Extended‐field radiotherapy for locally advanced cervical cancer

Komsan Thamronganantasakul; Narudom Supakalin; Chumnan Kietpeerakool; Porjai Pattanittum; Pisake Lumbiganon

doi:10.1002/14651858.CD012301.pub2

. 2018 Oct 26;2018(10):CD012301. doi: 10.1002/14651858.CD012301.pub2

Extended‐field radiotherapy for locally advanced cervical cancer

Komsan Thamronganantasakul ¹, Narudom Supakalin ¹, Chumnan Kietpeerakool ^2,^✉, Porjai Pattanittum ³, Pisake Lumbiganon ²

Editor: Cochrane Gynaecological, Neuro‐oncology and Orphan Cancer Group

PMCID: PMC6516992 PMID: 30362204

Abstract

Background

The para‐aortic lymph nodes (located along the major vessels in the mid and upper abdomen) are a common place for disease recurrence after treatment for locally advanced cervical cancer. The para‐aortic area is not covered by standard pelvic radiotherapy fields and so treatment to the pelvis alone is inadequate for women at a high risk of occult cancer within para‐aortic lymph nodes. Extended‐field radiotherapy (RT) widens the pelvic RT field to include the para‐aortic lymph node area. Extended‐field RT may improve outcomes in women with locally advanced cervical cancer by treating occult disease in para‐aortic nodes not identified at pretreatment imaging. However, RT treatment of the para‐aortic area can cause severe adverse effects, so may increase harms.

Studies of pelvic chemoradiotherapy (CRT) demonstrated improved survival rates compared to pelvic RT alone. CRT is now the standard of care in the treatment of locally advanced cervical cancer. Studies comparing pelvic RT alone (without concurrent chemotherapy) with extended‐field RT should therefore be viewed with caution, since they compare treatments against what is now substandard treatment (pelvic RT alone). This review should therefore be read with this in mind and comparisons with pelvic RT cannot be extrapolated to pelvic CRT.

Objectives

To evaluate the effectiveness and toxicity of extended‐field radiotherapy in women undergoing first‐line treatment for locally advanced cervical cancer.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2018, Issue 7), MEDLINE via Ovid (1946 to August week 4, 2018), and Embase via Ovid (1980 to 2018, week 35). We checked registers of clinical trials, grey literature, conference reports, and citation lists of included studies to August 2018.

Selection criteria

We included randomised controlled trials (RCTs) evaluating the effectiveness and toxicity of extended‐field RT for locally advanced cervical cancer.

Data collection and analysis

Two review authors independently selected potentially relevant RCTs, extracted data, assessed risk of bias, compared results, and made judgements on the quality and certainty of the evidence for each outcome. Any disagreements were resolved by discussion or consultation with a third review author.

Main results

Five studies met the inclusion criteria. Three included studies compared extended‐field RT versus pelvic RT, one included study compared extended‐field RT with pelvic CRT, and one study compared extended‐field CRT versus pelvic CRT.

Extended‐field radiotherapy versus pelvic radiotherapy alone  Compared to pelvic RT, extended‐field RT probably reduces the risk of death (hazard ratio (HR) 0.67, 95% confidence interval (CI) 0.48 to 0.94; 1 study; 337 participants; moderate‐certainty evidence) and para‐aortic lymph node recurrence (risk ratio (RR) 0.36, 95% CI 0.18 to 0.70; 2 studies; 477 participants; moderate‐certainty evidence), although there may or may not have been improvement in the risk of disease progression (HR 0.92, 95% CI 0.69 to 1.22; 1 study; 337 participants; moderate‐certainty evidence) and severe adverse events (RR 1.05, 95% CI 0.79 to 1.41; 2 studies; 776 participants; moderate‐certainty evidence).

Extended‐field radiotherapy versus pelvic chemoradiotherapy  In a comparison of extended‐field RT versus pelvic CRT, women given pelvic CRT probably had a lower risk of death (HR 0.50, 95% CI 0.39 to 0.64; 1 study; 389 participants; moderate‐certainty evidence) and disease progression (HR 0.52, 95% CI 0.37 to 0.72; 1 study; 389 participants; moderate‐certainty evidence). Participants given extended‐field RT may or may not have had a lower risk of para‐aortic lymph node recurrence (HR 0.44, 95% CI 0.20 to 0.99; 1 study; 389 participants; low‐certainty evidence) and acute severe adverse events (RR 0.05, 95% CI 0.02 to 0.11; 1 study; 388 participants; moderate‐certainty evidence). There were no clear differences in terms of late severe adverse events among the comparison groups (RR 1.06, 95% CI 0.69 to 1.62; 1 study; 386 participants; moderate‐certainty evidence).

Extended‐field chemoradiotherapy versus pelvic chemoradiotherapy  Very low‐certainty evidence obtained from one small study (74 participants) showed that, compared to pelvic CRT, extended‐field CRT may or may not have reduced risk of death (HR 0.37, 95% CI 0.14 to 0.96) and disease progression (HR 0.25, 95% CI 0.07 to 0.87). There were no clear differences between the groups in the risks of para‐aortic lymph node recurrence (RR 0.19, 95% CI 0.02 to 1.54; very low‐certainty evidence) and severe adverse events (acute: RR 0.95, 95% CI 0.20 to 4.39; late: RR 0.95, 95% CI 0.06 to 14.59; very low‐certainty evidence).

Authors' conclusions

Moderate‐certainty evidence shows that, compared with pelvic RT alone, extended‐field RT probably improves overall survival and reduces risk of para‐aortic lymph node recurrence. However, pelvic RT alone would now be considered substandard treatment, so this result cannot be extrapolated to modern standards of care. Low‐ to moderate‐certainty evidence suggests that pelvic CRT may increase overall and progression‐free survival compared to extended‐field RT, although there may or may not be a higher rate of para‐aortic recurrence and acute adverse events. Extended‐field CRT versus pelvic CRT may improve overall or progression‐free survival, but these findings should be interpreted with caution due to very low‐certainty evidence.

High‐quality RCTs, comparing modern treatment techniques in CRT, are needed to more fully inform treatment for locally advanced cervical cancer without obvious para‐aortic node involvement.

Plain language summary

Does extended‐field radiotherapy reduce death from locally advanced cervical cancer and what are the side effects?

The issue  Radiotherapy (RT) to the pelvis is used to treat cervical cancer. However, pelvic RT will not treat cancer that has spread to para‐aortic lymph nodes (lymph nodes lying along main blood vessels in the mid and upper stomach), since these are outside the target area (field) of RT. Extended‐field RT targets areas containing both pelvic and para‐aortic lymph nodes. Widening the RT field to include the para‐aortic area may reduce the risk of cancer returning.

