Article Text
Abstract
Objective To explore the clinical and biological prognostic factors for advanced ovarian cancer patients receiving first-line treatment with carboplatin, paclitaxel, and bevacizumab.
Methods A multicenter, phase IV, single arm trial was performed. Patients with advanced (FIGO (International Federation of Gynecology and Obstetrics) stage IIIB-IV) or recurrent, previously untreated, ovarian cancer received carboplatin (AUC (area under the curve) 5), paclitaxel (175 mg/m2) plus bevacizumab (15 mg/kg) on day 1 for six 3-weekly cycles followed by bevacizumab single agent (15 mg/kg) until progression or unacceptable toxicity up to a maximum of 22 total cycles. Here we report the final analysis on the role of clinical prognostic factors. The study had 80% power with a two-tailed 0.01 α error to detect a 0.60 hazard ratio with a factor expressed in at least 20% of the population. Both progression-free and overall survival were used as endpoints.
Results From October 2012 to November 2014, 398 eligible patients were treated. After a median follow-up of 32.3 months (IQR 24.1–40.4), median progression-free survival was 20.8 months (95% CI 19.1 to 22.0) and median overall survival was 41.1 months (95% CI 39.1 to 43.5). Clinical factors significantly predicting progression-free and overall survival were performance status, stage, and residual disease after primary surgery. Neither baseline blood pressure/antihypertensive treatment nor the development of hypertension during bevacizumab were prognostic. There were two deaths possibly related to treatment, but no unexpected safety signal was reported.
Conclusions Efficacy and safety of bevacizumab in combination with carboplatin and paclitaxel and as maintenance were comparable to previous data. Hypertension, either at baseline or developed during treatment, was not prognostic. Performance status, stage, and residual disease after primary surgery remain the most important clinical prognostic factors.
Trial registration number EudraCT 2012-003043-29; NCT01706120.
- ovarian cancer
- medical oncology
- ovarian neoplasms
Data availability statement
Data of this study will be shared with publication upon reasonable and motivated request to the Principal Investigator of the study (s.pignata@istitutotumori.na.it). The following IPD will be available for sharing: baseline characteristics of patients, treatment data, safety data, follow-up data. There will be no time limit for data sharing.
Statistics from Altmetric.com
HIGHLIGHTS
Age, histology, and size of center were not prognostic of disease-free survival or overall survival.
Baseline hypertension increased the probability of developing hypertension during treatment.
Development of hypertension was not prognostic for progression-free and overall survival.
Introduction
Ovarian cancer is the most lethal gynecological cancer with 677 700 new diagnoses, 445 800 estimated new deaths in Europe, and a 5-year overall survival of 47.6%. Approximately 80% of ovarian cancer cases are diagnosed at an advanced stage, when the standard treatment is a combination of surgery and platinum-based chemotherapy.1 2 In a landscape of few therapeutic options, the increasing understanding of ovarian cancer biology and resistance has led to the development of new biological compounds. Among these, anti-angiogenic agents play an important role. Aberrant angiogenesis is not only involved in progression and metastatic spread of tumors, but can also reduce drug delivery, therefore contributing to drug resistance.3 4
Bevacizumab, an anti-vascular endothelial growth factor (VEGF) antibody, has been demonstrated to prolong progression-free survival both in newly diagnosed and relapsed ovarian cancer.5–8 Bevacizumab was approved by the European Medicines Agency for the treatment of newly diagnosed, FIGO (International Federation of Gynecology and Obstetrics) stage IIIB-IV, ovarian cancer patients based on the results of two randomized phase III studies, with different schedules and inclusion criteria, showing a significant longer progression-free survival with the addition of bevacizumab to carboplatin/paclitaxel followed by bevacizumab maintenance.6 7 Unfortunately, both studies failed to demonstrate a significant advantage in overall survival with the addition of bevacizumab.6 7 9 10 In a changing landscape due to the introduction of PARP (poly ADP-ribose polymerase) inhibitors, the combination of carboplatin, paclitaxel, bevacizumab, and olaparib has recently been found to be effective and safe, and has been cleared by the Committee for Medicinal Products for Human Use (CHMP) at the European Medicines Agency as a possible new therapeutic option.11
In this scenario, an important question is the evaluation of molecular and clinical factors that may predict the prognosis of patients receiving combined chemotherapy plus bevacizumab. We designed the MITO-16A/MaNGO-OV2A trial (MITO: Multicentre Italian Trials in Ovarian Cancer; MaNGO: Mario Negri Gynecologic Oncology group), with the aim to evaluate potential clinical and molecular prognostic factors and to describe efficacy and safety of bevacizumab. Results might help identify patients who are best candidates for a bevacizumab combination.
