Articles: Background

Articles
Adjuvant chemotherapy following chemoradiotherapy as

primary treatment for locally advanced cervical cancer versus
chemoradiotherapy alone (OUTBACK): an international,
open-label, randomised, phase 3 trial
Linda R Mileshkin*, Kathleen N Moore*, Elizabeth H Barnes, Val Gebski, Kailash Narayan, Madeleine T King, Nathan Bradshaw, Yeh Chen Lee,
Katrina Diamante, Anthony W Fyles, William Small Jr, David K Gaffney, Pearly Khaw, Susan Brooks, J Spencer Thompson, Warner K Huh,
Cara A Mathews, Martin Buck, Aneta Suder, Thomas E Lad, Igor J Barani, Christine H Holschneider, Sylvia Van Dyk, Michael Quinn, Danny Rischin,
Bradley J Monk†, Martin R Stockler†
Summary
Lancet Oncol 2023; 24: 468–82 Background Standard treatment for locally advanced cervical cancer is chemoradiotherapy, but many patients
Published Online relapse and die of metastatic disease. We aimed to determine the effects on survival of adjuvant chemotherapy
April 17, 2023 after chemoradiotherapy.
https://doi.org/10.1016/
S1470-2045(23)00147-X
Methods The OUTBACK trial was a multicentre, open-label, randomised, phase 3 trial done in 157 hospitals
*Contributed equally
in Australia, China, Canada, New Zealand, Saudi Arabia, Singapore, and the USA. Eligible participants were
†Contributed equally
aged 18 year or older with histologically confirmed squamous cell carcinoma, adenosquamous cell carcinoma,
Department of Medical
Oncology
or adenocarcinoma of the cervix (FIGO 2008 stage IB1 disease with nodal involvement, or stage IB2, II, IIIB,
(Prof L R Mileshkin MD, or IVA disease), Eastern Cooperative Oncology Group performance status 0–2, and adequate bone marrow and
Prof D Rischin MD), Department organ function. Participants were randomly assigned centrally (1:1) using a minimisation approach and stratified
of Radiation Oncology by pelvic or common iliac nodal involvement, requirement for extended-field radiotherapy, FIGO 2008 stage, age,
(Prof K Narayan PhD,
P Khaw MBBS, S Van Dyk PhD),
and site to receive standard cisplatin-based chemoradiotherapy (40 mg/m² cisplatin intravenously once-a-week for
Peter MacCallum Cancer Centre 5 weeks, during radiotherapy with 45·0–50·4 Gy external beam radiotherapy delivered in fractions of 1·8 Gy to the
and University of Melbourne, whole pelvis plus brachytherapy; chemoradiotherapy only group) or standard cisplatin-based chemoradiotherapy
Melbourne, VIC, Australia; followed by adjuvant chemotherapy with four cycles of carboplatin (area under the receiver operator curve 5) and
Stephenson Cancer Center at
the University of Oklahoma,
paclitaxel (155 mg/m²) given intravenously on day 1 of a 21 day cycle (adjuvant chemotherapy group). The primary
Oklahoma City, OK, USA endpoint was overall survival at 5 years, analysed in the intention-to-treat population (ie, all eligible patients who
(K N Moore MD, were randomly assigned). Safety was assessed in all patients in the chemoradiotherapy only group who started
J S Thompson MD); National chemoradiotherapy and all patients in the adjuvant chemotherapy group who received at least one dose of adjuvant
Health and Medical Research
Council Clinical Trials Centre
chemotherapy. The OUTBACK trial is registered with ClinicalTrials.gov, NCT01414608, and the Australia
(E H Barnes MStat, New Zealand Clinical Trial Registry, ACTRN12610000732088.
Prof V Gebski PhD,
N Bradshaw BPsych, Findings Between April 15, 2011, and June 26, 2017, 926 patients were enrolled and randomly assigned to the
Y C Lee MBBS,
K Diamante MSCiMed,
chemoradiotherapy only group (n=461) or the adjuvant chemotherapy group (n=465), of whom 919 were eligible
Prof M R Stockler PhD) and (456 in the chemoradiotherapy only group and 463 in the adjuvant chemotherapy group; median age 46 years
School of Psychology [IQR 37 to 55]; 663 [72%] were White, 121 [13%] were Black or African American, 53 [6%] were Asian, 24 [3%] were
(Prof M T King PhD), University
Aboriginal or Pacific islander, and 57 [6%] were other races) and included in the analysis. As of data cutoff
of Sydney, Sydney, NSW,
Australia; National Cancer (April 12, 2021), median follow-up was 60 months (IQR 45 to 65). 5-year overall survival was 72% (95% CI 67 to 76)
Institute of Canada Clinical Trial in the adjuvant chemotherapy group (105 deaths) and 71% (66 to 75) in the chemoradiotherapy only group
Group, Radiation Medicine (116 deaths; difference 1% [95% CI –6 to 7]; hazard ratio 0·90 [95% CI 0·70 to 1·17]; p=0·81). In the safety
Program, Princess Margaret
population, the most common clinically significant grade 3–4 adverse events were decreased neutrophils (71 [20%]
Cancer Centre, Toronto, ON,
Canada (Prof A W Fyles MD); in the adjuvant chemotherapy group vs 34 [8%] in the chemoradiotherapy only group), and anaemia (66 [18%] vs
Department of Radiation 34 [8%]). Serious adverse events occurred in 107 (30%) in the adjuvant chemotherapy group versus
Oncology, Stritch School of 98 (22%) in the chemoradiotherapy only group, most commonly due to infectious complications. There were no
Medicine, Cardinal Bernadin
treatment-related deaths.
