Article Text
PDF
Abstract
Objective In this study we investigated response rates of bevacizumab in addition to weekly paclitaxel and carboplatin in neoadjuvant setting in cervical cancer stage IB–IIB.
Methods In this retrospective study we included patients with FIGO 2018 stage IB–IIB cervical cancer. Treatment consisted of 9 weeks' neoadjuvant paclitaxel and carboplatin (paclitaxel 60 mg/m2, carboplatin AUC 2.7; both weekly) and bevacizumab (15 mg/kg every 3 weeks). The radiologic response rate was analyzed using the Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 criteria. The definition of optimal pathological response was complete disappearance of tumor (complete response, pCR) or residual disease with less than 3 mm stromal invasion (pPR1). Suboptimal pathologic response (pPR2) was defined as persistent residual disease with more than 3 mm stromal invasion.
Results A total of 30 patients were included. Six patients had FIGO 2018 stage IB1–IB2 (20%), one had stage IB3 (3%), five had stage IIA (17%), and 18 had stage IIB (60%). After completing the neoadjuvant chemotherapy, all patients showed a RECIST response (seven (23%) complete response; 23 (77%) partial response). Six patients (20%) were judged to be still inoperable. After radical hysterectomy, optimal pathological response was observed in 11 patients (38%) (pCR in nine patients (29%) and pPR1 in two patients (8%)). Six patients (20%) received postoperative adjuvant chemoradiotherapy. Hematological toxicity was similar to neoadjuvant weekly paclitaxel and carboplatin, as we reported earlier. Grade IV proteinuria or hypertension was not observed and no administration of bevacizumab was delayed or dose-reduced.
Conclusion Bevacizumab in addition to weekly paclitaxel and carboplatin showed a 100% radiological RECIST response and an optimal pathological response of 38%. Although bevacizumab has an established role in the treatment of recurrent cervical cancer in combination with paclitaxel and carboplatin, we did not observe a tendency toward superior effect on the pathological response rate of bevacizumab in the neoadjuvant chemotherapy setting.
- cervical cancer
- uterine cervical neoplasms
Data availability statement
All data relevant to the study are included in the article.
Statistics from Altmetric.com
HIGHLIGHTS
No new unexpected adverse events were observed with the combination of bevacizumab and weekly paclitaxel carboplatin in the neoadjuvant setting in cervical cancer.
Bevacizumab with weekly paclitaxel carboplatin resulted in a RECIST response in all patients.
The optimal pathological response rate on bevacizumab with weekly paclitaxel carboplatin was 38%.
Introduction
Cervical cancer is the fourth most common cancer in females worldwide, with an estimated more than 500 000 new cases a year and 300 000 deaths.1 For early-stage cervical cancer (FIGO 2018 IA1–IB2), the standard treatment consists of primary surgery. Concomitant radiochemotherapy is an alternative if a patient is unfit for surgery. Locally advanced cervical cancer (FIGO 2018 IB3–IVA) is usually treated by concomitant radiochemotherapy, but neoadjuvant chemotherapy followed by radical surgery has been used as an alternative for cervical cancers FIGO 2018 stage IB3–IIB.2–4 Recent randomized studies showed a similar overall survival with neoadjuvant chemotherapy followed by radical hysterectomy compared with concomitant radiochemotherapy, however with a higher local recurrence rate, especially in FIGO 2018 stage IIB disease. However, neoadjuvant chemotherapy showed less long-term toxicity compared with concomitant radiochemotherapy, which might be important for quality of life. In addition, neoadjuvant chemotherapy has been used in the fertility-sparing management of cervical cancer patients who cannot be operated upfront with a fertility-sparing approach.5 Hence, there remains a need to optimize neoadjuvant regimens in order to improve response and local control.6 7
