Vascular endothelial growth factor (VEGF) inhibitors for the treatment of metastatic and recurrent cervical cancer

Objectives

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

To evaluate the benefits and harms of anti‐angiogenesis agents for the treatment of metastatic and recurrent cervical cancer.

Background

Historically, women with metastatic or recurrent cervical cancer have had limited treatment options with poor prognosis. New anti‐angiogenesis therapies act by inhibiting the growth of new blood vessels and thereby restricting tumour growth by cutting off their nutrient supply. These agents offer an alternative strategy to conventional chemotherapy in the clinical management of these women.

Description of the condition

Cervical cancer is the second most commonly diagnosed cancer to affect women worldwide and the third‐leading cause of cancer death amongst women in low‐income countries (Ferlay 2019; Siegel 2019; Torre 2015). Since the introduction of cytology‐based cervical screening, cervical cancer incidence and mortality rates have been reduced by approximately 75% in countries with co‐ordinated population vaccination and screening programmes (Cancer Research UK 2014; ISD Scotland 2011; Siegel 2019; Welsh Cancer Intelligence and Surveillance 2011).

The majority of cervical cancer cases (60%) are associated with inadequate screening and are therefore potentially preventable, or may be diagnosed at a very early stage, whilst still easily curable with surgery (Alldredge 2016; Eskander 2014; Rebolj 2019). However, even in women diagnosed at a relatively early stage who have undergone treatment with either surgery or radical chemoradiotherapy, the risk of recurrent disease is 10% to 20% for stages IB to IIA, and 28% to 64% in women with locally advanced FIGO stages IIB to IV, as per International Federation of Obstetrics and Gynaecology (FIGO) classification (see Figure 1 for FIGO cervical cancer staging) (Bhatla 2018; Bray 2018; Cancer Research UK 2014).

1.

FIGO stage 2018.

Very advanced cervical cancer (FIGO stage IVB), where disease has spread to areas outside of the pelvis at diagnosis (metastatic disease), has a very poor prognosis. Treatment options are limited and are unlikely to be curative, therefore management is aimed at prolonging life and managing symptoms. Similarly, in recurrent disease, unless exenterative surgery is possible, women are usually treated with palliative chemotherapy. Chemotherapy aims to prolong survival, provide symptom control, and improve quality of life. In patients not medically fit enough for chemotherapy, treatment options in both metastatic and recurrent disease may be limited to best supportive care, with the option of palliative radiotherapy, if appropriate (Eleje 2019; Fanfani 2016; Greer 2010; Matsuo 2015; Waggoner 2003).

Three phase III randomised trials (GOG‐204, GOG‐240, and Japanese Clinical Oncology Group 0505 (JCOG‐0505) have examined chemotherapy regimens for women with advanced and recurrent cervical cancer (Monk 2009; Saito 2010; Tewari 2014). The main concern was acquired resistance to platinum in patients with recurrent cervical cancer previously treated with chemotherapy. However, the regimens that did not include a platinum agent were not superior to cisplatin‐based treatments, therefore cisplatin plus paclitaxel remains the chemotherapy standard for recurrent and metastatic disease (Eleje 2019; Monk 2014).

Angiogenesis inhibitors are a class of promising biological agents that target the process by which a tumour forms new blood vessels to supply it with oxygen and nutrients. This process is key to cervical cancer growth, both locally and for metastatic deposits (Willmott 2009).

In 1971, Dr Folkman proposed a tumour angiogenesis‐dependent doctrine, which suggested that development of tumours was due not only to the ability of the individual cancer cells to grow without normal controls, but was also dependent on nutrient and oxygen supply via blood vessels (Folkman 1971). The transition of a tumour from an avascular to a vascular phase was termed the 'angiogenic switch' and was thought to be an important factor in the clinical progression from tumour dormancy to cancer invasion (Hanahan 1996). The angiogenic switch occurs through a dynamic process involving pro‐angiogenic and anti‐angiogenic factors that are normally maintained in balance (physiological homeostasis). In cancer there is an up‐regulation of pro‐angiogenic factors, such as vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), platelet‐derived growth factor (PDGF), and angiopoietins, and a down‐regulation of anti‐angiogenic factors, such as thrombospondin, angiostatin, and endostatin, leading to the development of new blood vessels (neovascularisation) and tumour progression (Hicklin 2005). Neovascularisation is therefore a vital process in metastatic spread to allow malignant cells to enter into the circulation (Falzone 2019; Folkman 1971; Folkman 1990; Pfaendler 2016).