Cancer Center, Loyola
University Chicago, Maywood,
IL, USA (Prof W Small Jr MD); Interpretation Adjuvant carboplatin and paclitaxel chemotherapy given after standard cisplatin-based chemoradiotherapy
Department of Radiation for unselected locally advanced cervical cancer increased short-term toxicity and did not improve overall survival;
Oncology, Huntsman Cancer
therefore, it should not be given in this setting.
Institute at the University of
Utah, Salt Lake City, UT, USA
(Prof D K Gaffney MD); Funding National Health and Medical Research Council and National Cancer Institute.
Oncology, Auckland City
Copyright © 2023 Elsevier Ltd. All rights reserved.
468 www.thelancet.com/oncology Vol 24 May 2023

Descargado para Eilyn Mora Corrales ([email protected]) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en mayo 18,
2023. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2023. Elsevier Inc. Todos los derechos reservados.
Articles
Hospital, Auckland,
Research in context New Zealand (S Brooks MD);
University of Alabama at
Evidence before this study plus paclitaxel given after standard chemoradiotherapy for Birmingham, Birmingham, AL,
We searched PubMed and Google Scholar for articles written in unselected locally advanced cervical cancer increases the USA (W K Huh MD); Program in
English from database inception to Oct 26, 2022, using the occurrence of adverse events without improving overall- Women’s Oncology,
Department of Obstetrics and
terms “cervical cancer”, “cervix cancer”, “locally-advanced”, survival or progression-free survival. Our study also reinforced
Gynecology, Women and
“radiotherapy” AND “adjuvant chemotherapy” and found only the value of strong multinational intergroup collaboration Infants Hospital, Brown
a small number of randomised trials and 2 prior meta-analyses. required to successfully complete randomised trials in cervical University, Providence, RI, USA
Although some previous individual trials have suggested a cancer and raises several questions about the optimal way to (C A Mathews MD); Department
of Medical Oncology,
benefit from the addition of adjuvant chemotherapy to design future trials in this disease.
Sir Charles Gairdner Hospital,
chemo-radiation, these data was considered controversial Perth, WA, Australia
Implications of all the available evidence
because of short follow-up times and increased adverse events, (M Buck MBBS); Department of
Standard treatment for locally advanced cervical cancer should Medical Oncology, Royal
and the results of meta-analyses have not confirmed a survival
remain cisplatin-based external beam chemoradiotherapy plus Brisbane and Women’s
benefit from this practise.