Chemotherapy is now normally given at the same time as RT for the treatment of cervical cancer (concurrent chemotherapy) and the combine treatment is called chemoradiotherapy (CRT). This is now standard treatment because studies have shown that addition of chemotherapy during RT improved survival for women with cervical cancer thought to be confined to the pelvis. Older studies, which compared treatments with pelvic RT alone, would not now be considered the standard of care for women well enough to have CRT. We cannot assume that results from studies which compared extended‐field RT with pelvic RT apply to modern CRT treatments.

The aim of this review  In women with locally advanced cervical cancer, does extending the RT field to cover the para‐aortic area reduce the risk of death from cervical cancer and what are the harms?

Study characteristics  We searched databases from their inception to August 2018 and found five studies that met the inclusion criteria. Three studies compared extended‐field RT versus pelvic RT. None of these three studies compared against the current gold‐standard of pelvic CRT. One study compared extended‐field RT versus pelvic CRT and one study compared extended‐field CRT versus pelvic CRT.

What were the main findings?  Compared with pelvic RT alone, women given extended‐field RT may have been less likely to die and probably were less likely to have a cervical cancer come back (recurrence) in the para‐aortic lymph nodes. However, extended‐field RT may have made little or no difference to how often their cancer recurred elsewhere and how often they experience severe side effects.

Pelvic CRT is the modern standard of treatment for locally advanced cervical cancer. In a comparison of extended‐field RT alone versus pelvic CRT, women given pelvic CRT were probably less likely to die or have recurrence of their cancer. Women given extended‐field RT alone may have been less likely to experience a recurrence within the para‐aortic lymph nodes and have had adverse events during or shortly after treatment. There were no clear differences regarding the late adverse events between the two groups.

Women given extended‐field CRT may or may not have been less likely to die or have cancer progression than those women pelvic CRT. There were no clear differences in the chances of experiencing a cancer recurrence in the para‐aortic lymph nodes and severe side effects between the groups.

Certainty of the evidence  The evidence for outcomes in the comparison of extended‐field RT alone versus pelvic RT alone were of moderate certainty. In the comparison of extended‐field RT versus pelvic CRT, the evidence regarding the survival and side effects were of moderate certainty. The evidence for para‐aortic recurrence was of low certainty. The evidence for all outcomes in a comparison of extended‐field CRT versus pelvic CRT were of very‐low certainty because of concerns regarding the high risk of bias and results coming from a single trial of very few women.

What were the conclusions?  We are moderately certain that, compared with pelvic RT alone, extended‐field RT probably improves overall survival and reduces risk of para‐aortic lymph node recurrence. However, pelvic RT alone would now not be considered the standard of care in women well enough to receive CRT, so these results should be reviewed with caution and cannot be extrapolated to modern treatment techniques.

Low‐ to moderate‐certainty evidence supports the use of pelvic CRT rather than extended‐field RT alone, as it appears to reduce the risk of death and cancer progression. The likelihood of experiencing unwanted side effects during treatment was higher among women receiving pelvic CRT than extended‐field RT. Evidence comparing extended‐field CRT to pelvic CRT was very low certainty regarding outcomes and it may or may not improve survival.

Summary of findings

Background

Description of the condition

Cervical cancer is the most common gynaecological cancer affecting women worldwide, with an estimated 528,000 new cases, and 266,000 cervical cancer deaths, globally in 2012 (GLOBOCAN 2012). Approximately 70% of all cervical cancer cases, and almost 90% of cervical cancer‐related deaths, occur in low‐income countries. The high death rate among women residing in low‐income countries is because most women present with advanced disease (Pisani 1999). A woman's risk of developing cervical cancer before the age of 75 years is estimated to range from 0.9% in high‐income to 1.9% in low‐income countries (Jemal 2011). This discrepancy in the incidence, risk, and stage at diagnosis between high and low‐income countries is due to lack of access to effective screening programmes that facilitate early diagnosis and treatment.

A summary of the International Federation of Gynecology and Obstetrics (FIGO) staging system for cervical cancer is displayed in Appendix 1 (FIGO Committee 2014). Women with FIGO stage IA1 are normally treated with local excision or simple hysterectomy (surgery to remove the uterus and the neck of the uterus). FIGO stage IA2 to IB1 cervical cancers have an increased risk of pelvic lymph node involvement (3.4% to 15.3%) (Mahawerawat 2013; Sakuragi 1999; Spirtos 2002; Suprasert 2010), and are usually treated with radical hysterectomy (the removal of the uterus, the cervix, the upper part of the vagina, and the tissues around the cervix), and pelvic lymph node dissection. However, treatment outcomes of pelvic radiotherapy (RT) are comparable to those of women who undergo surgical treatment (Landoni 1997). Therefore, decision‐making for early‐stage cervical cancer should be tailored based on individual patient characteristics, weighing up the risks of surgery with the longer‐term risks of radical RT. Pelvic RT was the principal treatment for more advanced cervical cancer, since the rate of positive pelvic lymph nodes is higher (22.7% to 71.4%) (Marnitz 2005; Sakuragi 1999; Srisomboon 2011; Suprasert 2010), although some clinicians recommend extended radical hysterectomy for locally advanced disease (Höckel 2008).

Randomised controlled trial (RCT) data demonstrated that giving chemotherapy at the same time as pelvic RT as a radiosensitiser, or the so‐called 'concurrent chemoradiation' or chemoradiotherapy (CRT) significantly improved the rates of local and distant disease control for women with all stages of cervical cancer compared to RT alone (Keys 1999; Morris 1999; Peters 2000; Rose 1999; Whitney 1999). Pelvic concurrent CRT has been recommended as the treatment of choice instead of pelvic RT alone in locally advanced cervical cancer, if clinically feasible (National Cancer Institute 1999). One Cochrane Review also endorsed the use of pelvic CRT versus pelvic RT alone for women with cervical cancer based on evidence obtained from individual patient data meta‐analysis (CCCMAC 2010).

Description of the intervention

Conventionally, whole pelvis external‐beam RT treatment field covers the space between L4 (fourth lumbar spine) to L5 (fifth lumbar spine) to the mid of pubis, or to a line 4 cm below the most distal vaginal or cervical site of disease. Lateral fields are designed to encompass S3 (third sacral spine) posteriorly, with a margin of at least 3 cm from the primary cervical tumour. The RT schedule is about 1.6 Gy to 2.0 Gy (unit of irradiation dose) per day, five days per week, with a total radiation dose of 45 Gy to 50 Gy. The prescribed dose is delivered to the mid‐depth on the central axis of the beams. Intracavitary brachytherapy is performed after the completion of pelvic RT and a total cumulative dose to point A is at least 80 Gy (Eifel 2004; Lim 2011; Rotman 1978; Rotman 1990).

Extended‐field or para‐aortic RT is the extension of whole pelvis external‐beam RT to also cover the para‐aortic lymph nodes (the nodes lying adjacent to the aorta and vena cava, the major vessels in the mid and upper abdomen). This additional RT is delivered in the same setting with the conventional whole pelvis external‐beam RT. The RT schedule and total dose of the two interventions are the same, but the extended‐field RT covers a wider area (Eifel 2004; Rotman 1978; Rotman 1990).