Several studies have investigated different chemotherapeutics in order to identify the most effective agents in the neoadjuvant chemotherapy setting of cervical cancer.8 9 The combination of cisplatin-paclitaxel-ifosfamide appeared to be the most active regimen according to the Studio Neo-Adjuvante Portio (SNAP) trials.10–14 In the SNAP-01 trial, an Italian study group showed that paclitaxel-ifosfamide-cisplatin had a higher response rate than ifosfamide-cisplatin. There was, however, no significant increase in overall survival.14 Lissoni et al compared, in the SNAP-02 trial, 3-weekly paclitaxel-cisplatin with 3-weekly paclitaxel-ifosfamide-cisplatin and concluded that with the addition of ifosfamide to paclitaxel-cisplatin, the pathological response was higher (25% vs 43% optimal pathological response), but was associated with more severe morbidity, namely grade 3–4 neutropenia, in 76% of patients treated with the ifosfamide-containing regimen vs 26% treated without ifosfamide.15
A recent study showed that 3-weekly paclitaxel-carboplatin had similar efficacy compared with 3-weekly paclitaxel-cisplatin in recurrent cervical cancer.16 Mori et al showed that in the neoadjuvant setting paclitaxel weekly at 60 mg/m2 in combination with carboplatin (AUC 2), resulted in a response rate of 87%.17 We also observed that dose-dense paclitaxel-carboplatin resulted in a Response Evaluation Criteria in Solid Tumors (RECIST) response rate of 87% in recurrent or primary advanced cervical cancer.18
Based on the meta-analysis of Rydzewska et al8 establishing the importance of dose-dense regimens in the treatment of primary cervical cancer and our results with dose-dense paclitaxel-carboplatin in the recurrent setting, we investigated the role of dose-dense paclitaxel-carboplatin as neoadjuvant chemotherapy in cervical cancer.19 In this study, 36 patients were treated with neoadjuvant paclitaxel-carboplatin, both weekly, for stage FIGO 2018 IB3–IIB cervical cancer.19 We concluded that weekly paclitaxel-carboplatin showed similar radiological and pathological responses as 3-weekly paclitaxel-ifosfamide-cisplatin, with less hematologic and non-hematologic toxicity. The recent meta-analysis of Yang et al indicated that, in cases of locally advanced cervical cancer (FIGO 2018 stages IB3, IIA2, and IIB), treatment with neoadjuvant chemotherapy followed by surgery, seemed to be an acceptable treatment option. Advantages were observed in reducing surgical morbidity without increasing the difficulty of surgery or decreasing survival rate.20
Bevacizumab, a humanized monoclonal antibody that is selectively binding to vascular endothelial growth factor, emerged in the first decade of the 21st century as an important molecule to prevent tumor angiogenesis. As tumor angiogenesis is known to drive HPV-mediated cervical carcinogenesis, the Oncology Group (GOG) set up a randomized controlled open-labeled phase 3 trial (GOG-240) evaluating the addition of bevacizumab to chemotherapy in metastatic or recurrent cervical cancer. A significant overall- and progression-free survival benefit was observed,21 resulting in a new standard of care in first-line recurrent or advanced cervical cancer, consisting of cisplatin, paclitaxel, and bevacizumab. Therefore, in the current study we evaluated whether adding bevacizumab to weekly paclitaxel-carboplatin would be associated with improved radiological and pathological responses in the neoadjuvant setting of FIGO 2018 stage I–IIB cervical cancer.
Methods
This retrospective cohort study was approved by the KU Leuven University Hospital’s Ethical Research Committee (MP002356). All consecutive patients with a newly diagnosis of cervical cancer FIGO 2018 stage IB–IIB, who were not candidates for primary surgery according to our institutional guidelines, and presented without co-morbidities excluding surgery, were recruited between December 2015 and October 2017 at the University Hospitals Leuven in Belgium.
Chemotherapy was administered at a weekly dose of paclitaxel 60 mg/m2 and weekly carboplatin AUC 2.7. Bevacizumab was added to the chemotherapy at a dose of 15 mg/kg every 3 weeks. This regimen of weekly paclitaxel-carboplatin and 3-weekly bevacizumab was administered for 9 consecutive weeks. Patients were planned to receive bevacizumab at week 1, 4, and 7 for a total of three times. Interval between last administration of chemotherapy and surgery was 4 to 8 weeks. A data analysis of all patients treated with neoadjuvant weekly paclitaxel-carboplatin and 3-weekly bevacizumab was made. All patients were planned to receive neoadjuvant chemotherapy followed by surgery. Only patients with squamous cell carcinoma, adenocarcinoma, clear cell carcinoma, or adenosquamous carcinoma of the cervix were included.