Methods
MITO-16A/MaNGO-OV2A (www.clinicaltrials.gov number: NCT01706120) is a multicenter, phase IV, single arm trial. The primary objective was to explore whether clinical and biological factors are able to identify a subset of patients, treated with chemotherapy and bevacizumab, with a better prognosis in terms of progression-free or overall survival. Secondary aims were: (i) to describe the safety of bevacizumab added to carboplatin-paclitaxel chemotherapy, and (ii) to investigate the prognostic value of hypertension. Two further aims (to test the prognostic value of circulating and in situ biomarkers, and to describe the prevalence of the use of oral antidiabetic and antithrombotic therapy) are not reported here. Patient enrollment and data collection were managed through the web-platform of the coordinating center (http:\\www.usc-intnapoli.net). The study was approved by ethical committees at each participating center.
Study Population
Main eligibility criteria were: FIGO stage IIIB-IV, previously untreated epithelial ovarian cancer, performance status 0–2 according to ECOG (Eastern Cooperative Oncology Group), and no contraindications to bevacizumab administration (including history of thrombotic or hemorrhagic disorders, abdominal fistulae, perforations or abscesses in the previous 6 months, non-healing wounds, ulcers, or bone fractures). Patients should also have adequate organ function, no history of other tumors, nor mild to severe cardiovascular disease (including uncontrolled hypertension). Patients could not be enrolled if less than 28 days had elapsed from a surgical procedure. Availability of tumor samples for molecular analyses was mandatory. All the patients signed an informed consent before study entry. Histology was reviewed at the coordinating center.
Study Treatment
Patients received carboplatin (AUC (area under the curve) 5) and paclitaxel (175 mg/m2) plus bevacizumab (15 mg/kg) on day 1 for six 3-weekly cycles, followed by bevacizumab monotherapy (15 mg/kg), up to a maximum of 22 total cycles. Treatment could be discontinued prematurely in the case of disease progression, occurrence of unacceptable toxicity, or withdrawal of consent. Adverse events were managed according to local practice. Following disease progression, treatment choice was left at the discretion of the investigators and, eventually, the enrollment of the patients in the MITO16/MANGO–OV2/ENGOT–ov17 randomized trial of bevacizumab beyond progression was allowed.12
Sample Size Calculation
The study had 80% power to identify a potential prognostic marker that was able to select a favorable subgroup with an HR (hazard ratio)=0.60 and was expressed in at least 20% of the population. Considering a 0.01 α error, 280 events (either progression-free or overall survival) were required for the final analysis, and a sample size of 400 patients was planned.13
Statistical Analysis
Continuous variables were described with median values and IQR; other variables were expressed in terms of absolute numbers and proportions over the total number of patients enrolled. Progression-free survival was defined as the time elapsing from the registration to the first occurring event among death (from any cause) or detection of progressive disease. Patients alive without progressive disease at the time of database lock were considered progression-free at the time of the last known follow-up date. Overall survival was defined as the time elapsing between the registration of the patient and the date of death.