Severe acute and late morbidities (adverse effects and complications of treatment) seem to be fundamental reasons limiting the use of extended‐field RT for cervical cancer, particularly among women receiving CRT. Severe acute toxicities in women who had undergone extended‐field RT with concomitant chemotherapy occur at approximately 33% to 80% and severe late toxicities (grade 3 or higher) occur in 10% to 15% (Kim 2009; Sood 2003; Varia 1998). Approximately 10% of women receiving extended‐field RT develop severe acute toxicities and 6% of women develop severe late toxicities, following extended‐field RT alone (Sood 2003).

How the intervention might work

Occult para‐aortic lymph node disease can occur in women thought to have locally advanced cervical cancer, which is confined to the pelvis. From a previous literature review, which was undertaken to evaluate the impact of pretreatment surgical para‐aortic lymph node staging on treatment outcomes of cervical cancer, para‐aortic lymph node metastasis were noted in 11% of women with stage I disease, 13% to16% with stage II, 29% with stage III, and 36% with stage IVA (Smits 2014). In one review of the literature, despite having normal findings on positron emission tomography (PET) or positron emission tomography‐computed tomography (PET‐CT), 4% to 15% of women thought to have negative para‐aortic lymph node on PET and PET‐CT had para‐aortic lymph node metastasis (Smits 2014).

Standard radical pelvic RT is likely to be inadequate for women with para‐aortic lymph node metastasis, since the para‐aortic area is outside of standard fields of traditional pelvic RT techniques. Para‐aortic lymph node recurrence is a frequent cause of extrapelvic failure among women with locally advanced cervical cancer who had undergone the standard technique of pelvic RT (21.6% to 57.1%) (Hong 2004; Sakurai 2001). Even in the era of concomitant CRT, para‐aortic lymph node metastasis has been noted in approximately 7% to 15% of women undergoing concomitant CRT for cervical cancer (Eifel 2004; Pearcey 2002; Toita 2012).

Previous studies have observed that extended‐field RT prolonged the survival of women with endometrial cancer (Rose 1992), and cervical cancer (Marnitz 2015; Varia 1998) who had para‐aortic lymph node metastasis. In addition, prophylaxis extended‐field RT to the para‐aortic region improved overall survival in women with cervical cancer (Rotman 1995). Based on these findings, delivering extended‐field RT to cover para‐aortic regions aimed at eliminating unrecognised metastatic lesions might be effective for reducing treatment failure at this region and thus improving treatment outcomes of women with cervical cancer.

Why it is important to do this review

The prognosis for women with recurrent cervical cancer is poor and para‐aortic lymph node recurrence is one of the most common sites of treatment failure in women with cervical cancer (Hong 2004; Sakurai 2001). However, there is significant morbidity from extended‐field RT (Kim 2009; Sood 2003; Varia 1998). Evidence for an effective strategy in reducing para‐aortic lymph node recurrence is therefore required to establish whether extended‐field RT may decrease the risk of para‐aortic recurrence. To date there has been no systematic review evaluating the impact of extended‐field RT on the treatment outcomes of women with locally advanced cervical cancer.

Objectives

To evaluate the effectiveness and toxicity of extended‐field radiotherapy in women undergoing first‐line treatment for locally advanced cervical cancer.

Methods

Criteria for considering studies for this review

Types of studies

We included RCTs. We excluded quasi‐randomised trials as these may have been subject to bias.

Types of participants

Women aged 18 years or older undergoing first‐line treatment for locally advanced cervical cancer. Cervical cancers with the largest diameter of 4 cm or larger or FIGO stage IIB to IVA (or both) were considered locally advanced. Details of the FIGO staging classification of cervical cancer are shown in Appendix 1 (FIGO Committee 2014).

Types of interventions

We compared extended‐field versus standard radical pelvic RT as follows:

extended‐field RT versus pelvic RT alone;
extended‐field RT versus pelvic CRT;
extended‐field CRT versus pelvic CRT.

Types of outcome measures

Primary outcomes

Overall survival (OS): survival until death from all causes. Survival was assessed from the time when women were enrolled in the study.
Para‐aortic lymph node recurrence: any relapse or persistent disease at the para‐aortic region.

Secondary outcomes

Progression‐free survival (PFS) or disease progression: survival until appearance of a new lesion. Survival was assessed from the time when women were enrolled in the study.
Non‐para‐aortic recurrence: recurrent disease other than the para‐aortic region was classified as locoregional and distant recurrences.
Adverse events: which were classified as:
- acute complications including digestive complications (e.g. nausea/vomiting, diarrhoea); urological complications (e.g. cystitis); haematological complications (e.g. anaemia, leukopenia, neutropenia, and thrombocytopenia); and cardiovascular and thromboembolic complications (e.g. myocardial infarction, arterial thrombosis, venous thrombosis, pulmonary embolism);
- late complications including digestive complications (e.g. proctitis, sigmoiditis, intestinal obstruction, and fistular formation); and urological complications (e.g. chronic cystitis, incontinence, ureteral stenosis, and fistula) (Cox 1995).
Quality of life: using validated scales (i.e. European Organisation for Research and Treatment of Cancer (EORTC) QLQ‐CX24; Greimel 2006).
Cost‐effectiveness: using a scale that was validated and reported in a peer‐reviewed publication (i.e. European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO‐MCBS; Cherny 2015).

We presented a 'Summary of findings' tables reporting the following outcomes listed in order of priority (see Data synthesis).

Overall survival.
Para‐aortic lymph node recurrence.
Progression‐free survival.
Non‐para‐aortic recurrence.
Adverse events.
Quality of life.
Cost‐effectiveness.

Search methods for identification of studies

Electronic searches

We searched the following electronic databases:

the Cochrane Central Register of Controlled Trials (CENTRAL; 2018, Issue 7) in the Cochrane Library;
MEDLINE via Ovid (1946 to August week 4, 2018);
Embase via Ovid (1980 to 2018, week 35).

Appendix 2; Appendix 3; Appendix 4 display the search strategies for CENTRAL, MEDLINE, and Embase.

Searching other resources

Ongoing studies and grey literature

We searched the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (www.who.int/ictrp/en/) and ClinicalTrials.gov to identify any ongoing trials. If we had identified ongoing trials that had not been published, we planned to approach the principal investigators, and major co‐operative groups active in this area, to ask for relevant data.

We searched the following databases for grey literature: Open‐Grey (www.opengrey.eu/) and Index to theses (ProQuest Dissertations & Theses: UK & Ireland).

Handsearching

We handsearched the citation lists of included studies and key textbooks, and contacted experts in the field to identify further reports of trials. We also handsearched the reports of conferences in the following sources.