brachytherapy. More efforts are needed to ensure that all Hospital, Brisbane, QLD,
Australia (A Suder MBBS);
Added value of this study women diagnosed with cervical cancer can receive standard
Division of Hematology-
OUTBACK is a large, adequately powered, randomised trial with cisplatin-based chemoradiotherapy globally. Adjuvant Oncology, Cook County
mature follow-up and overall survival as the primary endpoint. carboplatin and paclitaxel chemotherapy should not be used in Hospital, Chicago, IL, USA
This trial showed that adjuvant chemotherapy of carboplatin this setting. (T E Lad MD); Department of
Radiation Oncology, St Joseph’s
Hospital and Medical Centre,
Introduction with the same radiotherapy with concurrent cisplatin and Phoenix, AZ, USA
Cervical cancer is a substantial global health issue with gemcitabine, followed by two cycles of adjuvant cisplatin (I J Barani MD); Division of
Gynecologic Oncology,
more than half a million women diagnosed worldwide and gemcitabine.9 The trial reported a 9% improvement in University of California,
each year. In high-income countries the incidence has both progression-free survival and overall survival at Los Angeles, Los Angeles, CA,
decreased considerably following the widespread intro 3 years, and was highly influential in changing practise USA (Prof C H Holschneider MD);
duction of cervical screening programmes.1 However, at some centres. Oncology Unit, Royal Women’s
Hospital and University of
many women still die from cervical cancer, particularly However, criticisms of the trial included the short Melbourne, Melbourne, VIC,
in low-income and middle-income countries, making it follow-up of 3 years and significantly increased toxicity Australia (Prof M Quinn MGO);
the fourth leading cause of cancer-related death in in patients who received the additional concurrent and Division of Gynecologic
Oncology, HonorHealth
women worldwide.2 adjuvant chemotherapy. These limitations precluded the
Research Institute, University
For women presenting with early-stage disease, surgical widespread acceptance of the findings as standard of Arizona, Creighton
approaches or treatment with chemoradiotherapy provide treatment. However, an analysis of the US National University, Phoenix, AZ, USA
excellent outcomes. However, a substantial proportion Cancer Database, presented in 2022, suggested that one (Prof B J Monk MD)
of women, particularly those who have not participated in ten patients were receiving multidrug adjuvant Correspondence to:
in screening programmes, present with more locally chemotherapy in addition to chemoradiotherapy with Prof Linda R Mileshkin,
advanced disease and have much lower cure rates.3 no survival benefit.10 There was also a lower rate of Oncology, Peter MacCallum
Use of chemoradiotherapy for cervical cancer is proven brachytherapy completion in those treated with adjuvant Cancer Centre, University of
to improve survival, and became established as standard chemotherapy, which is known to result in inferior Melbourne, Melbourne 3000,
VIC, Australia
of care in 1999.4–6 Subsequently, a meta-analysis of outcomes.10 Consequently, the Gynecologic Cancer
[email protected]
individual patient data from 18 randomised trials found Intergroup (GCIG) envisaged the OUTBACK trial as a
that adding concurrent chemotherapy to radiotherapy confirmatory study. We aimed to test the potential
increased 5-year overall survival by 6% (60% with benefit of adjuvant chemotherapy following primary
radiotherapy alone vs 66% with chemoradiotherapy; chemoradiotherapy for locally advanced cervical cancer.
hazard ratio [HR] 0·81 [95% CI 0·71–0·91]).7 However,
a 5-year disease-free survival rate of 58% in those who Methods
received chemoradiotherapy left many women not cured, Study design and participants
with most deaths due to the subsequent development of The OUTBACK (ANZGOG 0902, RTOG 1174, NRG 0274)
distant metastatic disease.7,8 trial was a multicentre, open-label, randomised, phase 3
We hypothesised that giving additional adjuvant trial done at 157 hospitals in seven countries (Australia,
chemotherapy following chemoradiotherapy would China, Canada, New Zealand, Saudi Arabia, Singapore,
reduce distant relapses and improve overall survival. The and the USA; appendix pp 2–9). The trial was done See Online for appendix
meta-analysis by the Chemoradiotherapy for Cervical under the auspices of the GCIG, led by the Australia
Cancer Meta-Analysis Collaboration found improved and New Zealand Gynaecological Oncology Group
survival benefits in two trials that gave additional cycles of (ANZGOG), and coordinated by the National Health
adjuvant chemotherapy following chemoradiotherapy.7 and Medical Research Council Clinical Trials Centre,
Subsequently, a randomised trial done in South America University of Sydney (Sydney, NSW, Australia). Partici
compared standard cisplatin-based chemoradiotherapy pating cooperative groups or countries were NRG
www.thelancet.com/oncology Vol 24 May 2023 469

Articles
Oncology (USA, Saudi Arabia, Canada, and China) 5 weeks. Cisplatin was omitted or the dose reduced to
and Singapore. 30 mg/m² if participants had adverse events specified in
Patients aged 18 years or older with locally advanced the protocol (appendix). External beam radiotherapy
cervical cancer suitable for primary treatment with continued if cisplatin was withheld. Participants who had
chemoradiotherapy were considered for inclusion. not recovered from adverse events within 21 days did not
Participants were eligible if they had histologically receive further cisplatin.