Median follow-up was estimated by reverse Kaplan-Meier curve, according to Schemper and Smith.14 Time to event curves (progression-free and overall survival) were calculated by the Kaplan-Meier method. The prognostic value of clinical factors was evaluated by a Cox model including age (<65 vs ≥65 years), stage (III vs IV), histology (high grade serous vs other), performance status (0 vs 1–2), residual disease (none, ≤1 cm, >1 cm, not operated), and center (based on the enrolment volume, with a threshold of 25 patients, which was defined post hoc to separate the centers into two groups, each enrolling around 50% of the patients). Schoenfeld residuals test was used to judge the proportional-hazards assumption. Consistent with the predefined α-level of 0.01, 99% CI were calculated. All efficacy analyses were performed on an intention-to-treat basis.
As for the study of the prognostic value of hypertension, patients were classified into three categories after baseline measurement of blood pressure (BP): (i) no hypertension (no antihypertensive treatment and systolic BP <120 mmHg and diastolic BP <80 mmHg), (ii) pre-hypertension (no antihypertensive treatment and systolic BP 120–139 mmHg or diastolic BP 80–89 mmHg), or (iii) on-AHT (ongoing antihypertensive treatment and systolic BP <140 mmHg and diastolic BP <90 mmHg). During treatment with bevacizumab, the occurrence of at least one of the following events was required to define the occurrence of hypertension: systolic BP ≥140 mmHg or diastolic BP ≥90 mmHg or medical intervention indicated and not previously ongoing. The cumulative incidence of hypertension during bevacizumab treatment according to baseline BP categories was described with Kaplan-Meier curves. The prognostic role of hypertension was first investigated by adding baseline BP to the Cox prognostic model described above. Subsequently, the prognostic value of the development of hypertension during bevacizumab was similarly tested by adding the time-dependent variable to the latter model. Survival curves according to the development of hypertension (yes or no) were calculated with the Simon and Makuch method to account for time-dependency.15
The safety population included patients who received at least one dose of the study treatment. Toxicity was graded according to Common Terminology Criteria for Adverse Events (CTCAE) 4.03. The response rate was evaluated according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria. All the analyses were performed with STATA/MP 14.1 (StataCorp LP, College Station, TX).
Results
From October 2012 to November 2014, 401 patients were registered. One patient was registered twice and the second entry was deleted. Of the remaining 400, two patients were excluded from analysis: one who withdrew consent immediately after the registration, and one for whom ovarian cancer diagnosis was not confirmed at a pathological review done at the enrolling center soon after registration (online supplemental figure A1). The main baseline characteristics of the 398 eligible patients are reported in Table 1. Median age was 59 years (IQR 49–66); 125 patients (31.5%) were receiving antihypertensive treatment at the time of enrolment.
Supplemental material
Efficacy
At a median follow-up of 32.3 months (IQR 24.1–40.4; online supplemental figure A2), 292 progression-free survival events and 136 deaths were registered. Median progression-free survival was 20.8 months (95% CI 19.1-22.0), and median overall survival was 41.1 months (95% CI 39.1-43.5). Progression-free and overall survival curves are plotted in online supplemental figure A3. An objective response was reported in 176 (68.7%) of 256 patients eligible for RECIST assessment, with 103 complete and 73 partial responses. Table 2 summarizes the results of multivariable prognostic factor analysis for progression-free and overall survival. Stage at diagnosis, ECOG performance status, and residual disease after initial surgery were independently associated with both progression-free and overall survival; in contrast, age, histology and size of center were not prognostic. Proportionality assumption was met for all covariates. Kaplan-Meier plots of progression-free and overall survival for the three significant prognostic factors are reported in the online supplemental figures A4–A6.
Hypertension
Figure 1 reports the cumulative incidence of hypertension (grade ≥2) according to baseline categories of BP. As expected, the cumulative incidence of grade ≥2 hypertension was dependent on baseline starting point: 40 (57.1%) of 70 patients who had no hypertension at baseline developed hypertension after a median time of 3.2 (IQR 1.4–6.7) months; 141 (70.9%) of 199 patients who had prehypertension at baseline developed hypertension after a median time of 2.5 (IQR 1.4–6.3) months; 111 (91.0%) of 122 patients who were receiving antihypertensive treatment at baseline (On-AHT) developed hypertension after a median time of 1.4 (IQR 0.7–4.7) months. Neither the baseline categories of BP (online supplemental figure A7) nor the development of hypertension during treatment (online supplemental figure A8) were associated with prognosis, considering both progression-free and overall survival (Table 3).