Activation of VEGF promotes the proliferation and migration of endothelial cells, which line blood vessels, and the development of new blood vessels. Moreover, VEGF activation enhances the leakiness (permeability) of existing blood vessels and, as a consequence, causes increased leakage of multiple plasma proteins that are involved in the angiogenesis process (Nagy 2007; Turajlic 2018). These proteins are ligands to three VEGF receptors (VEGFR 1 to 3) and are responsible for initiating a cascade of reactions that promote migration, proliferation, and survival of endothelial cells, leading to the formation of new blood vessels. VEGFR‐1 and VEGFR‐2 are responsible for regulation of angiogenesis, whilst VEGFR‐3 plays a secondary role in angiogenesis, but it is essential for lymphangiogenesis (development of new lymphatic vessels) and is therefore involved in the production of ascites and lymphatic dissemination of metastatic factors (Ellis 2008).

Angiogenesis is present in pre‐malignant conditions of the cervix, as well as in invasive disease, which can be seen at colposcopy with the occurrence of vascular changes. Studies have shown that microvessel (small vessels) density increases progressively with grades of cervical intraepithelial neoplasia (CIN), suggesting that angiogenesis may be an early event in the development of cervical pre‐cancer and cancer (Dobbs 1997; Guidi 1994; Krill 2015; Smith‐McCune 1994). Cervical cancer has been shown to be more dependent on angiogenesis than most other solid cancers. This is because human papillomavirus (the cause of almost all cervical cancers) and hypoxia, with increased hypoxia‐inducible factor 1 alpha, are associated with higher levels of VEGF (Monk 2010 (a)). In addition, increased intratumoural microvessel density during histological examination and enhanced expression of proteins, such as the endothelial antigen CD31, have been associated with poor prognosis (Dellas 1997; Dobbs 1997; Tjalma 1999). Targeting the cancer micro‐environment, specifically new blood vessel formation, using VEGF pathway inhibition is the main goal of angiogenesis inhibition.

Description of the intervention

Anti‐angiogenesis factors

Anti‐angiogenic factors include inhibitors targeting either single or multiple angiogenic pathways. VEGF acts by binding to its receptor, which then signals within the cell via tyrosine kinase enzyme activity (tyrosine kinases). VEGF receptor‐1 (Flt‐1 or fms‐like‐tyrosine kinase) and VEGFR‐2 (KDR/flk‐1) belong to the best characterised angiogenesis signalling pathway and can be blocked by inhibitors, such as antibodies (bevacizumab) or soluble receptors, which bind to VEGF and prevent it from binding to its receptor on cells. VEGFR‐2 is thought to be the dominant receptor responsible to translate and regulate angiogenesis (Bellone 2007).

In 1993, it was shown that anti‐VEGF monoclonal antibodies exerted a potent inhibitory effect on the growth of three tumour cell lines injected subcutaneously into mice(Kim 1993). Interestingly, the antibody had no effects on the cell lines in vitro. However, several parallel studies confirmed in vivo growth inhibition, which correlated with decreased tumour microvessel density and inhibition of tumour angiogenesis (Borgstrom 1996; Borgstrom 1998; Borgstrom 1999).

Bevacizumab is a humanised monoclonal antibody directed against all major isoforms of VEGF (VEGF‐A, VEGF‐B, VEGF‐C, VEGF‐D, placental growth factor (PIGF), VEGF‐E, and VEGF‐F members), binding to VEGF and blocking its binding to the receptor VEGFR. This inhibits endothelial cell proliferation and the development of vessels (Ferrara 2004). Its use has been evaluated in many solid cancers. It is used extensively in the USA and is the most studied biological agent in gynaecological cancer (Grothey 2009; Tewari 2015 (a)). Several randomised controlled trials (RCTs) and non‐randomised trials have been published reporting the use of the binding and inactivation of VEGF by clinical use of bevacizumab as single agent (Monk 2009; Monk 2009 (a); Monk 2010; Monk 2010 (a); Schefter 2012; Schefter 2014; Takano 2013; Tan 2010; Tewari 2005; Tewari 2014; Tewari 2014 (a); Wright 2006).