confirmed squamous cell carcinoma, adenosquamous The standardised radiotherapy procedure used
cell carcinoma, or adenocarcinoma of the cervix megavoltage energy with a source-to-surface distance of
(International Federation of Gynecology and Obstetrics 80 cm or more. Use of linear accelerators or 60Co units
[FIGO] 2008 stage IB1 disease with nodal involvement, was allowed. A four-field box technique with parallel-
or stage IB2, IIA, IIB, IIIB, or IVA disease), Eastern opposed anterior-posterior and posterior-anterior and
Cooperative Oncology Group performance status two opposing lateral fields was recommended and
of 0–2, and adequate bone marrow and organ function. defined in the protocol; intensity modulated radiotherapy
Key exclusion criteria included previous hysterectomy, was not allowed. All patients received 45·0–50·4 Gy
para-aortic nodal involvement above the level of the external beam radiotherapy delivered in fractions of
common iliac nodes or above L3 or L4 (if biopsy proven, 1·8 Gy to the whole pelvis. Participants with common
positive on PET, or ≥15 mm short axis diameter on CT), iliac nodal disease received 45 Gy in 1·8 Gy fractions of
FIGO stage IIIA disease, and disease assessed at extended field radiotherapy. Parametrial or nodal boost
presentation as requiring interstitial brachytherapy. was allowed at the discretion of the treating radio-
A full list of eligibility criteria is in the protocol oncologist. The external beam target volume was to
(appendix). All participants had baseline CT imaging of encompass, with adequate margins, the gross tumour
the chest, abdomen, and pelvis, with or without MRI volume, including the primary cervical tumour, any
imaging of the pelvis and PET or PET/CT if available at gross extension, and any grossly involved lymph nodes.
the treating centre. The clinical target volume included the gross tumour
The protocol was approved by all participating groups volume; parametria; uterus; upper half of the vagina; the
and relevant institutional ethics review committees, and internal, external, and distal common iliac nodes; and
all participants gave written informed consent. The trial the utero-sacral ligaments.
was done according to Good Clinical Practice guidelines Intracavitary brachytherapy was delivered with
of the International Conference on Harmonisation and standard applicators using either tandem and ovoids or
the principles of the Declaration of Helsinki. tandem and ring. Brachytherapy could be delivered at
either a high-dose rate or low-dose rate to deliver a
Randomisation and masking total dose to the primary tumour of 80·0–86·4 Gy
Using a minimisation approach, participants were (equivalent dose in 2 Gy fractions), including external
randomly assigned (1:1) to receive either standard beam radiotherapy and brachytherapy. Brachytherapy
chemoradiotherapy (chemoradiotherapy only group) could be prescribed either to point A or to image-guided
or standard chemoradiotherapy followed by adjuvant target volumes.
chemotherapy (adjuvant chemotherapy group). Ran Radiation quality assurance was monitored by the
domi sation was stratified by pelvic or common iliac Imaging and Radiation Oncology Core administered by
nodal involvement, or both (yes vs no vs unknown), the American College of Radiology. Simulation films of
requirement for extended-field radiotherapy (yes vs no), digitally reconstructed radiographs (for all treatment
FIGO 2008 stage (IB or IIA vs IIB vs IIIB or IVA) age fields and phases of treatment) were submitted and
(<60 vs ≥60 years), and treating hospital or site. Random reviewed centrally for the first two participants at each
assignment was done by each site using a central web- participating site. If these were deemed acceptable,
based system. Nodal involvement was defined as any subsequent data from every tenth participant was
pelvic or common iliac nodes that were either PET submitted for central review. Centres with unsatisfactory
positive, had a short axis diameter of more than 15 mm results were required to send more participants for
on CT or MRI, or were histologically positive after review until deemed compliant with the protocol.
surgical sampling. Participants and investigators were Within 4 weeks of completing radiotherapy, including
not masked to treatment allocation given the alopecia brachytherapy, and following recovery from any
associated with paclitaxel that was administered to treatment-related adverse events, participants in the
patients in the adjuvant chemotherapy group. adjuvant chemotherapy received four 21-day cycles of
adjuvant chemotherapy using carboplatin (areas under
Procedures the receiver operator curve [AUC] 5, intravenously over
Participants in both groups received standard external 1 h) and paclitaxel (155 mg/m², intravenously over 3 h)
beam radiotherapy to the pelvis plus brachytherapy. administered on day 1 of each cycle. Hospira (Lake Forest,
Cisplatin was given concurrently during radiotherapy at IL, USA) provided paclitaxel for sites in Australia and
a dose of 40 mg/m² intravenously once-a-week for New Zealand. Before starting adjuvant chemotherapy,

Articles
any adverse events due to the previously completed All endpoints were determined by investigators at the
concomitant chemoradiotherapy (regardless of whether study sites. Several HRQoL questionnaires were used as
ceased early because of toxicity) were required to be outlined in the protocol (appendix); we report the global
resolved to less than Common Terminology Criteria for health status and quality of life scale from the EORTC
Adverse Events (CTCAE; version 4.0) grade 2 (except for QLQ-C30. Due to the large amount of HRQoL data
lymphocyte count). Adjuvant chemotherapy doses could collected, a comprehensive analysis of HRQoL will be
be delayed for up to 2 weeks or doses reduced, per reported elsewhere.11 The secondary endpoint of radiation
protocol-defined guidelines (appendix). protocol compliance will also be reported elsewhere.