Annual Meeting of the American Society of Gynecologic Oncology (ASGO).
Annual Meeting of the American Society of Radiation Oncology (ASTRO).
Annual Meeting of European Society of Radiotherapy and Oncology (ESTRO).
Annual Meeting of European Society of Medical Oncology (ESMO).
Annual Meeting of the American Society of Clinical Oncology (ASCO).
Annual Meeting of the British Gynaecological Cancer Society (BGCS).
Biennial Meeting of the Asian Society of Gynecologic Oncology (ASGO).
Biennial Meeting of Asia and Oceania Federation of Obstetrics and Gynaecology (AOFOG).
Biennial Meeting of the European Society of Gynaecological Oncology (ESGO).
Biennial Meeting of the International Gynecologic Cancer Society (IGCS).

Data collection and analysis

Selection of studies

We downloaded all titles and abstracts retrieved by electronic searching to a reference management database (Endnote). After removal of duplicates, we transferred these data to Covidence (www.covidence.org). Two review authors (KT and NS) independently examined the remaining references. We excluded studies that clearly did not meet the inclusion criteria, and we obtained copies of the full‐text of potentially relevant references. Two review authors (KT and NS) independently assessed the eligibility of the retrieved reports/publications. We resolved any disagreements through discussion or, when required, we consulted a third review author (CK). We identified and excluded duplicates and collated multiple reports of the same study so that each study rather than each report was the unit of interest in the review. We used the details regarding the selection process in Covidence 2018 to complete a PRISMA flow diagram and Characteristics of excluded studies table (Liberati 2009).

Data extraction and management

Two review authors (KT and NS) independently extracted study characteristics and outcome data from included studies using Covidence 2018. We noted in the Characteristics of included studies table if outcome data were not reported in a usable way. We resolved disagreements by consensus or by involving a third review author (PL or CK). A second review author (CK) checked study characteristics for accuracy against the trial report.

For included studies, we extracted the following data using a piloted data collection form.

Author, year of publication, and journal citation (including language).
Country.
Setting.
Inclusion and exclusion criteria.
Study methodology.
Study population and characteristics:
- total number enrolled;
- participant characteristics;
- age;
- co morbidities;
- other baseline characteristics;
- FIGO stage of cervical cancer;
- tumour size (largest diameter);
- histopathological type of cervical cancer;
- status of para‐aortic lymph node.

Intervention details:
- any regimens of pelvic RT or CRT plus extended‐field RT;
- RT technique;
- dose;
- duration;
- schedule.
Comparison:
- any regimens of pelvic RT or CRT.
Risk of bias in study (see below).
Duration of follow‐up.
Outcomes: for each outcome, we extracted the outcome definition and unit of measurement. For adjusted estimates, we planned to record variables adjusted for in analyses.
Results: we extracted 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 were extracted as follows.

For time to event data (survival and disease progression), we extracted the log of the hazard ratio [log(HR)] and its standard error (SE) from trial reports. If these were not reported, we estimated the log (HR) and its SE using the methods of Parmar 1998.
For dichotomous outcomes (e.g. adverse events or deaths, if it was not possible to use the HR), we extracted the number of participants in each treatment arm who experienced the outcome of interest and the number of participants assessed at endpoint, in order to estimate a risk ratio (RR) with 95% confidence interval (CI).

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

Assessment of risk of bias in included studies

We assessed and reported on the methodological quality and risk of bias in 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.

Selection bias: random sequence generation and allocation concealment.
Performance bias: blinding of participants and personnel (participants and treatment providers).
Detection bias: blinding of outcome assessment.
Attrition bias: incomplete outcome data.
Reporting bias: selective reporting of outcomes.
Other possible bias.

Two review authors (NS and CK) independently applied the Cochrane 'Risk of bias' tool and resolved differences by discussion or by appeal to a third review author (PL). We judged each item as being at high, low, or unclear risk of bias as set out in the criteria displayed in Appendix 5 (Higgins 2011). We provided a quote from the study report or a statement as justification for the judgement for each item (or both a quote and statement) in the 'Risk of bias' table. We summarised results in both a 'Risk of bias' graph and a 'Risk of bias' summary. When interpreting treatment effects and meta‐analyses, we took into account the risk of bias for the studies that contributed to that outcome. Where information on risk of bias related to unpublished data or correspondence with a trial authors, we noted this in the table.

Measures of treatment effect

We used the following measures of the effect of treatment.

For survival outcomes (e.g. OS and PFS), we used HR and 95% CI.
For dichotomous outcomes (e.g. adverse events, recurrences, and death), we analysed data based on the number of events and the number of people assessed in the intervention and comparison groups. We planned to use these to calculate the RR and 95% CI.
For continuous outcomes (e.g. quality of life and cost‐effectiveness measures), we planned to analyse data based on the mean, standard deviation (SD), and number of people assessed for both the intervention and comparison groups to calculate mean difference (MD) between treatment arms with a 95% CI. If the MD was reported without individual group data, we planned to use this to report the study results. If studies measured the same outcome using different tools, we planned to calculate the standardised mean difference (SMD) and 95% CI using the inverse variance method.

Unit of analysis issues

We planned to include studies where individual participants were randomised and cluster‐randomised studies. For studies that used a cluster‐randomised design but did not have any information related to the design effect, we planned to estimate the design effect based on a fairly large assumed intracluster correlation of 0.10. We planned to base this assumption by analogy on studies about implementation research (Campbell 2000; Ukoumunne 1999). We intended to follow the methods stated in the Cochrane Handbook for Systematic Reviews of Interventions for determining the calculations (Higgins 2011). In a study with multiple intervention groups, where possible, we planned to combine all relevant experimental intervention groups into a single group to create a single pair‐wise comparison (Higgins 2011).

Only studies that randomised individual participants into one of two comparison groups met our inclusion criteria. There was no requirement to review unit of analysis issues. This may be required in updates of this review, should additional studies become available.

Dealing with missing data

We did not impute missing outcome data for any of the outcomes.

Assessment of heterogeneity

We assessed heterogeneity between studies by visually inspecting forest plots, estimating the percentage of heterogeneity (I² statistic) and Chi² test between trials that could not be ascribed to sampling variation (Higgins 2003; Higgins 2011), and performing a formal statistical test of the significance of identified heterogeneity (Deeks 2001), and, where possible, by subgroup analyses. If there was evidence of substantial heterogeneity (I² greater than 60%), we investigated and reported the possible reasons for this.

Assessment of reporting biases

We planned to construct funnel plots to determine the possibility of publication bias. In future updates of this review, we will construct funnel plots corresponding to meta‐analysis of the primary outcomes to assess the potential for small‐study effects such as publication bias, if we identify more than 10 studies. We plan to assess funnel plot asymmetry visually (Sterne 2011).

We did not determine the potential of reporting bias, as there were too few studies available for this review.