Tyrosine kinase inhibitors

Tyrosine kinase inhibitors (TKI) bind to the ATP‐binding catalytic site of the tyrosine kinase domain of VEGF receptors (VEGFRs) and directly inhibit the activity of VEGFR.

Sorafenib is an inhibitor of VEGFR‐1, VEGFR‐2, VEGFR‐3, platelet‐derived growth factor receptor (PDGFR‐beta), and Raf‐1 tyrosine kinase activity.

Cediranib blocks VEGF signalling by inhibiting VEGFR1‐2‐3 tyrosine kinases as well as PDGFR‐alpha and stem cell factor receptor (c‐Kit) (Tewari 2015; Tewari 2015 (a); Tewari 2015 (b)).

Sunitinib blocks tumour proliferation and angiogenesis by acting on the intracellular kinase domains of the VEGF receptor isoforms (VEGFR‐1, VEGFR‐2, and VEGFR‐3) and on the PDGFR, c‐Kit, and class III tyrosine kinase (RTK) and fms‐inhibitor like tyrosine kinase III (FLT3) (Mendel 2003).

Pazopanib is a multi‐target TKI of VEGFR‐1, ‐2, and ‐3; PDGFR‐alpha and ‐beta; and c‐Kit (Hurwitz 2009).

How the intervention might work

VEGF‐targeted therapy can have an effect on number of blood vessels and their function as well as directly on tumour cells. It has been reported that vessel beds of normal adult tissues are quiescent and in a VEGF‐independent state. In contrast, the immature vessels in tumours are more susceptible to VEGF‐targeted therapy (Kerbel 2008).

Structural irregularities and insufficient vascular supply are common in tumour vasculature, which is thought to be due mainly to lack of oxygen, which then causes increased expression of hypoxia‐inducible factor‐1. This factor induces the expression of a larger number of genes that are involved in tumour progression and metastasis (Forsythe 1996), the most important of which is VEGF (Pennacchietti 2003). Over‐expression of VEGF is responsible for vascular hyperpermeability (leakiness), which may render the cancer less sensitive to therapies (Carpini 2010; Seebacher 2019;

Some animal studies support the idea that treatment with VEGF‐targeted agents encourages intratumoural delivery of therapeutic agents by causing vasoconstriction and flow dynamic changes (Wildiers 2003). The findings demonstrated that anti‐VEGF therapy changes not only vascular function, but also inhibits classical angiogenesis (Wildiers 2003).

Activation of VEGFR‐2 by VEGF also leads to expression of vasodilator factors, hence blocking VEGF signalling, which leads to a reduction in vasodilator factors This hypothesis of vascular normalisation fits well with the observation that bevacizumab monotherapy has produced only modest symptomatic release clinical benefits, but that it increases survival when used in combination with chemotherapy (Tewari 2014).

Finally, VEGF‐targeted therapy may have direct inhibitory effects on tumour cells. Studies on breast cancer cell lines reported a decrease in chemotactic migration when the VEGF was blocked (Nasarre 2003; Nasarre 2005). Studies in colon cancer and pancreatic cancer cell lines have shown that the activation of VEGFR‐1 leads to an increase in tumour cell invasion and migration, which was blocked by treatment with antibodies to VEGFR‐1 (Fan 2005; Liu 2015; Wey 2005). (See also Table 1 for drug actions.)

1. Actions of included drugs.

Name of drug	Actions
Bevacizumab	VEGF‐A‐B‐C‐D‐E‐F PIGF
Sorafenib	VEGFR 1‐2‐3 PDGFR‐beta Raf‐1
Cediranib	VEGFR 1‐2‐3 PDGFR‐alpha c‐Kit
Sunitinib	VEGFR 1‐2‐3 PDGFR c‐Kit RTK FLT3
Pazopanib	VEGFR 1‐2‐3 PDGFR‐alpha and ‐beta c‐kit
VEGFR: vascular endothelial growth factor receptors; VEGF: vascular endothelial growth factor; PDGFR: platelet‐derived growth factor receptor; c‐Kit: tyrosine kinase; RTK: tyrosine kinase; FLT3: tyrosine kinase; Raf 1: tyrosine kinase; PIGF: placental growth factor