After treatment, participants were followed up every
3 months for the first 2 years and then every 6 months Statistical analysis
for a minimum of 5 years. At each follow-up visit, The OUTBACK trial was originally powered to provide
a physical examination was done, and any laboratory- 80% power with a two-sided type 1 error rate of 5% if the
based monitoring was done as clinically indicted. true absolute difference in 5-year overall survival rates
Response rate using Response Evaluation Criteria in was 10% (range 63–73). This required 780 participants,
Solid Tumours (RECIST) version 1.1 was determined assuming 3 years of accrual plus 3 years of additional
locally by the site investigator in those with measurable follow-up. An Independent Data and Safety-Monitoring
disease, based on pelvis CT or MRI at baseline and Committee (IDSMC) monitored the study and advised
repeated 6 months after randomisation and if relapse the trial steering committee on the safety and feasibility
was suspected. If PET or PET and CT was done at of the study. An interim analysis for efficacy, planned
baseline, imaging was repeated 4–6 months after for after 120 deaths had occurred, was positive, and so
completing chemoradiotherapy. Metabolic PET response the was trial continued. In 2016, after consideration of
was assessed using Positron Emission Tomography the lower-than-expected progression-free survival rate,
Response Criteria in Solid Tumours (PERCIST) and substantial non-adherence to prescribed adjuvant
version 1.0. There was no restriction on treatment given chemotherapy, the IDSMC recommended that the
for relapse, which was as per investigator discretion. sample size be increased in a protocol amendment
Adverse events were classified and graded using the (version 5.0; May 9, 2016).
CTCAE (version 4.0), and assessed in person every week The revised sample size of 828 patients (414 per
during chemoradiotherapy, before each cycle of adjuvant group) provided 80% power with a two-sided type 1
chemotherapy, and at the end of study treatment, then error rate of 5% if the true absolute difference in 5-year
once every 3 months up to 2 years after randomisation, overall survival rates was 8% (range 72–80). This
and then once every 6 months up to 5 years after corresponded to a HR of 0·68 and assumed 48 months
randomisation. Health-related quality of life (HRQoL) for accrual plus 42 months of additional follow-up. The
was assessed at baseline, at the end of chemoradiotherapy, study sample size was increased to 900 (450 per group)
before each cycle of adjuvant chemotherapy, and then at to allow for non-adherence and loss to follow-up.
each follow-up visit until 36 months after randomisation. All statistical analyses were done with SAS
(version 9.4). Efficacy analyses were done in the
Outcomes intention-to-treat (ITT) population, which included all
The primary endpoint was overall survival at 5 years, eligible patients who were randomly assigned to
defined as the time from the date of randomisation to treatment and analysed according to their treatment
death from any cause or censoring at last known follow-up. allocation. The safety population included eligible
The secondary endpoints were progression-free survival at patients in the chemoradiotherapy only group who
3 years and 5 years; adverse events within 1 year of started chemoradiotherapy and eligible patients in
randomisation (short-term safety) and after 1 year (long- the adjuvant chemotherapy group who received at
term safety); the patterns of disease response and least one dose of adjuvant chemotherapy; patients
recurrence; radiation protocol compliance; and aspects of were analysed according to their randomly assigned
self-reported HRQoL related to both disease and treatment treatment group. Patterns of disease recurrence were
given. A tertiary endpoint was complete and partial measured in the ITT population. The secondary
metabolic response by PERCIST 1.0 criteria on a PET scan endpoints of acute and long-term safety were defined
done 4–6 months after completion of chemoradiotherapy. according to worst grade experienced within 1 year
Progression-free survival was defined as the time from after randomisation (acute safety) and worst grade
the date of randomisation to tumour progression or experienced at any time after 1 year (long-term saftery).
recurrence at any site, commencement of non-protocol For the acute and long-term safety analyses, participants
anticancer therapy, or death from any cause, whichever in the safety population with no recorded safety
occurred first. Recurrences were analysed according to assessment 1 year after randomisation were excluded
the first site of recurrence, and defined as either from the analyses of long-term adverse events. For
persistent disease (present at treatment completion), endpoints of response, these were calculated according
locoregional, or distance recurrence. to RECIST 1.1 in participants with measurable disease

Articles
at baseline, and patients who did not have a scan after was set to have occurred 0·5 days after randomisation.
baseline were treated as non-evaluable and excluded When these patients were removed, the proportional
from this analysis. hazards assumption was upheld, but all participants
The primary outcome of overall survival and its 95% CI are included in all survival analyses reported, unless
at 3 and 5 years were estimated in each randomised treat otherwise indicated.
