Anti‐angiogenic therapy for persistent, recurrent and metastatic cervical cancer

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effectiveness and safety of angiogenesis inhibitors in the management of persistent, recurrent and metastatic cervical cancer.

Background

Description of the condition

Cervical cancer ranks as the fourth most common cancer and the fourth leading cause of cancer death in women, with an estimated 570,000 new cases and 311,000 deaths reported worldwide in 2018 (Bray 2018; Ferlay 2018). Almost 70% of the global burden falls in low‐and middle‐income countries, where 9 out of 10 cervical cancer deaths occur (Ferlay 2013). A woman's risk of developing cervical cancer to the age of 74 years is estimated to be 1.36% worldwide, ranging from 0.9% in more developed areas to 1.6% in less developed areas (Ferlay 2018; Torre 2015). The World Health Organization (WHO) recognises three pathological types of cervical cancer: squamous cell carcinoma (70% to 80%); adenocarcinoma (20% to 25%); and others, including adenosquamous carcinoma, neuroendocrine tumours and undifferentiated carcinoma (Marth 2017).

Early stage cancer (FIGO stage IA, IB1, IB2, and IIA1) can be treated by either surgery or chemoradiotherapy (Bhatla 2018; Marth 2017; NCCN 2018). For locally advanced disease (FIGO stage IB3 to IVA), standard treatment is concurrent platinum‐based chemoradiotherapy i.e. weekly cisplatin during radiotherapy (Bhatla 2018; Marth 2017; NCCN 2018). Generally, five‐year survival rates are 99% for stage IA1 and 98% for stage IA2; for stages IB to IIA, five‐year survival rates are 88% to 95% without lymph node metastasis and 51% to 78% with lymph node metastasis; for stage IIB rates are 65%, for stage III 40%, and for stage IV 10% to 20% (Cao 2014; Kim 2000). See FIGO staging 2018 (Bhatla 2018) Appendix 1.

Persistent, recurrent and metastatic diseases are the main causes of death in women with cervical cancer (Scatchard 2012). In persistent disease, visible lesions appear within six months after primary surgery, or within three to six months after primary radiotherapy; and in recurrent disease, visible lesions appear within an interval of more than six months after primary surgery, or three to six months after primary chemoradiotherapy (Cao 2011). More than 60% of recurrent disease is detected within two years after primary treatment, and about 10% after five years (Cao 2011). Persistent or recurrent cancers occur in about 35% of women with stages IB to IVA diseases, regardless of surgery or radiotherapy, giving five‐year survival rates of between 5% to 10% (Cao 2017). More than 60% of recurrent disease originates in the pelvic cavity, including the central pelvic cavity (uterine cervix, vagina, and uterine body), parametrium, and pelvic wall (Cao 2014). Women who experience distant metastases (i.e. cancer that has spread from the original (primary) tumour location to distant organs or distant lymph nodes, outside of the pelvis), either at initial diagnosis (FIGO stage IVB) or at recurrence, have poor prognoses. In recurrent disease, the metastatic rates (16% to 31% for IB and IIA, 26% for IIB, 39% for III, and 75% for IVA) increase with the FIGO stages (Cao 2017). The distant metastatic regions most often include lung (21%), bone (16%), para‐aortic region (11%), abdominal cavity (8%), and supraclavicular region (7%) (Cao 2014).

Description of the intervention

Angiogenesis, the development of new blood vessels, is essential for tumour initiation, growth, and metastasis (Tan 2010). Tumours require angiogenesis to receive enough nutrients and oxygen, to grow beyond 1 to 2 mm in diameter, and to facilitate metastasis (Wagner 2009). The process of tumour‐related angiogenesis is regulated by various pro‐angiogenic factors, such as vascular endothelial growth factor (VEGF), and their cognate receptors which are the dominant regulators of endothelial cells proliferation and new blood vessels formation (Tan 2010). Unlike chemotherapy agents that attack the tumour cells, main angiogenesis inhibitors are developed to block the VEGF signalling pathways (Jayson 2016). Angiogenesis inhibitors, anti‐angiogenic agents, starve the tumour by preventing blood vessels formation, and inhibit tumour growth and spread. Angiogenesis inhibitors have been tested in some solid tumours, including metastatic colorectal cancer, metastatic thyroid cancer, ovarian cancer, high‐grade glioblastoma, and endocrine refractory or resistant metastatic breast cancer (Gaitskell 2011; Khasraw 2014; Tan 2010; Wagner 2009; Wagner 2012). Phase I to III clinical trials of some angiogenesis inhibitors (bevacizumab, cediranib, pazopanib, sunitinib, and sorafenib) have been completed in women with persistent/recurrent, metastatic and locally advanced cervical cancer (Mackay 2010; Milosevic 2016; Monk 2009; Monk 2010; Schefter 2014; Symonds 2015; Tewari 2014; Zighelboim 2013), and bevacizumab has been approved by more than 60 countries for the treatment of persistent, recurrent, or metastatic disease, in combination with paclitaxel and cisplatin, or paclitaxel and topotecan, on the basis of improved overall survival (OS) with typical adverse events (hypertension, thromboembolic events, and gastrointestinal fistulae) in the Gynecologic Oncology Group (GOG) 240 trial (Tewari 2017).