Why it is important to do this review

Increasing knowledge about the molecular biology of gynaecological cancers has led to the development of biological therapies. Treatment options for metastatic and recurrent cervical cancer are limited, and other agents have not shown a survival advantage compared with standard therapy. However, a phase III clinical trial showed that the addition of bevacizumab to chemotherapy improved survival from 13.3 months to 16.8 months (hazard ratio for death 0.71, 98% confidence interval 0.54 to 0.95; P = 0.004 in a one‐sided t test) (Penson 2015; Tewari 2014; Tewari 2017). Since publication of the early results of Tewari 2014, bevacizumab has become the standard of care in addition to chemotherapy for individuals with metastatic and recurrent cervical cancer in some high income countries Other anti‐angiogenic agents have been studied to a lesser degree than bevacizumab.

We aim to perform a systematic review to assess the benefits and harms of angiogenesis inhibitors for the treatment of cervical cancer with the goal of assisting clinicians and patients in the decision‐making process.

Objectives

To evaluate the benefits and harms of anti‐angiogenesis agents for the treatment of metastatic and recurrent cervical cancer.

Methods

Criteria for considering studies for this review

Types of studies

RCTs that evaluate the use of anti‐angiogenesis therapy for the treatment of recurrent or metastatic cervical cancer. We will exclude observational studies such as case reports, case series, case‐control and case‐cohort and non‐RCTs. If no RCTs are identified, we will summarise the outcomes of non‐randomised studies in a narrative review and in tabular format.

Types of participants

We will include adult women over 18 years of age, diagnosed with recurrent or metastatic cervical cancer of any stage.

Types of interventions

Anti‐angiogenic therapies, single agent or combined with other chemotherapeutic agents, such as platinum‐based chemotherapy, taxane and/or topotecan for the treatment of advanced cervical cancer versus other platinum‐based or non‐platinum chemotherapy regimens (see Table 1).

Types of outcome measures

We will consider overall survival, progression‐free survival, adverse events, and quality of life.

Primary outcomes

• Overall survival (OS)

• Progression‐free survival (PFS)

Secondary outcomes

• Response rates assessed using Response Evaluation Criteria in Solid Tumours (RECIST) v 1.1 cervical cancer

• Disease‐specific survival

• Adverse events during treatment such as dry mouth, cough, voice changes, loss of appetite, mouth sores, headache, back pain, flu‐like symptoms (stuffy nose, sneezing, sore throat), dry or watery eyes, dry or flaky skin, hair loss, and change of taste classified according to World Health Organization (WHO) or National Cancer Institute Common Toxicity Criteria (NCI‐CTC) (www.ctep.cancer.gov).

• Serious adverse events and side effects of treatment (i.e. hypertension, fistula formation, etc.)

• Quality of life (QoL) on treatment and post‐treatment when reported, and measured by the Functional Assessment of Cancer Therapy‐Cervix Trial Outcome Index (FACT‐Cx TOI) (http://www.facit.org/facitorg/) to assess physical and functional well‐being specific to cervical cancer. European Organisation for Research and Treatment of Cancer Quality Questionnaire Core 30 Items (EORTC QLQ‐C30), the Functional Assessment of Chronic Illness Therapy (FACIT) Measurement System, the Rotterdam Symptom Checklist (RSCL), and the Symptom Distress Scale (SDS) (Tamburini 2001).

Search methods for identification of studies

We will identify all relevant RCTs regardless of language and publication status (published, unpublished, in press, or in progress). The search date will commence in 1990 as we do not expect to find any trials investigating anti‐angiogenesis inhibition before this date (Rodriguez‐Freixinos 2015).

Electronic searches

We will search the following databases:

• the Cochrane Central Register of Controlled Trials (CENTRAL; latest issue) in the Cochrane Library;

• MEDLINE via Ovid (1946 to present);

• Embase via Ovid (1980 to present).

We will also search online registers of clinical trials, abstracts of scientific meetings, and reference lists of included studies, and we will use the 'related articles' feature in PubMed of the studies we are including to avoid missing studies. See search strategies designed for MEDLINE (Appendix 1), Embase (Appendix 2), and CENTRAL (Appendix 3).