At diagnosis, a pretreatment evaluation with medical history, physical examination including pelvic examination under anesthesia, and cystoscopy, preoperative blood test including tumor markers (squamous cell carcinoma antigen if squamous histology and CA-125 if adenocarcinoma), MRI scan of the pelvis, and CT of the thorax and abdomen or whole-body positron emission tomography (PET-CT), was performed in all patients.
We repeated the clinical examination of the pelvis, MRI, or PET-CT, 1 to 3 weeks after the last chemotherapy course, to evaluate the response using the RECIST criteria (version 1.1).22 Patients were deemed inoperable if after neoadjuvant chemotherapy no RECIST response was observed, lateral infiltration of the parametrium was still present, or large volume (more than two centimeters) metastatic lymph nodes were present. If deemed operable, the patient underwent radical hysterectomy. Patients with early-stage cervical cancer wishing to preserve fertility underwent a conization. Patients with FIGO 2018 stage IB3–IIB cervical cancer underwent a Querleu type B1 or C1 Wertheim–Meigs radical hysterectomy. Patients considered to be inoperable were treated with concomitant radiochemotherapy (fractions of 1.8 Gy, for a total of 45 Gy or 25 sessions, combined with six-times weekly cisplatin at a dose of 40 mg/m2).
Similarly to the SNAP-01 trial, we defined pathological response on the resection specimen to be an optimal pathological response if complete disappearance of tumor in the cervix with negative pelvic lymph nodes (pCR) or a residual disease with less than 3 mm stromal invasion or in situ carcinoma (pPR1) was observed. Suboptimal response consisted of persistent residual disease with more than 3 mm stromal invasion at the surgical specimen (pPR2).14 Hematologic and non-hematologic toxicities were evaluated according to the Common Terminology Criteria for Adverse Events version 3.0.23 The overall- and progression-free survival was calculated according to Kaplan–Meier. The statistical calculations were made by the SPSS Statistics version 21 by IBM Corporation (Armonk, NY).
Results
A total of 30 patients were included (Table 1). At baseline prior to neoadjuvant chemotherapy, three patients had suspicious lymph nodes (10%). The median age of the patients was 49 years (range; 26–76). Most patients had American Society of Anesthesiologists (ASA)-score 2 (60%). There were 19 patients (63%) with squamous carcinoma, eight patients (27%) with adenocarcinoma, and three patients (10%) with adenosquamous carcinoma. Of the six patients with FIGO 2018 stage IB1–IB2, four patients underwent fertility-sparing treatment, two patients underwent radical hysterectomy after neoadjuvant chemotherapy (one patient because of the presence of two metastatic lymph nodes after neoadjuvant chemotherapy, one patient because of severe comorbidity at diagnosis preventing primary surgery). After receiving neoadjuvant chemotherapy, evaluation according to the RECIST criteria showed seven patients with complete response (23%) and 23 with partial remissions (77%) (online supplemental table 1). No patient had progressive disease. Twenty patients (67%) underwent a radical hysterectomy and four patients (13%) had a conization. The median time between the last course of neoadjuvant chemotherapy and surgery was 40 days (range; 19–85). Two patients had a longer interval than 8 weeks between the last administration of chemotherapy and surgery, one patient because of lung embolism, and one because of patient’s preference. Six patients (20%) were judged to be inoperable due to insufficient response of the tumor; all had FIGO 2018 Stage IIB.
Supplemental material
Baseline characteristics
Pathological complete response was observed in seven patients (29%) (five patients with FIGO 2018 stage IB1–IB2 stage, one patient with stage IB3, and one patient with stage IIA1), pPR1 was seen in two patients (8%) (one patient with FIGO 2018 stage IB3 and one with stage IIB) and 15 patients (63%) showed a pPR2 (one patient with FIGO 2018 stage IB2, two patients with stage IIA1, two patients with stage IIA2, 10 patients with stage IIB). The six inoperable patients after neoadjuvant chemotherapy underwent chemoradiotherapy. Among the 24 patients who underwent surgery, seven patients (29%) received adjuvant chemoradiotherapy, two because of close resection margins and five patients (21%) because of metastatic lymph nodes on pathological examination. None underwent adjuvant chemotherapy.