The treatment plan for persistent, recurrent, or metastatic cervical cancer depends on the performance status of the person with cancer (i.e. their general well‐being), the disease site and extent of disease, and previous treatment (Bhatla 2018). For women with isolated distant metastases or locoregional recurrences (i.e. original microscopic cancer cells that have resisted primary therapy and become visible at the original site), treatment includes individualised radiotherapy (external radiation or brachytherapy (intracavity radiotherapy)), or both, with or without cisplatin chemotherapy, and surgical resection can be used for control of local disease and symptoms (NCCN 2018). For women with central pelvic recurrence after primary radiotherapy, pelvic exenteration (i.e. removal of part or all of the pelvic organs), with or without intraoperative radiotherapy, may be considered. Surgical mortality of less than 5% and long‐term survival rates of approximately 50% have been reported (Morley 1989). Radical hysterectomy or brachytherapy may also be considered if the central pelvic lesion is small (less than 2 cm) and isolated (NCCN 2018). For women with extrapelvic metastases or recurrence, who are unsuitable to undergo surgery or radiotherapy, combination cisplatin‐based chemotherapy regimens are often recommended (i.e. paclitaxel and cisplatin combined with bevacizumab) rather than the use of single agents (NCCN 2018). However, overall response rates to chemotherapy are about 20% (15% to 46%) and the median overall survival (OS) six to eight months and, for progression‐free survival (PFS), three months (Marth 2017; Moore 2006; Tewari 2005).

How the intervention might work

Over‐expression of VEGF has been found in most human cancers compared with normal tissues (Jayson 2016). VEGF is released by tumour cells and interacts with its receptor (VEGF‐R), a transmembrane protein, on nearby endothelial cells of normal blood vessels (see Gaitskell 2011 for figures describing receptor function). VEGF binds to the extracellular domain of VEGF‐R, stimulates tyrosine kinase activity in the intracellular domain of VEGF‐R, and triggers the downstream network of cell signalling pathways that promote endothelial cells proliferation and new blood vessels formation (Kerbel 2008). VEGF and VEGF‐R have already become the dominant targets for clinical angiogenesis inhibitors (Jayson 2016). VEGF signalling pathways can be blocked by several mechanisms, including blockade of VEGF (e.g. anti‐VEGF antibodies (bevacizumab), VEGF‐trap (aflibercept)), blockade of VEGF‐R (e.g. anti‐VEGF‐R antibodies (ramucirumab)), and interruption of intracellular signalling pathways (e.g. tyrosine kinase inhibitors (cediranib, pazopanib, sorafenib, sunitinib, etc.)) (Alldredge 2016). Results of xenograft models of human tumours in mice showed that angiogenesis inhibitors caused regression of microvascular tissues and reduction of metastatic disease progression (Avastin 2018; Morton 2012).

Bevacizumab (avastin) is a VEGF‐directed antibody and has been given as an intravenous infusion, monotherapy or with other therapies to treat cervical cancer in clinical trials. In a phase II study of 46 women with persistent or recurrent squamous cell carcinoma, single‐agent bevacizumab was active and well tolerated. The median OS was 7.29 months and the median PFS was 3.40 months, and the common severe adverse events were hypertension, thromboembolic events, and gastrointestinal disorders (Monk 2009). In a phase II study of 27 women with persistent or recurrent disease (66.6% were squamous), bevacizumab was used with cisplatin and topotecan, the median OS was 13.2 months and the median PFS was 7.1 months , and severe haematologic toxicities were common, including thrombocytopenia , leukopenia, anaemia, and neutropenia. Most women required at least one unexpected hospital admission for supportive care (Zighelboim 2013). In a phase II study of 49 women with locally advanced disease (63% were IIB, 80% were squamous), bevacizumab (10 mg/kg every 14 days for three cycles) was used with concurrent platinum‐based chemoradiotherapy, the 3‐year OS and PFS were 81.3% and 68.7% respectively, and no significant treatment‐related adverse events occurred (Schefter 2014). In the GOG 240 phase III study of 452 women with persistent, recurrent, or metastatic cervical cancer (68.6% were squamous), participants were randomised to receive bevacizumab with chemotherapy (paclitaxel and cisplatin, or paclitaxel and topotecan) or chemotherapy alone. The addition of bevacizumab to chemotherapy improved the median OS to 17.0 months versus 13.3 months, and PFS median to 8.2 months versus 5.9 months compared with chemotherapy alone; however, there was an increase in adverse events including hypertension, severe thromboembolic events, and severe gastrointestinal fistulas (Tewari 2014).