Searching other resources

We will search for study records from different registries: the ISRCTN registry (www.isrctn.com) and from the National Institute for Health Research www.isrctn.com) US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov) and the National Cancer Institute (www.nci.nih.gov), for ongoing trials. We will search abstracts from the American Society of Clinical Oncology (ASCO), European Society Gynaecological Oncology (ESGO), International Gynaecologic Cancer Society (IGCS), British Gynaecological Cancer Society (BGCS) and Society of Gynecologic Oncology (SGO). We will also review other relevant websites: the WCG Center Watch (www.isrctn.com), the European Organisation for Research and Treatment of Cancer (www.isrctn.com) and the Cochrane library (www.cochranelibrary.com),

Data collection and analysis

Selection of studies

We will download all titles and abstracts retrieved by electronic searching to the EndNote database version X9 (EndNote X9)and remove duplicates. Two review authors (IR and SO) will examine the remaining articles to identify potentially relevant trials. Any disagreements will be resolved by discussion or by consulting a third review author (RB) if necessary. We will exclude studies that clearly do not meet the inclusion criteria and obtain the full texts of potentially relevant trials. We will document the reasons for exclusion of any excluded studies.

Data extraction and management

Two review authors will independently extract data (IR and RB) onto a pre‐designed data collection form. We will assess the risk of bias of the included studies according to the Cochrane Handbook for Systematic Reviews of Interventions and analyse data using Review Manager 5 (Review Manager 2014).

We will collect the following data for the included studies when reported.

• Author, year of publication and journal citation, and country where the trial was performed.

• Inclusion and exclusion criteria of the study.

• Study design and methodology.

• Study population including: total number of women enrolled, women characteristics, age, previous therapy, and comorbidities.

• Cervical cancer characteristics such as FIGO stage, histological type, tumour grade, and extent of disease.

• Total number of intervention groups.

• Intervention details such as the anti‐angiogenetic factor studied, dose, regimen, frequency and number of cycles.

• Details about the comparison when applicable such as type of control and dose, regimen, frequency and number of cycles.

• Proportion of participants who received all/part/none of the intended treatment.

• Duration of follow‐up.

• Outcomes such as overall survival, PFS, QoL, symptom control, and adverse events.

In the case that cluster‐randomised trials are eligible, we will carry out analysis using the inverse‐variance method, and appropriately choose intraclass correlations. If there are women with more than one recorded time‐to‐recurrence interval, we will include only the first interval where possible.

We will extract data on outcomes as follows.

• For time‐to‐event data (OS and PFS), we will abstract the hazard ratio (HR), log of the hazard ratio (log(HR)) and its standard error (SE) from trial reports where possible. If these are not reported, we will attempt to estimate them from other reported statistics using the methods of Parmar 1998 (Parmar 1998) (e.g. number of events in each arm and log‐rank P value comparing the relevant outcomes in each arm). If it is not possible to estimate the HR, we will abstract the number of women in each treatment arm who experienced the outcome of interest at a specific time point, in order to estimate a risk ratio (RR).

• For dichotomous outcomes (e.g. adverse events), we will abstract the number of women in each treatment arm who experienced the outcome of interest and the number of women assessed at endpoint, in order to estimate an RR.

• For continuous outcomes (e.g. QoL measures), we will abstract the mean difference (MD) and standard deviation (SD) between the final value of the outcome measure in each treatment arm at the end of follow‐up. If the SDs of final values are not available, we will use change scores if their SDs are available. If no SDs are available, we will omit these trials from the analyses and include them only in a narrative analysis with a table.

Where possible, we will extract data relevant to an intention‐to‐treat analysis (ITT), analysing participants in the groups to which they were assigned. Where time‐to‐event outcomes are assessed by more than one method (e.g. radiology reviews, investigator assessment or oncology review), we will use the radiology review data. We will collect the time points at which outcomes are collected and reported. Where data from several time points are reported, we will use the data from the last assessment in our meta‐analyses if appropriate. Where a trial evaluates the same drug in two or more different doses versus the use of anti‐angiogenic factor, we will collect the combined data and the individual data of the most efficacious regimen versus the intervention.