Grade 1 and 2 anemia was the most frequently observed adverse effect of the neoadjuvant chemotherapy occurring in 27 patients (90%). Only two patients (7%) had grade 2 anemia. Erythropoietin treatment was not used. Grade 3 and 4 neutropenia was observed in 16 patients (53%). Neutropenic fever was not observed. Granulocyte colony stimulating factor (G-CSF) was not used as primary prevention of neutropenia, but was administered to five patients for secondary neutropenia. Dose delays were necessary in four patients and dose reduction in two patients. All patients but one (due to poor performance status) received the planned 9-weekly courses of paclitaxel-carboplatin and the planned three bevacizumab courses. Median-administrated dose of paclitaxel was 60 mg/m2, carboplatin AUC 2.7, and bevacizumab 15 mg/kg. Grade 3 thrombocytopenia was seen in one patient (3%) and grade 4 in two patients (6%). Twenty-six patients (86%) showed grade 1 or 2 arterial hypertension related to bevacizumab (13 (43%) grade 1 and 13 (43%) grade 2). No fistulas were diagnosed during neoadjuvant chemotherapy nor during the 3 months' postoperative period. Grade 3 hypertension was observed in three patients (10%). Grade 1 proteinuria, measured by urine dipstick before each course of bevacizumab,was observed in two patients (6%), grade 2 proteinuria in one patient (3%), and grade 3 in one patient (3%). Neuropathy grade 3 with severe paresthesia in fingers and toes was seen in three patients (10%) (online supplemental table 2).
The median follow-up time was 29 months (range; 7–47). In this period, eight patients (30%) had disease recurrence. None of these patients had an optimal pathological response initially. The mean time to progression was 10.4 months (range; 6–18). In online supplemental table 3, the different sites of recurrence are shown. The estimated Kaplan–Meier overall survival in all patients was 37 months (95% CI 32 to 43 months). The estimated Kaplan–Meier progression-free survival was 35 months (95% CI 29 to 41 months) (Figure 1).
Overall- and progression-free survival according to Kaplan–Meier.
Discussion
Summary of Main Results
When comparing the results in the current study with weekly paclitaxel-carboplatin and 3-weekly bevacizumab and our prior study with weekly paclitaxel-carboplatin,19 similar optimal pathologic responses were observed for weekly paclitaxel-carboplatin and 3-weekly bevacizumab (37%) and weekly paclitaxel-carboplatin (50%) (for FIGO 2018 stage IB1–IB2 67% and 88%, for stage IB3 in 100% and 33% for stage IIA–IIB in 18% and 22%, respectively). The optimal pathological response rates in both series are, however, comparable with the paclitaxel-ifosfamide-cisplatin regimen in the SNAP-01 trial, showing an optimal pathological response of 48%. RECIST responses after completing the neoadjuvant chemotherapy were seen in 90% for weekly paclitaxel-carboplatin and bevacizumab and 89% for weekly paclitaxel-carboplatin. The overall survival was 87% in the weekly paclitaxel-carboplatin and 3-weekly bevacizumab group vs 70% with weekly paclitaxel-carboplatin. The progression-free survival was 77% and 66%, respectively.
Most adverse events observed with weekly paclitaxel-carboplatin and 3-weekly bevacizumab were low-grade, with limited grade 3 and 4 adverse events. The overall adverse event profile was comparable with our former study with weekly paclitaxel-carboplatin, except for hypertension and proteinuria, related to the use of bevacizumab. No cases of neutropenic fever or sepsis were reported.
Results in the Context of the Published Literature
The study of Gupta et al4 randomized patients with cervical cancer FIGO 1994 stage IB2–IIB between concomitant radiochemotherapy and neoadjuvant chemotherapy with 3-weekly carboplatin-paclitaxel followed by surgery and resulted in similar 5-year overall survival in both arms. A recent meta-analysis of 16 studies on this topic by Yang et al20 confirmed these findings and concluded that neoadjuvant chemotherapy followed by surgery resulted in similar oncological outcomes as radiochemotherapy in cervical cancer stage I–II. In addition, the meta-analysis of Rydzewska et al8 on neoadjuvant chemotherapy followed by surgery vs surgery alone in cervical cancer, established the importance of dose-dense regimens in the treatment of primary cervical cancer (as used in the current study).