Cediranib (recentin) is a small molecule tyrosine kinase inhibitor of VEGF‐R. In a phase II study of 69 women with metastatic or recurrent disease, participants were randomised to receive cediranib with chemotherapy (paclitaxel and carboplatin) or chemotherapy alone. The addition of cediranib to chemotherapy improved the median PFS (8.1 months versus 6.7 months) as compared with chemotherapy alone, but did not improve the median OS (13.6 months versus 14.8 months), and increased adverse events including hypertension, febrile neutropenia, and diarrhoea (Symonds 2015).

Pazopanib (votrient) is an oral multi‐targeted tyrosine kinase inhibitor of VEGF‐R, platelet‐derived growth factor receptor (PDGF‐R), and c‐kit. Lapatinib is an oral dual tyrosine kinase inhibitor of epidermal growth factor receptor and human epidermal growth factor receptor 2. In a phase II study of 228 women with persistent, recurrent, or metastatic cervical cancer, participants were randomised to receive single‐agent pazopanib (800 mg once daily), single‐agent lapatinib (1500 mg once daily), or pazopanib plus lapatinib (this arm was terminated due to toxicity). Pazopanib improved the median OS (50.7 weeks versus 39.1 weeks) and PFS (18.1 weeks versus 17.1 weeks) as compared with lapatinib, and the most common adverse events were similar in both arms: diarrhoea, nausea, hypertension, and anorexia (Monk 2010).

Sunitinib (sutent) is an oral multi‐targeted tyrosine kinase inhibitor of VEGF‐R, PDGF‐R, and c‐kit. In a phase II study of 19 women with incurable locally advanced or metastatic disease, single‐agent sunitinib was administered in a 6‐week cycle for up to a maximum of six cycles; no tumour response was observed during the follow‐up (median 4.4 months) and five cases of fistula formation were observed (Mackay 2010).

Sorafenib (nexavar) is an oral multi‐targeted inhibitor that targets both VEGF‐R and PDGF‐R tyrosine kinase and Raf serine/threonine kinase. In a phase I study of 13 women with locally advanced disease, sunitinib was administered before and during concurrent platinum‐based chemoradiotherapy, but the study was terminated after a preliminary analysis because sorafenib increased tumour hypoxia, which was a strong predictor of disease progression (Milosevic 2016).

Why it is important to do this review

Persistent, recurrent, or metastatic cervical cancer is a refractory and symptomatic condition, and the purpose of current treatment is to relieve symptoms and improve quality of life (Bhatla 2018; Marth 2017; NCCN 2018). As opposed to conventional chemotherapy that kills the tumour cells, angiogenesis inhibitors are designed to block the tumour blood supply (Avastin 2018). Cochrane reviews of anti‐angiogenic therapy have been produced for some solid tumours (Gaitskell 2011; Khasraw 2014; Tan 2010; Wagner 2009; Wagner 2012). Clinical trials showed that angiogenesis inhibitors improved OS or PFS but increased tolerable toxicities in women with persistent, recurrent, or metastatic cervical cancer (Monk 2010; Symonds 2015; Tewari 2014). A Cochrane review is needed to evaluate the effectiveness and safety of angiogenesis inhibitors alone or adjuvant in the treatment of cervical cancer.

Objectives

To assess the effectiveness and safety of angiogenesis inhibitors in the management of persistent, recurrent and metastatic cervical cancer.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs). We will include studies reported as full‐text, those published as abstract only, and unpublished data. We will not include quasi‐randomised trials, cluster‐randomised trials, or cross‐over trials.

To evaluate the economics, full/partial economic evaluation studies and single effectiveness studies will be considered for inclusion. The review will consider only health economics studies conducted alongside effectiveness studies included in the effectiveness component of the review.

Types of participants

We will include adult women (aged 18 years or over) with persistent, recurrent, or metastatic cervical cancer. We will not include women with locally advanced disease. We will not include women with any other concurrent malignancy.

Types of interventions

• Angiogenesis inhibitors plus other treatments versus the other treatments only;

• Angiogenesis inhibitors versus other treatments;

• Angiogenesis inhibitors versus placebo or no treatment;

• Angiogenesis inhibitors versus another angiogenesis inhibitor.