Two review authors (IR and SO) will independently collect data using a piloted data extraction form designed specifically for this review.

Assessment of risk of bias in included studies

We will assess risk of bias of the included RCTs using the Cochrane 'Risk of bias' tool and the criteria specified in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). This will include an assessment of the following.

• Selection bias (random sequence generation and allocation concealment)

• Performance bias (blinding of participants and personnel)

• Detection bias (blinding of outcome assessment)

• Attrition bias (incomplete outcome data; we will record the proportion of participants whose outcomes are not reported at the end of the study, considering greater than 20% attrition to be at high risk of bias)

• Reporting bias (selective reporting of outcomes)

• Other possible sources of bias.

Two review authors (IR and RB) will apply the 'Risk of bias' tool, resolving any differences by discussion or by consultation with a third review author (SO) if necessary. We will present results in a 'Risk of bias' summary graph, and interpret the results of the meta‐analyses in light of the findings with respect to risk of bias.

Measures of treatment effect

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

• For time‐to‐event data, we will use HR and 95% confidence interval (CI).

• For dichotomous outcomes, we will use RR and 95% CI.

• For continuous outcomes, we will use MD between treatment arms and the 95% CI if outcomes are measured on the same scale; otherwise we will calculate standardised mean difference (SMD).

• For studies where there is no control,we will perform subgroup analysis according to FIGO stage..

Unit of analysis issues

We do not expect any unit of analysis issues.

Dealing with missing data

We will attempt to contact study authors to obtain missing data (participant, outcome, or summary data). For participant data, we will conduct analysis on an intention‐to‐treat basis where possible; otherwise we will analyse data as reported. We will not impute missing outcome data.

Assessment of heterogeneity

We will assess heterogeneity between trials by visual inspection of forest plots, estimation of the percentage of heterogeneity between trials that cannot be ascribed to sampling variation (Higgins 2011), and a formal statistical test of the significance of the heterogeneity (Deeks 2001). If there is evidence of moderate heterogeneity, we will investigate the possible reasons for this and report this information.

Assessment of reporting biases

We will evaluate the potential small‐study effects such as publication bias using funnel plots if more than 10 studies are included in the review.

Data synthesis

If sufficient clinically similar trials are available, we will pool their results and perform meta‐analysis. For time‐to‐event data we will use HRs and analyse the data using the generic inverse‐variance method (Review Manager 2014).

For dichotomous outcomes and for continuous outcomes, we will pool RR. We will use the random‐effects method for the meta‐analysis, as a degree of heterogeneity is expected. In the case of multi‐arm studies, we will combine all relevant interventional groups into a single group and compare to the standard treatment as recommended in Section 16.5.4 of the Cochrane Handbook for Systematic Reviews of Interventions.

We will assess the overall certainty of the evidence using the GRADE approach, employing GRADEpro GDT (GRADEpro; Schunemann 2011), and we will distinguish:

• high certainty: further research is very unlikely to change our confidence in the estimate of effect;

• moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate;

• low certainty: 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 certainty: we are very uncertain about the estimate.

We will group studies according to the type of drug used in the 'Summary of findings' tables.

We will only include studies reporting data on the use of bevacizumab, sorafenib, cediranib, sunitinib, pazopanib as single agent or in combination for the treatment of metastatic and recurrent cervical cancer.

'Summary of findings' table

We will present the overall certainty of the evidence for each outcome according to the 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; Schnemann 2011). We will create a 'Summary of findings' table (Table 2) based on the methods described in the Cochrane Handbook for Systematic Reviews, (Higgins 2011a), employing GRADEpro GDT (GRADEpro). We will use the GRADE checklist and GRADE Working Group definitions for certainty of the evidence. We will downgrade the evidence from 'high' certainty by one level for serious (or by two levels for very serious) concerns for each limitation.

2. Draft 'Summary of findings' table.

Vascular endothelial growth factor (VEGF) inhibitors for the treatment of metastatic and recurrent cervical cancer
Patient or population: metastatic or recurrent cervical cancer Settings: hospital Interventions: bevacizumab plus standard chemotherapy Comparison: standard chemotherapy
Outcomes	Illustrative comparative risks*		Relative effect (95% CI)	No. of participants (studies)	Certainty of evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
Overall survival
Median progression‐free survival
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and it's 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
GRADE: Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are uncertain about the estimate.