ment group using the Kaplan-Meier method, and SE was Given substantial rates of non-adherence in the adjuvant
estimated using the Greenwood method. An unstratified chemotherapy group, and our previous findings that non-
log-rank test was used to quantify the evidence for overall completion of chemoradiation was the strongest predictor
survival differences between the two groups. For survival of not starting any adjuvant chemotherapy,13 a prespecified
outcomes, patients who had not experienced the event of sensitivity analysis for overall survival and progression-free
interest were censored at their last known event-free date. survival was was done based on those who did or did not
Proportional hazards regression models were used to complete the initial chemoradiotherapy (appendix), using
calculate HRs and 95% CIs for overall survival and the Kaplan-Meier method to estimate survival outcomes
progression-free survival at 3 years and 5 years. The and a proportional hazards regression model with
proportional hazards assumption for overall survival and randomised treatment, completion or non-completion of
progression-free survival was tested with the method chemoradiotherapy and their interaction as predictors.
introduced by Li and colleagues12 based on Martingale Subgroup analyses were done for the outcomes of 5 year
residuals. Calculations for progression-free survival used overall survival and progression-free survival, according to
the same method. Tests of the proportional hazards prespecified baseline characteristics (ie, nodal involvement
assumption showed no evidence of non-proportionality [yes vs no vs unknown], requirement for extended-field
for overall survival, but some evidence for progression- radiotherapy [yes vs no], FIGO 2008 stage [IB1 with nodal
free survival. However, 20 patients had persistent disease involvement or IB2 or IIA vs IIB vs IIIB or IVA], age [<60 vs
(five in the adjuvant chemotherapy group and 15 in the ≥60 years], country [the USA and Canada vs Australia and
chemoradiotherapy only group) whose progression New Zealand vs China, Saudi Arabia, and Singapore],
and tobacco smoking status [never smoker vs current or
former smoker or unknown]) and one post-hoc predictor
926 patients randomly assigned (tumour histology [squamous or adenosquamous cell
carcinoma vs adenocarcinoma]). The Cox proportional
hazards regression models for these analyses included
terms for the subgroup, randomised treatment group, and
461 assigned to the chemoradiotherapy only 465 assigned to the adjuvant chemotherapy
group group their interaction, with p values calculated.
Cervical cancer-specific mortality was compared
between the two treatment groups was done with Gray’s
5 excluded after randomisation 2 excluded after randomisation
(ineligible) (ineligible)
method. The HR and 95% CI was calculated using the
Fine-Gray method, after masked central adjudication of
the cause of death, before the primary analysis was done.
456 assigned chemoradiotherapy 463 assigned adjuvant chemotherapy For cervical-cancer-specific mortality, death from other
causes was treated as a competing risk and patients alive
3 did not start chemoradiotherapy* 10 did not start chemoradiotherapy* at last follow-up were censored. A prespecified analysis in
participants who underwent a PET scan 168 to 365 days
after randomisation explored the association between
453 started chemoradiotherapy 453 started chemoradiotherapy
353 completed chemoradiotherapy 356 completed chemoradiotherapy
PET response (complete metabolic response [CMR] vs any
other result [non-CMR, progressive metabolic disease,
result unknown]) and the outcomes of overall survival
92 did not start adjuvant and progression-free survival. The Kaplan-Meier method
chemotherapy
was used to calculate survival estimates separately for
participants with and without CMR in each treatment
361 started adjuvant chemotherapy group. Proportional hazards regressions model tested for
285 completed adjuvant chemotherapy
interaction between CMR and treatment group and
calculated HRs for the effect of treatment and CMR in a
456 included in survival analyses 463 included in survival analyses multivariable model.