Other treatments may include chemotherapy, radiotherapy, other molecular targeted therapy, or surgery.

Types of outcome measures

Primary outcomes

• Overall survival (OS), defined as the time interval from randomisation to death from any cause, or to last follow‐up;

• Special adverse events, including gastrointestinal perforations and fistulae, haemorrhage, thromboembolic events, hypertension.

Secondary outcomes

• Progression‐free survival (PFS), defined as the time interval from randomisation to progression of disease or death;

• Quality of life (QoL), measured using a scale that has been validated through reporting of norms in a peer‐reviewed publication, i.e. European Organization for Research and Treatment of Cancer (EORTC) QLQ‐CX24 cervical cancer‐specific quality of life questionnaire (Greimel 2006);

• Adverse events, classified according to CTCAE 2017;

• Economic evaluation, as defined by the individual studies, including full economic evaluation (cost‐benefit analysis, cost‐effectiveness analysis, and cost‐utility analysis), partial economic evaluation (cost analyses, cost‐description studies, and cost‐outcome descriptions), or single effectiveness evaluation (the description, measurement or valuation of resource use associated with interventions) (Shemilt 2011).

We will present a 'Summary of findings' table reporting the following outcomes listed in order of priority.

• Overall survival; follow‐up: 48 months;

• Total adverse events;

• Adverse events: gastrointestinal perforations and fistulae;

• Adverse events: haemorrhage;

• Adverse events: thromboembolic events;

• Adverse events: hypertension;

• Economic evaluation.

Search methods for identification of studies

There will be no language or date restrictions for our searches. We will search for papers in all languages and have them translated as necessary.

Electronic searches

We will search the following electronic databases:

• the Cochrane Central Register of Controlled Trials (CENTRAL), in The Cochrane Library;

• MEDLINE via Ovid (1946 to present);

• Embase via Ovid (1980 to present).

The MEDLINE search strategy is presented in Appendix 2. For databases other than MEDLINE, we will adapt the search strategy accordingly.

Economics search filter will not be used since most of the economics studies to be considered will be retrieved using the study design search filter (Shemilt 2011).

Searching other resources

All relevant articles will be identified on PubMed and we will make a further search for newly‐published articles using the 'related articles' feature.

Unpublished and grey literature

We will search the following for ongoing trials:

• ISRCTN ‐ metaRegister of Controlled Trials(http://www.isrctn.com/page/mrct);

• Physicians Data Query (http://www.cancer.gov/publications/pdq);

• ClinicalTrials.gov (http://www.clinicaltrials.gov);

• International Clinical Trials Registry Platform (ICTRP) (http://www.who.int/ictrp/en/).

If ongoing trials that have not been published are identified through these searches, we will approach the principal investigators, and major co‐operative groups active in this area, to ask for relevant data.

We will also search the reports of conferences in the Conference Proceedings Citation Index‐Science (CPCI‐S), and the bibliographic and abstracts in the NHS Economic Evaluation Database (NHS EED).

Handsearching

We will handsearch the citation lists of included studies, key textbooks, and previous systematic reviews and contact experts in the field to identify further reports of trials. We will also handsearch the reports of conferences from the following sources:

• Gynecologic Oncology (Annual Meeting of the American Society of Gynecologic Oncology);

• International Journal of Gynecological Cancer (Annual Meeting of the International Gynecologic Cancer Society);

• British Journal of Cancer;

• British Cancer Research Meeting;

• Annual Meeting of European Society of Medical Oncology (ESMO);

• Annual Meeting of the American Society of Clinical Oncology (ASCO).

Data collection and analysis

Selection of studies

We will download all titles and abstracts retrieved by electronic searching to a reference management database (Endnote) and remove duplicates. Two review authors (YC and XZ) will examine the remaining references independently. We will exclude those studies that clearly do not meet the inclusion criteria, and obtain copies of the full‐text of potentially relevant references. Two review authors (YC and XZ) will independently assess the eligibility of the retrieved reports/publications. We will resolve any disagreement through discussion or, if required, we will consult a third person (GD). We will identify and exclude duplicate reports and collate multiple reports of the same study so that each study rather than each report is the unit of interest in the review. We will record the selection process in sufficient detail to complete a PRISMA flow diagram and a 'Characteristics of excluded studies' table (Liberati 2009).

Data extraction and management

Two review authors (YC and XZ) will independently extract study characteristics and outcome data from included studies on to a piloted data collection form. We will note in the 'Characteristics of included studies' table if outcome data were not reported in a usable way. We will resolve disagreements by consensus or by involving a third person (GD). One review author (GD) will transfer data into Review Manager 2014. We will double‐check that data has been entered correctly by comparing the data presented in the systematic review with the study reports. A second review author (YL) will 'spot‐check' the accuracy of the study characteristics against the trial report.