Data synthesis

We pooled the results in a meta‐analyses. We used the random‐effects model with inverse variance weighting for all meta‐analyses (DerSimonian 1986). We performed statistical analysis using Review Manager 5 (Review Manager 2014).

For time‐to‐event data, we pooled HRs using the generic inverse variance.
For dichotomous outcomes, we calculated the RRs for each study and pooled them.
For continuous outcomes, we planned to pool the MDs between the treatment arms, if trials measured the outcome on the same scale; if trials measured the outcome on a different scale, we planned to pool SMDs.

We prepared 'Summary of findings' tables to display the results of the meta‐analysis, based on the methods described in Chapter 11 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2011). We presented the results of the meta‐analysis for the outcomes and harms as outlined in the Types of outcome measures section. If we are unable to pool the results using meta‐analysis methods, we planned to conduct a narrative review of the available results.

Main outcomes of 'Summary of findings' tables for assessing the certainty of the evidence

We presented the overall certainty of evidence for each outcome according to the GRADE approach, which takes into account issues relating to internal validity (risk of bias, inconsistency, imprecision, publication bias), and external validity such as directness of results (Langendam 2013). We created 'Summary of findings' tables based on the methods described the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and using GRADEpro GDT. We used the GRADE checklist and GRADE Working Group definitions (Meader 2014). We downgraded the evidence from high certainty by one level for serious (or by two for very serious) limitations. See Table 1; Table 2; Table 3.

Summary of findings for the main comparison. Extended‐field radiotherapy compared to pelvic radiotherapy for locally advanced cervical cancer.

Extended‐field radiotherapy versus pelvic radiotherapy for locally advanced cervical cancer
Participant: women with locally advanced cervical cancer undergoing radiotherapy as a primary treatment  Setting: specialised hospital  Intervention: extended‐field radiotherapy  Comparison: pelvic radiotherapy
Outcomes	*Anticipated absolute effects (95% CI)**		Relative effect  (95% CI)	No of participants  (studies)	Certainty of the evidence  (GRADE)	Comments
Outcomes	Risk with pelvic radiotherapy	Risk with extended‐field radiotherapy	Relative effect  (95% CI)	No of participants  (studies)	Certainty of the evidence  (GRADE)	Comments
Overall survival	—	—	HR 0.67  (0.48 to 0.94)	337 (1 study)	⊕⊕⊕⊝ Moderate^a	As a result of the way HR was calculated, assumed and corresponding risks were not estimated.
Para‐aortic lymph node recurrence	126 per 1000	45 per 1000 (23 to 88)	RR 0.36 (0.18 to 0.70)	477  (2 studies)	⊕⊕⊕⊝ Moderate^a	—
Progression‐free survival	—	—	HR 0.92  (0.69 to 1.22)	377 (1 study)	⊕⊕⊕⊝ Moderate^b	As a result of the way, HR was calculated, assumed and corresponding risks were not estimated.
Adverse events	357 per 1000	375 per 1000  (282 to 503)	RR 1.05  (0.79 to 1.41)	776  (2 study)	⊕⊕⊕⊝ Moderate^b	See Differences between protocol and review.
*The risk in the intervention group (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; RR: risk ratio.
GRADE Working Group grades of evidence  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.

Extended‐field CRT versus pelvic CRT for locally advanced cervical cancer
Participant: women with locally advanced cervical cancer undergoing CRT as a primary treatment  Setting: specialised hospital  Intervention: extended‐field CRT  Comparison: pelvic CRT
Outcomes	*Anticipated absolute effects (95% CI)**		Relative effect  (95% CI)	No of participants  (studies)	Certainty of the evidence  (GRADE)	Comments
Outcomes	Risk with pelvic CRT	Risk with extended‐field CRT	Relative effect  (95% CI)	No of participants  (studies)	Certainty of the evidence  (GRADE)	Comments
Overall survival	—	—	HR 0.37  (0.14 to 0.96)	74 (1 study)	⊕⊝⊝⊝ Very low^a,b	As a result of the way HRs are calculated, assumed, and corresponding risks were not estimated.
Para‐aortic lymph node recurrence	139 per 1000	26 per 1000  (3 to 214)	RR 0.19  (0.02 to 1.54)	74  (1 study)	⊕⊝⊝⊝ Very low^a,b	—
Progression‐free survival	—	—	HR 0.25  (0.07 to 0.87)	74 (1 study)	⊕⊝⊝⊝ Very low^a,b	As a result of the way HRs are calculated, assumed and corresponding risks were not estimated
Acute adverse events	83 per 1000	79 per 1000  (17 to 366)	RR 0.95  (0.20 to 4.39)	74  (1 study)	⊕⊝⊝⊝ Very low^a,b	—
Late adverse events	28 per 1000	26 per 1000  (2 to 405)	RR 0.95  (0.06 to 14.59)	74  (1 study)	⊕⊝⊝⊝ Very low^a,b	—
*The risk in the intervention group (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; CRT: chemoradiotherapy; RR: risk ratio.
GRADE Working Group grades of evidence  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

FIGO Stage			Description
I	—	—	Carcinoma confined to the cervix.
—	IA	—	Invasive cancer identified only microscopically. (All gross lesions even with superficial invasion are stage IB cancers.) Invasion is limited to measured stromal invasion with a maximum depth of 5 mm and no wider than 7 mm.
—	—	IA1	Measured invasion of stroma ≤ 3 mm in depth and ≤ 7 mm width.
—	—	IA2	Measured invasion of stroma > 3 mm and < 5 mm in depth and ≤ 7 mm width.
—	IB	—	Clinical lesions confined to the cervix, or preclinical lesions greater than stage IA.
—	—	IB1	Clinical lesions no greater than 4 cm in size.
—	—	IB2	Clinical lesions > 4 cm in size.
II	—	—	Carcinoma extends beyond the uterus, but has not extended onto the pelvic wall or to the lower third of vagina.
—	IIA	—	Involvement of up to the upper 2/3 of the vagina. No obvious parametrial involvement.
—	—	IIA1	Clinically visible lesion ≤ 4 cm.
—	—	IIA2	Clinically visible lesion > 4 cm.
—	IIB	—	Parametrial involvement but not onto the pelvic sidewall.
III	—	—	Carcinoma has extended onto the pelvic sidewall. On rectal examination, there is no cancer‐free space between the tumour and pelvic sidewall. The tumour involves the lower 1/3 of the vagina. All cases of hydronephrosis or non‐functioning kidney should be included unless they are known to be due to other causes.
—	IIIA	—	Involvement of the lower vagina but no extension onto pelvic sidewall.
—	IIIB	—	Extension onto the pelvic sidewall, or hydronephrosis/non‐functioning kidney.
IV	—	—	Carcinoma has extended beyond the true pelvis or has clinically involved the mucosa of the bladder or rectum (or both).
—	IVA	—	Spread to adjacent pelvic organs.
—	IVB	—	Spread to distant organs.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Overall survival	1		Hazard Ratio (Random, 95% CI)	Subtotals only
2 Para‐aortic lymph node recurrence	2	477	Risk Ratio (IV, Random, 95% CI)	0.36 [0.18, 0.70]
3 Progression‐free survival	1		Hazard Ratio (Random, 95% CI)	Subtotals only
4 Locoregional recurrence	3	814	Risk Ratio (IV, Random, 95% CI)	1.16 [0.91, 1.48]
5 Distant recurrence	3	814	Risk Ratio (IV, Random, 95% CI)	0.90 [0.65, 1.24]
6 Combined locoregional/distant recurrence	1		Risk Ratio (IV, Random, 95% CI)	Subtotals only
7 Adverse events (onset not specified)	2	776	Risk Ratio (IV, Random, 95% CI)	1.05 [0.79, 1.41]

