GOG-3016 study, the OS at a median follow-up of 18 months for patients treated with cemiplimab was similar to that revealed in our study. Moreover, CPA-BEV treatment appears to be welltolerated. The only grade 3 or higher adverse event in our study was neutropenia, occurring in one patient. By contrast, in the above randomized controlled trials, grade 3 or higher adverse events occurred in 45% and 53.4% of patients in the cemiplimab and anticancer chemotherapy groups, respectively. This suggests that CPA-BEV is an effective and less toxic treatment for recurrent cervical cancer in the second-line and beyond, compared with not only single-agent anticancer chemotherapy but also immune checkpoint agents. In addition to inhibiting angiogenesis, CPA-BEV is believed to enhance tumor immunity by optimizing vascularization within the tumor microenvironment and modulating regulatory T cells (Simsek et al., 2019). Although immune checkpoint agents for cervical cancer have not been approved in Japan, pembrolizumab has recently been approved for solid tumors presenting with high tumor mutation burden (TMB-H, ≥10 Muts/Mb). TMB-H has been reported in 21% of cervical cancer cases, and the response rate of cervical cancer to immune checkpoint inhibitor treatment is as high as 31% (Marabelle et al., 2020). Before the results of GOG-3016, the usefulness of 1st line pembrolizumab was also reported, and expectations for immune-mediated drug therapy for the treatment of cervical cancer are high (Colombo et al., 2021). Although the present study could not investigate CPA-BEV therapy and its immunostimulatory effects in cervical cancer, a phase II trial of CPA-BEV plus pembrolizumab has been reported in recurrent ovarian cancer, (Zsiros et al., 2021) and there is potential for future development of combination therapy with CPA-BEV and immune checkpoint agents in cervical cancer as well. The limitation of this study is that only one patient who received prior bevacizumab treatment was enrolled in this study. The concept of 'bevacizumab beyond bevacizumab' was established after randomized control study of platinum sensitive recurrent ovarian cancer (Pignata, Lancet 2021), however, it remains obscure for cervical cancer.

Conclusion
The CPA-BEV regimen appears to be a promising treatment for recurrent cervical cancer with minimal toxicity and antitumor effects.

Consent
Written informed consent was obtained from all patients for publication of the case series.