For included studies, we will extract the following data:

• Author, year of publication and journal citation (including language);

• Country;

• Setting;

• Inclusion and exclusion criteria;

• Study design, methodology;

• Study population:

  • total number enrolled

  • participant characteristics

  • age

  • co‐morbidities (e.g. gastrointestinal diseases, cardiovascular diseases)

  • stage, grade, histology of cervical cancer

  • details about persistent/recurrent and metastatic disease

    • disease site and extent

    • PFS after primary treatment

    • PFS after initial diagnosis of stage IVb disease.

    • previous treatment

  • Other baseline characteristics;

• Intervention details

  • type of angiogenesis inhibitor

  • dose of angiogenesis inhibitor

  • treatment duration of angiogenesis inhibitor

  • type of combination treatment;

• Comparison

  • type of control: conventional treatment or no treatment

  • details of comparison treatment

• Risk of bias in study (see below);

• Duration of follow‐up;

• Outcomes: for each outcome, we will extract the outcome definition and unit of measurement (if relevant). For adjusted estimates, we will record variables adjusted for in analyses;

• Results: we will extract the number of participants allocated to each intervention group, the total number analysed for each outcome, and the missing participants;

• Notes: funding for trial, and notable conflicts of interest of trial authors.

Results will be extracted as follows:

• For time‐to‐event data (survival and disease progression), we will extract the log of the hazard ratio [log(HR)] and its standard error from trial reports. If these are not reported, we will attempt to estimate the log (HR) and its standard error using the methods of Parmar 1998.

• For dichotomous outcomes (e.g. adverse events or deaths, if it is not possible to use a HR) we will extract the number of participants in each treatment arm who experienced the outcome of interest and the number of participants assessed at end point, in order to estimate a risk ratio (RR).

• For continuous outcomes (e.g. QoL measures), we will extract the final value and standard deviation of the outcome of interest and the number of participants assessed at end point in each treatment arm at the end of follow‐up, in order to estimate the mean difference between treatment arms and its standard error.

If reported, we will extract both unadjusted and adjusted statistics. If available, we will use adjusted results, otherwise we will use unadjusted results.

Where possible, all data extracted will be those relevant to an intention‐to‐treat analysis, in which participants will be analysed in the groups to which they were assigned.

We will note the time points at which outcomes were collected and reported.

In the event of duplicate publications, companion documents or multiple reports of a primary study, we will maximise yield of information by collating all available data and use the most complete data set aggregated across all known publications. In case of doubt, we will give priority to the publication reporting the longest follow‐up associated with our primary or secondary outcomes.

Assessment of risk of bias in included studies

We will assess and report on the methodological risk of bias of included studies in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), which recommends the explicit reporting of the following individual elements for RCTs.

1. Selection bias: random sequence generation and allocation concealment;

2. Performance bias: blinding of participants and personnel;

3. Detection bias: blinding of outcome assessment;

4. Attrition bias: incomplete outcome data;

5. Reporting bias: selective reporting of outcomes;

6. Other possible sources of bias.

Two review authors (YC and XZ) will apply the 'Risk of bias' tool independently and resolve differences by discussion or by appeal to a third review author (GD). We will judge each item as being at high, low or unclear risk of bias as set out in the criteria provided by Higgins 2011, and provide a quote from the study report or a statement as justification for the judgement for each item in the 'Risk of bias' table or both. We will summarise results in both a 'Risk of bias' graph and a 'Risk of bias' summary. When interpreting treatment effects and meta‐analyses, we will take into account the risk of bias for the studies that contribute to that outcome. Where information on risk of bias relates to unpublished data or correspondence with a trialist, we will note this in the 'Risk of bias' table. See 'Risk of bias' items (Appendix 3).

We defined the following end points as subjective outcomes: PFS, QoL.

We defined the following end points as objective outcomes: OS, adverse events (gastrointestinal perforations and fistulae, haemorrhage, thromboembolic events, hypertension), economic evaluation.

Measures of treatment effect

We will use the following measures of the effect of treatment.

• For time‐to‐event data, we will use the HR, if possible.

• For dichotomous outcomes, we will analyse data based on the number of events and the number of women assessed in the intervention and comparison groups. We will use these to calculate the RR and 95% confidence interval (CI).

• For continuous outcomes, we will analyse data based on the mean, standard deviation (SD) and number of women assessed for both the intervention and comparison groups to calculate mean difference (MD) between treatment arms with a 95% CI. If the MD is reported without individual group data, we will use this to report the study results. If more than one study measures the same outcome using different tools, we will calculate the standardised mean difference (SMD) and 95% CI using the inverse variance method in Review Manager 5 (Review Manager 2014).