• 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 the effect.

We will include the following outcomes in the 'Summary of findings' table:

• overall survival;

• median progression‐free survival.

If meta‐analysis is not possible, we will present the results in a narrative 'Summary of findings' table format (Chan 2011).

Subgroup analysis and investigation of heterogeneity

We will group the RCTs by type of intervention, and subgroup analysis will include advanced cervical cancer and single use of anti‐angiogenesis therapy and different drug combinations.

Sensitivity analysis

We will exclude studies at high risk of bias.

Summary of findings and assessment of the certainty of the evidence

See Table 2 for an example of the summary of findings example (Table 2)

What's new

Date	Event	Description
25 May 2021	Amended	Merged with Vascular endothelial growth factor (VEGF) targeting therapy for persistent, recurrent, or metastatic cervical cancer. See www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD013348.pub2/full.

History

Protocol first published: Issue 5, 2020

Notes

Merged with Vascular endothelial growth factor (VEGF) targeting therapy for persistent, recurrent, or metastatic cervical cancer. See www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD013348.pub2/full.

Appendices

Appendix 1. MEDLINE search strategy

1. Uterine Cervical Neoplasms/
2. (cervi* adj5 (cancer* or tumor* or tumour* or neoplas* or malignan* or carcinoma* or adenocarcinoma*)).mp.
3. 1 or 2
4. exp Angiogenesis Inhibitors/
5. (angiogene* adj5 (inhibit* or antagoni*)).mp.
6. (anti‐angiogene* or antiangiogene*).mp. [mp=title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms]
7. (bevacizumab or sorafenib or cediranib or avastin or sunitinib or sutent or pazopanib or votrient or brivanib or "BMS‐582664").mp.
8. exp Vascular Endothelial Growth Factors/
9. (vascular endothelial growth factor* or VEGF*).mp.
10. 4 or 5 or 6 or 7 or 8 or 9
11. 3 and 10

Key:

[mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

Appendix 2. Embase search strategy

1. exp uterine cervix tumor/
2. (cervi* adj5 (cancer* or tumor* or tumour* or neoplas* or malignan* or carcinoma* or adenocarcinoma*)).mp.
3. 1 or 2
4. exp angiogenesis inhibitor/
5. (angiogene* adj5 (inhibit* or antagoni*)).mp.
6. (anti‐angiogene* or antiangiogene*).mp.
7. (bevacizumab or sorafenib or cediranib or avastin or sunitinib or sutent or pazopanib or votrient or brivanib or "BMS‐582664").mp.
8. exp vasculotropin/
9. (vascular endothelial growth factor* or VEGF*).mp.
10. 4 or 5 or 6 or 7 or 8 or 9
11. 3 and 10

Key:

[mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

Appendix 3. CENTRAL search strategy

#1 MeSH descriptor: [Uterine Cervical Neoplasms] explode all trees
#2 cervi* near/5 (cancer* or tumor* or tumour* or neoplas* or malignan* or carcinoma* or adenocarcinoma*)
#3 #1 or #2
#4 MeSH descriptor: [Angiogenesis Inhibitors] explode all trees
#5 angiogene* near/5 (inhibit* or antagoni*)
#6 anti‐angiogene* or antiangiogene*
#7. bevacizumab or sorafenib or cediranib or avastin or sunitinib or sutent or pazopanib or votrient or brivanib or "BMS‐582664"
#8 MeSH descriptor: [Vascular Endothelial Growth Factors] explode all trees
#9 vascular endothelial growth factor* or VEGF*
#10 #4 or #5 or #6 or #7 or #8 or #9
#11 #3 and #10

Contributions of authors

IR wrote the first draft. SO, RB, and AS edited the draft of the protocol.

Sources of support

Internal sources

• None to declare, Other

  Not applicable

External sources

• New source of support, UK

  Ipswich library who provided the papers needed to complete the protocol

Declarations of interest

Ivana Rizzuto: nothing to declare
Sophie Otter: nothing to declare
Rasiah Bharathan: nothing to declare
Alexandra Stweart: nothing to declare

Edited (no change to conclusions)