Safety
Data on adverse events for 391/398 patients are summarized in Table 4 and online supplemental table A1. There were two deaths without progression classified as possibly treatment related: one patient died 1 week after beginning the first cycle of treatment, due to a suspected pulmonary embolism; another patient suffered intestinal perforation after the fifth cycle of treatment, and died 45 days later due to further complications.
Adverse events of grade 3 or higher were reported in 302 of 391 (77.2%) patients. The most frequent adverse events were hypertension (grade 2 in 38.9% and grade 3 in 35.2%), neutropenia (grade ≥1 in 67.1%, grade ≥3 in 47.5%), fatigue (grade ≥1 in 58.0%, grade ≥3 in 3.0%), and anemia (grade ≥1 in 54.8%, grade ≥3 in 5.1%). Adverse events of special interest were proteinuria (grade ≥1 in 28.4%, grade ≥3 in 2%), thromboembolic events (grade ≥1 in 5.6%, grade ≥3 in 3.3%), fistulas (grade ≥1 in 1.3%, grade ≥3 in 0.8%), and bleedings (grade ≥1 in 3.3%, no event grade ≥3).
Treatment compliance
Two patients never started bevacizumab, due to rapid progression after the first cycle of therapy in which bevacizumab was not included due to recent surgery and vaginal bleeding. Two other patients never started paclitaxel. The median number of chemotherapy cycles was 6 (IQR 6–6); the median number of bevacizumab administrations was 6 (IQR 5–6) during the combination phase and 15.5 (IQR 9–16) during maintenance. Overall, 33 (8.3%) of 398 patients stopped treatment due to toxicity, and 11 (2.8%) of 398 patients interrupted for refusal. Administered relative dose intensities were 89% for both carboplatin and paclitaxel, and 82% and 89% for bevacizumab during the combination phase and maintenance, respectively (see Table A1 for details of drug exposure).
Discussion
Summary of Main Results
In the MITO-16A/MaNGO-OV2A phase IV study we found that the efficacy and safety of bevacizumab in combination with carboplatin and paclitaxel and as maintenance were comparable to previously available data, the risk of developing hypertension was higher for patients with hypertension at baseline, even if controlled, and the development of hypertension was not prognostic for progression-free and overall survival.
Results in the Context of Published Literature
The progression-free survival (20.7 months) in this trial was longer than in the registrative trials. Indeed, the median progression-free survival was 14.1 months in GOG-218, and 15.9 months in the high-risk population of the ICON-7 trial.6 7 Such difference clearly depends on the partially different populations enrolled in the three trials. The GOG-218 trial recruited 1873 women with stage III with residual disease after upfront surgery or stage IV epithelial ovarian cancer, randomizing them into three arms. Patients in the control arm received carboplatin AUC 6 and paclitaxel 175 mg/m2 every 3 weeks for six cycles plus placebo; patients in the two experimental arms also received bevacizumab (15 mg/kg) from cycle 2 to 6 (bevacizumab initiation arm), or from cycle 2 up to 21 cycles (bevacizumab throughout arm).6 In the ICON-7 trial, patients with early stage disease (FIGO stage I or IIA and clear-cell or grade 3 tumors) or advanced (FIGO stage IIB to IV) ovarian cancer, were randomly assigned to receive carboplatin AUC 5–6 and paclitaxel 175 mg/m2 every 3 weeks for six cycles with or without the addition of bevacizumab (7.5 mg/kg), given concurrently for five or six cycles and then continued for 12 additional cycles. This study recruited 1528 patients, but only 30% of patients had advanced disease at high risk of progression, that is, a FIGO stage III ovarian cancer with a residual disease >1 cm, or a FIGO stage IV ovarian cancer.7
Therefore, it is extremely difficult to compare the results of these two trials with MITO-16A/MaNGO-OV2A, where 56.5% of women had a residual disease ≤1 cm as compared with 34.7% patients in the GOG-218. More recently, preliminary results of the OTILIA study, an observational study that recruited a population similar to MITO-16A, were presented, reporting a median progression-free survival of 21.3 months.16