We will undertake meta‐analyses only where this is meaningful, i.e. if the treatments, participants and the underlying clinical question are similar enough for pooling to make sense. We will describe skewed data that are reported as medians and interquartile ranges narratively. Where multiple trial arms are reported in a single trial, we will include only the relevant arms.

Unit of analysis issues

We do not anticipate unit of analysis issues.

Dealing with missing data

For missing outcome or summary data, we will not impute missing data and we will report any assumptions we make in the review. We will attempt to contact study authors to obtain missing data (participants, outcomes, or summary data). We will report on the levels of loss to follow‐up and assess this as a source of potential bias.

Assessment of heterogeneity

Where studies are considered similar enough to allow pooling of data using meta‐analysis, we will assess the degree of heterogeneity by visual inspection of forest plots, by estimation of the percentage heterogeneity (I² measurement) between trials which cannot be ascribed to sampling variation (Higgins 2003), by a formal statistical test of the significance of the heterogeneity (Chi²) (Deeks 2001) and, if possible, by subgroup analyses. We will regard heterogeneity to be substantial if I² is greater than 30% and either T² is greater than zero, or there is a low P value (< 0.10) in the Chi² test for heterogeneity.

If there is evidence of substantial clinical, methodological or statistical heterogeneity across included studies, we will not pool results for meta‐analysis but instead will use a narrative approach to our data synthesis. In this event, we will investigate and report the possible clinical or methodological reasons for this level of heterogeneity. Although, statistical heterogeneity is a post hoc measure, we will discuss and pool trials if they are deemed similar enough to conduct the meta‐analysis.

Assessment of reporting biases

If we include 10 or more studies that investigate a particular outcome, we will examine funnel plots corresponding to meta‐analysis of the outcome to assess the potential for small study effects such as publication bias. We plan to assess funnel plot symmetry visually, and, if asymmetry is suggested, we will perform exploratory analyses to investigate it.

Data synthesis

If a sufficient number of clinically similar studies (in terms of participants, settings, intervention, comparison and outcome measures) are available to ensure meaningful conclusions, we will pool their results in Review Manager 5 (Review Manager 2014. We will synthesize the outcomes using a random‐effects model with inverse variance for meta‐analysis to incorporate heterogeneity (statistical or clinical, or both) among studies (DeSimonian 1986).

• For time‐to‐event data, we will pool HRs using the generic inverse variance facility in Review Manager 2014.

• For any dichotomous outcomes, we will calculate the RR for each study and these will then be pooled.

• For continuous outcomes, we will calculate the summary MD between the treatment arms at the end of follow‐up, if all trials measure the outcome on the same scale; otherwise, we will calculate the summary SMD.

If any trials have multiple treatment groups, we will divide the 'shared' comparison group into the number of treatment groups and treat the split comparison group as independent comparisons.

If we are unable to pool the data statistically using meta‐analysis, we will conduct a narrative synthesis of results. We will present the major outcomes and results, organised by intervention categories according to the major types or aims of the identified interventions, or both. Depending on the assembled research, we may also explore the possibility of organising the data by population. Within the data categories, we will explore the main comparisons of the review.

Assessing the certainty of the evidence

We will present the overall certainty of the evidence for prespecified outcomes (Types of outcome measures), including OS, adverse events, and economic evaluation, according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, which takes into account issues not only related to internal validity (risk of bias, inconsistency, imprecision, publication bias) but also to external validity such as directness of results (Langendam 2013). We will create a 'Summary of findings' table (Appendix 4) based on the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and using GRADEpro GDT. We will use the GRADE checklist and GRADE Working Group certainty of evidence definitions (Meader 2014). We will downgrade the evidence from 'high' certainty by one level for serious (or by two for very serious) concerns for each limitation:

• High‐certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate‐certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

• Low‐certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

• Very low‐certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

If meta‐analysis is not possible, we will present results in a narrative 'Summary of findings' table format.

Subgroup analysis and investigation of heterogeneity

We will perform subgroup analyses for the following factors:

• different angiogenesis inhibitors (anti‐VEGF antibodies, VEGF trap, anti‐VEGF‐receptor antibodies, tyrosine kinase inhibitors, and others).

• different histologic types (squamous, adenocarcinoma, and others).

We will consider factors such as age, stage, type of intervention, length of follow‐up and risk of bias status in interpretation of any heterogeneity.

Sensitivity analysis

Sensitivity analyses will be performed excluding studies at high risk of bias and using unadjusted results.

What's new

Date	Event	Description
14 April 2020	Amended	Sources of support updated.