Prior to the current study, our institution investigated the role of dose-dense paclitaxel-carboplatin as neoadjuvant chemotherapy in cervical cancer in a study with 36 patients, who were treated by neoadjuvant paclitaxel-carboplatin, both weekly, for stage FIGO 2018 IB3–IIB cervical cancer.19 This study showed similar radiological and pathological responses as the chemotherapy scheme (paclitaxel-ifosfamide-cisplatin) in the randomized SNAP-02 trial of Lissoni et al.15
When we compare the adverse events seen in the current study with those in the SNAP-01 trial, we observed lower rates of grade 3 and 4 hematologic toxicity with weekly paclitaxel-carboplatin and 3-weekly bevacizumab than with paclitaxel-ifosfamide-cisplatinum. In addition, there were four toxicity-related deaths in the SNAP-01 trial, and none in the current study. Finally, we need to bear in mind that the addition of ifosfamide in the paclitaxel-ifosfamide-cisplatinum regimen can induce an important gonadotoxic effect, which should be avoided in young patients, especially when fertility-sparing treatment is aimed.5 24
Strengths and Weaknesses
The strength of this study is that this is, to the best of our knowledge, the first study investigating the role of bevacizumab in addition to neoadjuvant chemotherapy in cervical cancer, and that this group of patients was uniformly treated in a monocentric study. The weaknesses are that this study was not randomized and that only 30 patients were included. Furthermore, it is difficult to compare our results with historical data, e.g., with our former study reporting on the neoadjuvant use of weekly paclitaxel-carboplatin. Obviously, comparability of our current and the former series (weekly paclitaxel-carboplatin and 3-weekly bevacizumab vs weekly paclitaxel-carboplatin) is limited because patients were treated in different time periods and more patients with FIGO 2018 stage IIB were included in the weekly paclitaxel-carboplatin and 3-weekly bevacizumab series compared with the weekly paclitaxel-carboplatin cohort.
Implications for Future Research
The current series shows that in cervical cancer, bevacizumab can be safely added to neoadjuvant paclitaxel-carboplatin and that this combination resulted in excellent RECIST response rates, at least comparable to older series using more toxic chemotherapy regimens. However, the pathological response rate did not seem to be superior to the older series. Hence, we believe that this combination deserves further evaluation in larger prospective studies.
Conclusion
Although bevacizumab has an established role in the treatment of recurrent cervical cancer, in combination with paclitaxel and carboplatin,21 we did not observe a tendency toward a superior effect of bevacizumab in combination neoadjuvant chemotherapy in this study. This might be due to the different effect of bevacizumab in the primary neoadjuvant setting, compared with the recurrent setting, or due to the small sample size of this study.
Data availability statement
All data relevant to the study are included in the article.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors Dr CM and Prof IV designed the study. They performed data collection from medical files and performed statistical analysis. All authors performed data analysis, interpretation, and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests Prof. Em IV reports grants and other from Amgen (Europe) GmbH, personal fees and other from AstraZeneca, other from Clovis Oncology Inc. (2018-2019), other from Carrick Therapeutics (2019), other from Debiopharm International SA (2018), other from Deciphera Pharmaceuticals (2020), personal fees from Elevar Therapeutics (2020), other from F. Hoffmann-La Roche Ltd (2018-2020), other from Genmab (2018–2020), personal fees and other from GSK (2019–2020), personal fees and other from Immunogen Inc. (2018–2020), other from Medical University of Vienna (2018), other from MSD (2018–2020), other from Octimet Oncology NV (2019), personal fees and other from Oncoinvent AS (2018–2020), other from Pharmamar (2018), other from Sotio a.s. (2019), other from Tesaro Inc.(2018–2019), other from Verastem Oncology (2020), grants from Roche, personal fees from Mersana (2020), personal fees from Zentalis (2020), personal fees from Novocure (2020), outside the submitted work.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.