Standard maintenance therapy after platinum-based chemotherapy in patients with advanced ovarian cancer is not yet well defined. Several studies have demonstrated that bevacizumab might improve the outcome of patients with ovarian carcinoma molecular subtypes associated with the poorest survival, such as proliferative or mesenchymal.17 Although this is not part of the current study, the gene expression analysis in samples collected from patients enrolled in this trial is ongoing, to confirm the available information in the literature. At the same time, BRCA-mutated patients do not seem to benefit from anti-angiogenic treatment to the same degree as non-mutated patients.12 18
In our study, there were no unexpected adverse events, and only 32 patients (8%) interrupted treatment because of toxicity, a result that is comparable to OTILIA (7%) and lower than the GOG-218 (17% in bevacizumab throughout arm) study.6 16 In our study, the incidence of adverse events such as fistulae, thromboembolism, proteinuria, or neutropenia was comparable with ICON-7 and GOG-218, but, notably, a higher percentage of hypertension was recorded (38.9% grade 2 and 35.2% grade 3). This could be related to the enrolment of a population not strictly selected (indeed 50% of our patients had prehypertension at baseline, and a further 30.7% were on active antihypertensive treatment), but also to the more frequent measurement of BP (both before and after drug administration) planned in the MITO-16A/MaNGO-OV2A.
Lastly, neither hypertension at baseline nor development of grade ≥2 hypertension during treatment was prognostic in our study. Overall, our findings on hypertension are consistent with the interim analysis from OTILIA. In this trial, baseline hypertension was not associated with a difference in progression-free survival, while developing hypertension without the baseline condition was associated with a prolonged progression-free survival. Interestingly, in the same trial, the development or worsening of hypertension during treatment was more frequent among patients who already had hypertension at baseline (72%) than for those with no baseline hypertension (33%).16 In contrast, our study confirms that FIGO stage at diagnosis and residual disease after primary surgery are the main prognostic factors in newly diagnosed ovarian cancer.19 20
Strengths and Weaknesses
When the MITO-16A/MaNGO-OV2A study was designed, bevacizumab, although approved by the European Medicines Agency, was not available in Italy. Therefore, notwithstanding the obvious limitations of a single-arm non-comparative phase IV trial, this study represents a report of the first user’s experience of this treatment in Italy. However, this trial also represents an outlook on the treatment in clinical practice conditions of newly diagnosed ovarian cancer patients with bevacizumab plus chemotherapy.
Implications for Practice and Future Research
Translational analyses of MITO-16A/MaNGO-OV2A are ongoing and will possibly provide some insights into the molecular biomarkers that might influence the prognosis of patients treated with this schedule in first line. Genomic and proteomic profiling, gene expression to define the molecular subtypes, and angiogenic and immunological biomarker characterization are going to be defined in a collaborative network of 11 Italian research laboratories. Multi-omic bioinformatics analysis is also planned.
Conclusions
In conclusion, even with the limitations of a single-arm non-comparative study, MITO-16A/MaNGO-OV2A supports the efficacy and safety of the addition of bevacizumab to first line chemotherapy in a population of patients with newly diagnosed advanced ovarian cancer, in a setting similar to clinical practice. Ongoing translational analysis will hopefully clarify whether molecular prognostic biomarkers exist in this group of patients.
Data availability statement
Data of this study will be shared with publication upon reasonable and motivated request to the Principal Investigator of the study (s.pignata@istitutotumori.na.it). The following IPD will be available for sharing: baseline characteristics of patients, treatment data, safety data, follow-up data. There will be no time limit for data sharing.
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
FP and SP are co-last authorship.