Appendices

Appendix 1. International Federation of Gynecology and Obstetrics (FIGO) staging of cancer of the cervix uteri (2018)

Stage	Description
I	The carcinoma is strictly confined to the cervix (extension to the uterine corpus should be disregarded)
IA	Invasive carcinoma that can be diagnosed only by microscopy, with maximum depth of invasion < 5 mma
IA1	Measured stromal invasion < 3 mm in depth
IA2	Measured stromal invasion ≥ 3 mm and < 5 mm in depth
IB	Invasive carcinoma with measured deepest invasion ≥ 5 mm (greater than Stage IA), lesion limited to the cervix uterib
IB1	Invasive carcinoma ≥ 5 mm depth of stromal invasion, and < 2 cm in greatest dimension
IB2	Invasive carcinoma ≥ 2 cm and < 4 cm in greatest dimension
IB3	Invasive carcinoma ≥ 4 cm in greatest dimension
II	The carcinoma invades beyond the uterus, but has not extended onto the lower third of the vagina or to the pelvic wall
IIA	Involvement limited to the upper two‐thirds of the vagina without parametrial involvement
IIA1	Invasive carcinoma < 4 cm in greatest dimension
IIA2	Invasive carcinoma ≥ 4 cm in greatest dimension
IIB	With parametrial involvement but not up to the pelvic wall
III	The carcinoma involves the lower third of the vagina and/or extends to the pelvic wall and/or causes hydronephrosis or nonfunctioning kidney and/or involves pelvic and/or para‐aortic lymph nodesc
IIIA	The carcinoma involves the lower third of the vagina, with no extension to the pelvic wall
IIIB	Extension to the pelvic wall and/or hydronephrosis or nonfunctioning kidney (unless known to be due to another cause)
IIIC	Involvement of pelvic and/or para‐aortic lymph nodes, irrespective of tumour size and extent (with r and p notations)c
IIIC1	Pelvic lymph node metastasis only
IIIC2	Para‐aortic lymph node metastasis
IV	The carcinoma has extended beyond the true pelvis or has involved (biopsy‐proven) the mucosa of the bladder or rectum. (A bullous oedema, as such, does not permit a case to be allotted to Stage IV)
IVA	Spread to adjacent pelvic organs
IVB	Spread to distant organs

When in doubt, the lower staging should be assigned.

^aImaging and pathology can be used, where available, to supplement clinical findings with respect to tumour size and extent, in all stages.

^bThe involvement of vascular/lymphatic spaces does not change the staging. The lateral extent of the lesion is no longer considered.

^cAdding notation of r (imaging) and p (pathology) to indicate the findings that are used to allocate the case to Stage IIIC. Example: If imaging indicates pelvic lymph node metastasis, the stage allocation would be Stage IIIC1r, and if confirmed by pathologic findings, it would be Stage IIIC1p. The type of imaging modality or pathology technique used should always be documented.

Appendix 2. MEDLINE search strategy

1. exp Uterine Cervical Neoplasms/

2. (cervi* adj5 (cancer* or tumor* or tumour* or neoplas* or carcinoma* or adenocarcinoma* or malignan*)).mp.

3. 1 or 2

4. exp Angiogenesis Inhibitors/

5. (angiogenesis adj5 inhibit*).mp.

6. exp Vascular Endothelial Growth Factors/

7. vascular endothelial growth factor*.mp.

8. VEGF.mp.

9. (VEGFR or VEGF‐R).mp.

10. exp Antibodies, Monoclonal/

11. monoclonal antibodies.mp.

12. (bevacizumab or avastin).mp.

13. (VEGF‐Trap or aflibercept or AVE0005).mp.

14. (ramucirumab or gyramza).mp.

15. exp Protein‐Tyrosine Kinases/

16. (tyrosine kinase adj5 inhibit*).mp.

17. (sorafenib or nexavar or BAY 43‐0006 or NSC724772).mp.

18. (cediranib or AZD2171 or recentin).mp.

19. (sunitinib or sutent or SU11248).mp.

20. (pazopanib or votrient or GW‐786034).mp.

21. (nintedanib or vargatef or BIBF 1120).mp.

22. (regorafenib or stivarga or BAY 73‐4506).mp.

23. (axitinib or lnlyta or AG‐013736).mp.

24. (vandetanib or caprelsa or ZD6474).mp.

25. (lenvatinib or lenvima or E7080).mp.

26. 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25

27. 3 and 26

28. randomized controlled trial.pt.

29. controlled clinical trial.pt.

30. randomized.ab.

31. placebo.ab.

32. drug therapy.fs.

33. randomly.ab.

34. trial.ti.