Methods	2‐armed parallel, RCT Study duration: July 2007 and April 2008
Participants	74 women with histopathologically confirmed squamous cell carcinoma, adenocarcinoma, or adenosquamous cell carcinoma, and radiologically negative para‐aortic lymph node cervical cancer stage IIB‐IVA. WP‐CCRT (50 participants) vs EF‐CCRT (52 participants), followed by HDR brachytherapy. Eligibility criteria: histologically confirmed squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma of the cervix uteri; clinical and radiological FIGO stage IIB–IVA, with no other evidence of distant metastasis outside the pelvis; ECOG performance status 0–2; negative para‐aortic lymph node on imaging CT, MRI, or FDG‐PET; and normal haematological, hepatic, and renal functions tests (white blood cell count ≥ 4000/mm³, absolute neutrophil count ≥ 1500/mm³, platelet count ≥ 100,000/mm³, total bilirubin ≤ 1.5 mg/dL, alanine transaminase ≤ 2 × normal, creatinine ≤ 1.5 mg/dL). Exclusion criteria: history of hysterectomy, retroperitoneal surgery, abdominal or pelvic RT, prior chemotherapy, pregnancy, or positive para‐aortic nodes on imaging or biopsy‐confirmed para‐aortic lymph node metastasis. 66% of participants were diagnosed with stage IIB. 38 (51.4%) participants had iliac/common iliac nodal involvement detected by pretreatment MRI.
Interventions	Control arm: WP‐CCRT Intervention arm: EF‐CCRT RT technique: all participants were scanned, for simulation, on a CT simulator. 91.7% of participants; equally spaced, coplanar 3D‐CRT field plans (box‐field) were applied; however, IMRT was also generated in 8.3% of participants to achieve better dose distribution. Pelvic RT technique applied in EF‐CCRT was similar to those for the WP‐CCRT group, and para‐aortic fields were added as a continuous area or with a half‐beam block, with a superior field border at the junction of T12/L1, to cover para‐aortic lymph node up to the level of the renal hila. HDR‐brachytherapy with iridium‐192 sources was given once per week, following 45 Gy of EBRT. A dose of 7 Gy per fraction, using 3 insertions to point A with a total dose of 21 Gy, was delivered, based on the dose limit derived from the treatment plan for the rectum and bladder. Both groups of participants received weekly cisplatin 40 mg/m² before 6 doses of RT.
Outcomes	Primary endpoint: locoregional recurrence (pelvic and paraaortic control) Other endpoints: distant metastasis, survival outcomes (disease‐free survival and overall survival), treatment‐related toxicities
Notes	102 participants randomised. However, 28 (27.5%) participants were excluded after randomisation due to incomplete treatment protocol (11 participants), incomplete staging (7 participants), irregular follow‐up (6 participants), and social reason (4 participants), leaving 74 participants for analyses (36 in WP‐CCRT and 38 in EF‐CCRT). Imbalance of participant allocation was noted. Participants assigned to EF‐CCRT group had a higher rate of common iliac node involvement determined by pretreatment MRI than those allocated to WP‐CCRT group (36.8% with EF‐CCRT vs 5.6% with WP‐CCRT). Technique of RT was different between the groups. In WP‐CCRT, all participants underwent 3D‐CRT. Approximately 20% of participants allocated to EF‐CCRT, however, underwent IMRT technique while the remaining 80% of participants underwent 3D‐CRT which could have resulted in a differences of treatment‐related toxicity.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No statement regarding method of random sequence generation provided.
Allocation concealment (selection bias)	Unclear risk	No statement regarding allocation concealment provided.
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	No statement regarding blinding of participants and personnel. However, the outcomes of interest were unlikely to be affected by lack of blinding of participants and personnel.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	No statement regarding blinding of outcome assessment. However. the outcomes of interest were unlikely to be affected by lack of blinding of outcome assessor.
Incomplete outcome data (attrition bias)   All outcomes	High risk	102 participants randomised. 28 (27.5%) participants excluded after randomisation due to incomplete treatment protocol (11 participants), incomplete staging (7 participants), irregular follow‐up (6 participants), and social reason (4 participants), leaving 74 participants for final analyses (36 in WR‐CCRT group and 38 in EF‐CCRT group).
Selective reporting (reporting bias)	High risk	No data on quality of life and cost‐effectiveness reported.
Other bias	High risk	Participants assigned to EF‐CCRT group had a higher rate of common iliac node involvement determined by pretreatment MRI than those allocated to WP‐CCRT group (36.8% with EF‐CCRT vs 5.6% with WP‐CCRT).

Methods	Randomised study conducted at a University Hospital in Japan. Study duration: November 1986 and October 1990.
Participants	93 women with cervical carcinoma randomly allocated. 36 participants received external RT and intracavitary brachytherapy as a primary therapy and 57 participants received RT as an adjuvant treatment following extended radical hysterectomy and external RT. All participants had histologically confirmed carcinoma of the uterine cervix with negative para‐aortic lymph node status detected by CT. Positive para‐aortic status defined as a single node of ≥ 15 mm and multiple nodes of ≥ 10 mm in diameter. This study excluded women who had RTCOG status > 3, and aged > 75 years, and women with active gastric and duodenal ulcer. The distribution of pretreatment characteristics of the participants including stage of disease, histological subtypes of cancer, age, and performance status were well balance between groups.
Interventions	Control arm: pelvic RT Intervention arm: extended‐field RT RT technique: source of beam energy for external RT was a megavoltage machine. Brachytherapy was conducted following external RT using ⁶⁰Co HDR remote after loading. Extended‐field RT was delivered through anterior‐posterior fields.
Outcomes	3‐year cause‐specific survival, local failure, distant metastasis, and complications
Notes	No data regarding overall and progression‐free survival. In addition, the authors did not state the severity of complications observed.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No statement regarding the method used to generate the allocation sequence. The authors only stated that participants were randomised based on a randomised scheme of Peto 1977.
Allocation concealment (selection bias)	Unclear risk	No statement regarding the method used to conceal the allocation sequence. The authors only stated that participants were randomised based on a randomised scheme of Peto 1977.
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	No statement regarding the blinding of participants and personnel. However, outcomes of interest were unlikely to be influenced by lack of blinding.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	No statement regarding blinding of outcome assessment. However, outcomes of interest were unlikely to be affected by lack of blinding of outcome assessor.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Data analyses based on all participants.
Selective reporting (reporting bias)	High risk	No data regarding overall, progression‐free survival, quality of life, and cost‐effectiveness. In addition, the authors did not state the severity of complications observed in this study.
Other bias	Low risk	Analyses based on an intention‐to‐treat basis.