334 alive at last known follow-up 354 alive at last known follow-up Changes in mean HRQoL scores from baseline for
participants with at least one follow-up questionnaire
Figure 1: Trial profile were compared between groups using two-sample
*The most common reason for not starting chemoradiotherapy in both groups was withdrawl of patient consent
due to patient preference. In the adjuvant chemotherapy group, two patients were deemed to have disease not
Student’s t tests and described with difference and
suitable for treatment with radiotherapy as per the study protocol, and one patient had a haemoglobin level too 95% CIs. Categorical outcomes were compared with
low for eligibility. the χ² test. The OUTBACK trial is registered with

Articles
ClinicalTrials.gov, NCT01414608, and the Australia New

Chemoradiotherapy Adjuvant
Zealand Clinical Trial Registry, ACTRN12610000732088. only group (n=456) chemotherapy
group (n=463)
Role of the funding source Age, years 45 (38–54) 46 (37–55)
The funding bodies had no role in study design, data ECOG performance status
collection, data interpretation or analysis, or writing of 0 344 (75%) 337 (73%)
the report. 1 94 (21%) 117 (25%)
2 18 (4%) 9 (2%)
Results Race*
Between April 15, 2011, and June 26, 2017, 926 participants
White 326 (71%) 337 (73%)
were randomly assigned to treatment. Seven participants
Black or African American 68 (15%) 53 (11%)
were found to be ineligible after randomisation and
Asian 22 (5%) 31 (7%)
were excluded from analyses (figure 1). 919 women
Aboriginal or Pacific Islander 11 (2%) 13 (3%)
(99%; median age 46 years [IQR 37–55]; 663 [72%] of
Other 28 (6%) 29 (6%)
919 were White, 121 [13%] were Black or African
Geographical region
American, 53 [6%] were Asian, 24 [3%] were Aboriginal
Australia and New Zealand 84 (18%) 81 (17%)
or Pacific islander, and 57 [6%] were other races) were
USA and Canada 366 (80%) 373 (81%)
included in the ITT population (456 [50%] randomly
China, Saudi Arabia, and 6 (1%) 9 (2%)
assigned to the chemoradiotherapy only group and Singapore
463 [50%] to the adjuvant chemotherapy group. As
Tobacco smoking
of data cutoff (April 12, 2021), the median duration
Never smoker 237 (52%) 224 (48%)
of follow-up was 60 months (IQR 45–65). Baseline
Current or former smoker 219 (48%) 239 (52%)
characteristics are reported in table 1. or unknown
Adherence to the standard chemoradiotherapy protocol Histological type
was similar in the two treatment groups. Dose Squamous cell carcinoma 358 (79%) 383 (83%)
modification or delays due to adverse events occurred for Adenocarcinoma 79 (17%) 68 (15%)
162 (35%) of 463 patients in the adjuvant chemotherapy Adenosquamous cell 19 (4%) 12 (3%)
group and 144 (32%) of 456 patients in the chemoradio carcinoma
therapy only group. One (<1%) participant in the adjuvant FIGO 2008 stage
chemotherapy group received radiotherapy but no IB1 (all node positive), IB2, 152 (33%) 154 (33%)
concurrent cisplatin. External beam radiotherapy was or IIA
given without interruption in 92% of participants, and IIB 196 (43%) 197 (43%)
brachytherapy was delivered as planned in 95%. All IIIB or IVA 108 (24%) 112 (24%)
components of chemo-radiation were completed by Maximum tumour diameter, 5·0 (4·0–6·0) 5·0 (4·0–6·0)
77% of participants, including at least 45 Gy of external cm
beam radiotherapy, all planned brachytherapy, and all Nodal involvement
5 cycles of concurrent cisplatin (table 2). Pelvic alone 144 (32%) 149 (32%)
102 (22%) of 465 patients in the adjuvant chemotherapy Common iliac alone 33 (7%) 31 (7%)
group did not initiate any adjuvant chemotherapy; the Pelvic and common iliac 44 (10%) 44 (10%)
most common reason for which was patient preference Neither 225 (49%) 231 (50%)
(appendix p 72). 200 (70%) of 285 patients in the Unknown 10 (2%) 8 (2%)
adjuvant chemotherapy who completed treatment Extended field radiotherapy planned
completed all four cycles of carboplatin without dose No 397 (87%) 404 (87%)
reduction or delay, and 197 (69%) completed all four Yes 59 (13%) 59 (13%)
cycles of paclitaxel without dose reduction or delay Data are median (IQR) or n (%). ECOG=Eastern Cooperative Oncology Group.
(appendix p 10). FIGO=International Federation of Gynecology and Obstetrics. *Self-reported.