Twitter @fperrone62
Contributors Gennaro Daniele, Ciro Gallo, Francesco Perrone and Sandro Pignata, wrote the protocol and the first draft of the manuscript. Ciro Gallo, Gennaro Daniele, Laura Arenare and Francesco Perrone did the statistical analyses. Francesco Raspagliesi, Giovanni Scambia, Carmela Pisano, Nicoletta Colombo, Simona Frezzini, Germana Tognon, Grazia Artioli, Angiolo Gadducci, Rossella Lauria, Annamaria Ferrero, Saverio Cinieri, Andrea De Censi, Enrico Breda, Paolo Scollo, Ugo De Giorgi, Andrea Alberto Lissoni, Dionissyos Katsaros, Domenica Lorusso, Vanda Salutari, Sabrina Chiara Cecere, Eleonora Zaccarelli, Margherita Nardin, Giorgio Bogani, Mariagrazia Distefano, Stefano Greggi, Maria Carmela Piccirillo, Roldano Fossati, and Gaia Giannone contributed to the results’ interpretation and to the final version of the manuscript.
Funding The MITO-16A/MaNGO-OV2A is a multicenter, national, academic trial sponsored by NCI Naples. Roche Italy provided bevacizumab and partial funding for trial activities and for the translational project. AIRC (Associazione Italiana per la Ricerca sul Cancro) supported translational studies with the IG 5776. AIOM (Associazione Italiana di Oncologia Medica) supported biomarker analysis with a grant. Roche, AIRC and AIOM did not play any role in study design, protocol writing, data collection, data analysis and interpretation, and manuscript writing. The corresponding author had full access to all of the data and the final responsibility to submit for publication.
Competing interests Dr Daniele reports travel and accommodations from Roche, outside the submitted work. Dr Raspagliesi reports grants from Roche, grants from GSK, grants from MSD, grants from Pharmamar, outside the submitted work. Dr Colombo reports personal fees from Roche, personal fees from Pharmamar, personal fees from AstraZeneca, personal fees from MSD, personal fees from Tesaro, personal fees from GSK, personal fees from Clovis, personal fees from Amgen, personal fees from Pfizer, personal fees from Biogen, personal fees from BIOCAD, outside the submitted work. Dr De Giorgi reports personal fees from Astellas - Pharma, personal fees from Bayer, personal fees and non-financial support from Bristol Myers-Squibb, personal fees and non-financial support from Ipsen, personal fees and non-financial support from Janssen, personal fees and non-financial support from Pfizer, grants and personal fees from Sanofi, personal fees from Novartis, personal fees from MSD, grants and non-financial support from Roche, grants from AstraZeneca, personal fees from Pharmamar, outside the submitted work. Dr Lorusso reports grants, personal fees and non-financial support from Pharmamar, grants and personal fees from MSD, grants and personal fees from Clovis, grants, personal fees and non-financial support from GSK, personal fees and non-financial support from AstraZeneca, personal fees from Merck Serono, personal fees from Amgen, non-financial support from Roche, outside the submitted work. Dr Salutari reports personal fees from MSD, personal fees from GSK, personal fees from Tesaro, personal fees from AstraZeneca, personal fees from Roche, personal fees from Eisai, personal fees from Clovis, outside the submitted work. Dr Piccirillo reports personal fees from Daichii Sankyo, personal fees from GSK, personal fees from MSD, grants from Roche, grants and personal fees from AstraZeneca, non-financial support from Bayer, outside the submitted work. Dr Perrone reports grants from Roche, during the conduct of the study; grants, personal fees and non-financial support from Bayer, personal fees from Sandoz, grants and personal fees from Incyte, personal fees from Celgene, grants and personal fees from AstraZeneca, personal fees from Pierre Fabre, personal fees from Janssen Cilag, grants from Roche, grants from Pfizer, outside the submitted work. Dr Pignata reports grants from Roche, during the conduct of the study; personal fees from AZ, personal fees from MSD, personal fees from Clovis, personal fees and non-financial support from Tesaro GSK, personal fees from Roche, outside the submitted work. The other authors do not declare conflicts of interests.
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.