Methods	Multicentre, 2‐armed parallel, RCT conducted by the Radiotherapy Cooperative Group of EORTC. Study duration: November 1977 to July 1981
Participants	441 previously untreated women with histologically confirmed carcinoma of the cervix with high risk of subclinical para‐aortic node metastases. 3 high‐risk groups of para‐aortic node metastasis applied in this study including: group 1: stages I and IIB proximal, i.e. with proximal vaginal or parametrial (or both) involvement, with histologically positive pelvic lymph nodes after surgery; group 2: stages I and IIB proximal without surgery and with positive pelvic lymph nodes on lymphangiogram; group 3: stages IIB distal, i.e. with distal vaginal or parametrial (or both) involvement and IIIb regardless of pelvic nodal status on lymphangiogram. Exclusion criteria: para‐aortic node involvement confirmed either by surgery or diagnosed on lymphangiogram; no lymphangiogram results or with a non‐interpretable lymphangiogram; urinary tract obstruction; higher risk of postoperative RT complications, i.e. previous abdominopelvic surgery or inflammatory bowel disease; previous history of other primary cancer, excluding basal cell carcinoma of the skin; pregnancy; and aged > 75 years.
Interventions	Control arm: pelvic RT. The upper limit of the field was the lower border of L4. Intervention arm: pelvic and para‐aortic RT. The pelvic field was extended to cover the para‐aortic nodes to the upper border of L1. RT technique: either 2 fields or 4 fields (box technique) for field configuration during EBRT. Source of beam energy was a megavoltage machine. The dose delivered to the pelvis was 40–50 Gy over 4–6 weeks using megavoltage machines. Further pelvic RT was given either by intracavitary brachytherapy or localised EBRT depending on each centre's practice. The external beam dose to the para‐aortic area was fixed at 45 Gy in 5 weeks. No chemotherapy administered.
Outcomes	Primary endpoint: 4‐year disease‐free survival Other outcomes: pelvic failure (no complete disappearance of all detectable pelvic disease or local recurrence); para‐aortic metastasis; distant metastasis other than para‐aortic metastasis; and 4‐year severe complication (grade 3 or 4) rate. However, this study did not report overall survival which is the most important outcome in research on cancer treatment.
Notes	Analyses based on an intention‐to‐treat basis. The distributions of important baseline characteristics of participants including participant age, histological subtypes, and FIGO stage were similar between groups.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Authors stated that order of treatment allocation was determined by computer‐generated random numbers.
Allocation concealment (selection bias)	Low risk	Authors stated that participants were randomised by drawing a sealed envelope.
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	No statement regarding blinding of participants and personnel; however, outcomes of interest were unlikely to be affected by lack of blinding of participants.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	No statement regarding blinding of outcome assessor; however, outcomes of interest were unlikely to be affected by lack of blinding of outcome assessment. There was a well‐defined protocol for post‐treatment surveillance.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	214 (94%) participants in the pelvic RT group and 201 (94%) in the extended‐field RT group were followed for a minimum of 4 years and available for analysis of primary outcome.
Selective reporting (reporting bias)	High risk	No overall survival, quality of life, and cost‐effectiveness were reported which are important oncological outcomes.
Other bias	Low risk	Analyses based on an intention‐to‐treat basis.

Methods	2‐armed parallel, RCT conducted by the RTOG; study number RTOG 90‐01. Study duration: 1990–1997
Participants	388 women with FIGO stages IIB–IVA, squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma of the cervix or stage IB or IIA of 1 of these cancers with a tumour diameter ≥ 5 cm or biopsy‐confirmed metastasis to pelvic lymph nodes. Other inclusion criteria included Karnofsky performance score ≥ 60 and normal results of full blood counts and serum levels of urea, creatinine, and bilirubin. Exclusion criteria: disease outside the pelvic area or spread to para‐aortic lymph nodes; prior cancer other than cutaneous basal‐cell carcinoma; medical contraindications to chemotherapy; other rare histological subtypes; and prior hysterectomy or transperitoneal staging procedure for cervical cancer, pelvic RT, or systemic chemotherapy. Participants in each treatment group were stratified according to the tumour stage (IB, IIA, or IIB vs III or IVA) and the staging method used for para‐aortic lymph nodes (clinical vs surgical). 73.7% of participants underwent lymphangiography for para‐aortic lymph node evaluation. Approximately 40% of participants had cervical cancer stage III–IVA. Median tumour diameter in participants with stage IB and IIA was 6 cm. Allocations of participants were well balance between groups determined by equal distribution of characteristics of participants.
Interventions	Intervention arm: 45 Gy of EBRT given to the pelvis and para‐aortic lymph nodes. Control arm: 45 Gy of EBRT given to the pelvis alone plus 3 cycles of fluorouracil and cisplatin ( intravenous infusion of cisplatin 75 mg/m² of body‐surface area over a 4‐hour period followed by an intravenous infusion of fluorouracil 4000 mg/m² over a 96‐hour period; days 1–5; days 22–26 of RT; and during the time of the second intracavitary RT session). RT technique: either 2 fields or 4 fields (box technique) for field configuration during EBRT. Source of beam energy was a megavoltage machine. Brachytherapy using low‐dose‐rate intracavitary RT was performed within 2 weeks (preferably < 1 week) after the completion of pelvic radiation. For participants given RT plus systemic chemotherapy, the treatment field began from the space between L4 and L5 to the mid pubis or to a line 4 cm below the most distal vaginal or cervical lesion. Lateral fields were designed to encompass S3 posteriorly, with a margin of ≥ 3 cm from primary lesion. Shielding was designed to cover the areas of the pelvic lymph nodes, with a margin of ≥ 1–1.5 cm. For participants given RT alone, the pelvic and para‐aortic areas were treated as a continuous area, with a superior field border at the space between L1 and L2. The radiation dose was calculated at the midplane in the central ray of the field (for anteroposterior‐posteroanterior fields) or to the isocentre of the beams.
Outcomes	Primary endpoint: 5‐year and 8‐year overall survival Other study endpoints: 5‐year and 8‐year disease‐free survival rates of locoregional failure, para‐aortic failure, and distance metastasis, and grade 3–4 treatment‐related adverse effects within 60 days
Notes	The second round of interim analysis of this study, which was conducted when achieving the enrolment goal (July 1998), found the significant difference in terms of survival between groups that met the requirement for early reporting. The results presented in this study were updated using the results received by 11 November 1998. Median follow‐up time in Morris 1999 was 43 months. Eifel 2004 is an updated result of Morris 1999. The median follow‐up time of the participants in Eifel 2004 was 6.6 years.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	No statement regarding the method applied for random sequence generation provided in the published article. Kathryn Winter, Co‐Director, Division of Biostatistics and Science, NRG Oncology SDMC/RTOG and Senior Director of Statistics, American College of Radiology, confirmed that this study applied an algorithm that implemented a permuted block randomisation scheme (Winter 2017 [pers comm]).
Allocation concealment (selection bias)	Low risk	No statement regarding allocation concealment provided in the published article. Kathryn Winter, Co‐Director, Division of Biostatistics and Science, NRG Oncology SDMC/RTOG and Senior Director of Statistics, American College of Radiology, confirmed that central allocation was applied in this study (RTOG 9001) (Winter 2017 [pers comm]).
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	No statement regarding blinding of participants and personnel; however, outcomes of interest were unlikely to be affected by lack of blinding of outcome assessment.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	No statement regarding blinding of outcome assessor; however, outcomes of interest were unlikely to be affected by lack of blinding of outcome assessment. There was a well‐defined protocol for post‐treatment surveillance.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Only 15 (4%) participants were excluded after randomisation (6 in chemoradiation group and 9 in extended‐field RT group). Data used during the analyses of primary outcomes were obtained from approximately 96.5% of participants.
Selective reporting (reporting bias)	High risk	No data on quality of life and cost‐effectiveness reported.
Other bias	Low risk	Data analysed according to an intention‐to‐treat basis. The distributions of baseline characteristics of the participants which have had an impact on the treatment outcomes including age, performance status, histological subtypes, FIGO stage, and pelvic lymph node involvement were balanced between groups.