35. groups.ab.

36. 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35

37. (animals not (humans and animals)).sh.

38. 36 not 37

39. 27 and 38

Key

mp = title, original title, abstract, name of substance word, subject heading word
 ab = abstract
 pt = publication type
 fs = floating subheading

Appendix 3. 'Risk of bias' items

• Random sequence generation

  1. Low risk of bias e.g. participants assigned to treatments on basis of a computer‐generated random sequence or a table of random numbers

  2. High risk of bias e.g. participants assigned to treatments on basis of date of birth, clinic identity‐number or surname, or no attempt to randomise participants

  3. Unclear risk of bias e.g. not reported, information not available

• Allocation concealment

  1. Low risk of bias e.g. where the allocation sequence could not be foretold

  2. High risk of bias e.g. allocation sequence could be foretold by women, investigators or treatment providers

  3. Unclear risk of bias e.g. not reported

• Blinding of participants and personnel

  1. Low risk of bias, if participants and personnel were adequately blinded

  2. High risk of bias, if participants were not blinded to the intervention that the participant received

  3. Unclear risk of bias, if this was not reported or unclear

• Blinding of outcomes assessors

  1. Low risk of bias, if outcome assessors were adequately blinded

  2. High risk of bias, if outcome assessors were not blinded to the intervention that the participant received

  3. Unclear risk of bias, if this was not reported or unclear

• Incomplete outcome data: we will record the proportion of participants whose outcomes were not reported at the end of the study. We will code a satisfactory level of loss to follow‐up for each outcome as:

  1. Low risk of bias, if fewer than 20% of women were lost to follow‐up and reasons for loss to follow‐up were similar in both treatment arms, or missing data had been imputed using appropriate methods (at least 80% women were assessed in the analysis)

  2. High risk of bias, if more than 20% of women were lost to follow‐up or reasons for loss to follow‐up differed between treatment arms, or potentially inappropriate application of simple imputation

  3. Unclear risk of bias if loss to follow‐up was not reported

• Selective reporting of outcomes

  1. Low risk of bias e.g. review reports all outcomes specified in the protocol

  2. High risk of bias e.g. It is suspected that outcomes had been selectively reported

  3. Unclear risk of bias e.g. It is unclear whether outcomes had been selectively reported

• Other bias

  1. Low risk of bias, if you did not suspect any other source of bias and the trial appeared to be methodologically sound

  2. High risk of bias, if you suspected that the trial was prone to an additional bias

  3. Unclear risk of bias, if you were uncertain whether an additional bias might have been present

Appendix 4. Draft Summary of findings table

Angiogenesis inhibitors compared to placebo for the treatment of persistent, recurrent, or metastatic cervical cancer
Patient or population: adult women (aged 18 years or over) with persistent, recurrent, or metastatic cervical cancer Settings: hospital Intervention: treatment with angiogenesis inhibitors Comparison: standard care, placebo or none.
Outcomes	Illustrative comparative risks*		Relative effect (95% CI)	No of participants (studies)	Quality of evidence (GRADE)	Comment
	Assumed risk	Corresponding risk
Overall survival Follow‐up: 48 months
Total adverse events
Adverse events Gastrointestinal perforations and fistulae
Adverse events Haemorrhage
Adverse events Thromboembolic events
Adverse events Hypertension
Economic evaluation
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; HR: hazard ratio; MD: mean difference; RR: risk ratio; OR: odds ratio
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

Contributions of authors

Yunhai Chuai (YC): drafted the protocol.
 YC and Xia Zhang (XZ): will develop and run the search strategy.
 Guanghai Dai (GD) and Aiming Wang (AW) and Ying Li (YL): will obtain copies of studies.
 YC and XZ: will select which studies to include.
 YC and XZ: will extract data from studies.
 GD and AW and YL: will enter data into RevMan.
 GD and AW and YL: will carry out the analysis.
 YC and XZ: will interpret the analysis.
 YC: will draft the final review.
 YC and XZ: will update the review.

Sources of support

Internal sources

• Chinese PLA General Hospital, Medical School of Chinese PLA, China.

  Financial support: salary of the team
 Principal Investigator: Xia Zhang is the first lead researcher for this Cochrane project.
 Sub‐Investigator: Yunhai Chuai, Aiming Wang, Ying Li, and Guanghai Dai

External sources

• Cochrane Gynaecological, Neuro‐oncology and Orphan Cancers, UK.

  Editorial support

Declarations of interest

Yunhai Chuai (YC): none known
 Aiming Wang (AW): none known
 Ying Li (YL): none known
 Guanghai Dai (GD): none known
 Xia Zhang (XZ): none known

Edited (no change to conclusions)