Study	Reason for exclusion
Du 2010	Prospectively examined 60 women with cervical cancer who had para‐aortic lymph node metastasis and who were undergoing whole‐pelvis radiotherapy followed by brachytherapy. The results of participants undergoing extended‐field radiotherapy using IMRT were compared with those given the 3D technique. Not a randomised study.
Kim 2016	Randomised phase II multi‐institutional study conducted by the Korean Radiation Oncology Group to determine the survival benefit of prophylactic extended‐field radiotherapy for locally advanced cervical cancer relative to the hypoxic level. Participants were divided into 2 comparison groups by the result of CA9 immunohistochemical staining (CA9‐positive and CA9‐negative) and then were further randomly allocated to extended‐field radiotherapy and pelvic radiotherapy arms. However, excluded because the results were reported according to the status of CA9 expression. No data available for comparing extended‐field radiotherapy and pelvic radiotherapy.
Liang 2014	Prospectively included 32 women with stage IB2–IIIB cervical cancer with positive para‐aortic lymph node and negative para‐aortic lymph node. All participants underwent low‐dose prophylactic extended‐field, IMRT + concurrent weekly cisplatin. Non‐randomised study design.
Lin 2015	Updated report of a previously published randomised trial (Tsai 2010) that was undertaken to assess the impact of FDG‐PET on the detection of extrapelvic disease and improvement of survival among people with stage I–IVA cervical cancer who had enlarged pelvic lymph node on MRI. Pelvic radiotherapy was given to participants found to have no extrapelvic findings on PET. Extended‐field radiotherapy was given for the remainder of the participants. All participants received 6 cycles of weekly intravenous infusion of cisplatin (50 mg/m² body surface area) as a concurrent chemotherapy. Thus, the comparisons in this randomised study were not the comparisons that this review aimed to evaluate.
Sood 2003	Included 54 women with biopsy‐confirmed carcinoma of the cervix using extended‐field radiotherapy and high‐dose‐rate brachytherapy with or without concomitant chemotherapy. Non‐randomised study design containing a single‐arm study without a comparator arm.
Tsai 2010	Randomised trial to determine the usefulness of FDG‐PET for assessing extra‐pelvic metastases, designed radiotherapy field technique, and improved survival among people with stage I–IVA cervical cancer who had enlarged pelvic lymph node on MRI. Pelvic radiotherapy was given for participants who found to have no extrapelvic findings on PET. An extended‐field radiotherapy was given for the remainder of the participants. All participants received 6 cycles of weekly intravenous infusion of cisplatin (50 mg/m² body surface area) as a concurrent chemotherapy. Thus, the comparison in this randomised study was not the comparison that this review aimed to evaluate.
Vargo 2014	61 women with cervical cancer (stage IB1–IVA) who had PET‐positive para‐aortic nodes treated with extended‐field intensity modulated radiation therapy (IMRT). Non‐randomised study design.
Varia 1998	Evaluated efficacy and safety of extended‐field radiotherapy with 5‐FU and cisplatin among 86 participants with biopsy‐confirmed para‐aortic node metastases from cervical carcinoma. It was excluded because of a non‐randomised study design. All participants underwent extended‐field radiotherapy.
Wakatsuki 2015	Evaluated efficacy and the toxicity of prophylactic extended‐field carbon‐ion radiotherapy among 26 women with locally advanced squamous cell carcinoma of the uterine cervix. Non‐randomised study design. All participants underwent extended‐field carbon‐ion radiotherapy.
Yoon 2014	Conducted to assess the efficacy and toxicity of extended‐field radiotherapy for 101 women with stage IB–IVA cervical cancer and positive para‐aortic nodes. Retrospective descriptive study. All women underwent extended‐field radiotherapy.

PERMALINK

Extended‐field radiotherapy for locally advanced cervical cancer

Komsan Thamronganantasakul

Narudom Supakalin

Chumnan Kietpeerakool

Porjai Pattanittum

Pisake Lumbiganon

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Ongoing studies and grey literature

Handsearching

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Main outcomes of 'Summary of findings' tables for assessing the certainty of the evidence

Summary of findings for the main comparison. Extended‐field radiotherapy compared to pelvic radiotherapy for locally advanced cervical cancer.

Summary of findings 2. Extended‐field radiotherapy compared to pelvic chemoradiotherapy for locally advanced cervical cancer.

Summary of findings 3. Extended‐field chemoradiotherapy compared to pelvic chemoradiotherapy for locally advanced cervical cancer.

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Results

Description of studies

Results of the search

1.

Included studies

Participants

Interventions

Outcomes

Excluded studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Extended‐field radiotherapy versus pelvic radiotherapy alone

Primary outcomes

Overall survival

1.1. Analysis.

Para‐aortic lymph node recurrence

1.2. Analysis.

Secondary outcomes

Progression‐free survival

1.3. Analysis.

Locoregional recurrence