As of data cut-off, 109 (24%) of 463 patients in the
Table 1: Baseline demographic and disease characteristics, intention-to-
adjuvant chemotherapy group and 123 (27%) of treat population
456 patients in the chemoradiotherapy only group had
died; median overall survival time was not reached in
either group (figure 2). 5-year overall survival was 72% the adjuvant chemotherapy group versus 168 events
(95% CI 67 to 76; 105 deaths) in the adjuvant (18 deaths; 150 progressions) in the chemoradiotherapy
chemotherapy group versus 71% (66 to 75; 116 deaths) in only group. The 3-year and 5-year progression-
the chemoradiotherapy only group (difference 1% free survival are shown in figure 3. There were
[95% CI –6 to 7]; HR 0·90 [95% CI 0·70 to 1·17]; p=0·81). 84 cervical-cancer specific deaths and 25 with other
As of data cut-off, 151 progression-free survival or unknown cause in the adjuvant chemotherapy
events (20 deaths; 131 progressions) were reported in group versus 102 cervical cancer-specific deaths and

Articles
Forest plots and p values for interactions showed no

Chemoradiotherapy Adjuvant
only group (n=456) chemotherapy evidence that adjuvant chemotherapy had different
group (n=463) effects on 5-year overall survival or 5-year progression-
Chemotherapy free survival in prespecified subgroups defined by nodal
Number of cisplatin cycles commenced status, requirement for extended field radiotherapy,
0–3 27 (6%) 49 (11%) FIGO stage, country, and smoking status, or post-hoc
4 46 (10%) 32 (7%)
subgroup histological subtype (figure 4). A sensitivity
5 383 (84%) 382 (83%)
analysis showed that in those who completed initial
Cisplatin cycles completed at full dose with no delays
chemoradiotherapy, the 5-year overall survival was
74% in the adjuvant chemotherapy group versus 71% in
Cycle one 450 (100%) 451 (99%)
the chemoradiotherapy only group, with an absolute
Cycle two 438 (98%) 434 (96%)
difference of 3% (95% CI –4 to 11) in 5-year overall
Cycle three 431 (97%) 423 (95%)
survival between the treatment groups (HR 0·81 [95% CI
Cycle four 420 (95%) 409 (93%)
0·60 to 1·08]; p=0·15). A similar pattern was seen for
Cycle five 383 (88%) 375 (87%)
progression-free survival (appendix p 11).
Cisplatin dose intensity, 28 (27–29) 28 (26–29)
mg/m² per week
Sites of disease recurrence were similar in the randomly
Radiation technique
assigned groups; most patients were disease free at last
Four field 413 (91%) 416 (92%)
follow-up. In the ITT population, persistent disease after
Two field 7 (2%) 5 (1%)
treatment was reported in five (1%) of 463 patients in the
adjuvant chemotherapy group versus 15 (3%) of 456
Other 33 (7%) 33 (7%)
patients in the chemoradiotherapy only group. Isolated
Radiotherapy
locoregional recurrence was reported in 54 (12%) patients
Mean EBRT dose given, Gy 45·6 45·7
in the adjuvant chemotherapy group versus 50 (11%) in the
EBRT nodal boost given 145 (32%) 135 (30%)
chemoradiotherapy only group. Distant recurrence was
EBRT parametrial boost 161 (36%) 165 (36%)
given reported in 61 (13%) patients in the adjuvant chemotherapy
EBRT without interruption 417 (92%) 418 (92%)
group versus 70 (15%) in the chemoradiotherapy only
Brachytherapy given 429 (95%) 426 (94%)
group. Sites of other or unknown recurrence were reported
Brachtherapy dose rate
in 11 (2%) patients in the adjuvant chemotherapy group
High-dose rate 393 (92%) 384 (90%)
versus 15 (3%) in the chemoradiotherapy only group.
There was no evidence of difference in sites of recurrence
Low-dose rate 25 (6%) 24 (6%)
(none vs locoregional vs distant vs other or unknown)
Pulse-dose rate 9 (2%) 16 (4%)
between the randomly assigned groups (p=0·12).
Not recorded 1 (<1%) 4 (1%)
Chemotherapy was the most common treatment given for
Brachytherapy prescription
recurrent disease (appendix p 73).
Point A 292 (64%) 292 (63%)
In the safety population, grade 2 or worse adverse
Image-guided 134 (29%) 131 (28%)
events were reported in the first year after random
Not recorded 30 (7%) 40 (9%)
isation in 356 (99%) of 361 patients in the adjuvant
Duration of radiation
chemotherapy group versus 409 (90%) of 453 patients in
<8 weeks 278 (63%) 281 (64%)
the chemoradiotherapy only group. Grade 3 or worse
8–10 weeks 141 (32%) 143 (33%)
adverse events were reported in 292 (81%) patients in
>10 weeks 19 (4%) 14 (3%) the adjuvant chemotherapy group versus 280 (62%)
Chemoradiotherapy completed patients in the chemoradiotherapy only group
Five cisplatin cycles, 45 Gy 353 (77%) 356 (77%) (p<0·0001). Grade 1–2 adverse events reported in at
EBRT, and brachytherapy
least 10% of patients and grade 3–5 events reported in
Minimum four cisplatin 396 (87%) 383 (83%)
cycles, 45 Gy EBRT, and 1% of patients are shown in table 3. The most common
brachytherapy grade 3–4 adverse events were decreased lymphocyte
Data are median (IQR) or n (%), unless otherwise indicated. EBRT=external beam
count (211 [58%] of 361 patients in the adjuvant
radiotherapy. chemotherapy group vs 208 [46%] of 453 in the
chemoradiotherapy only group), decreased neutrophils
Table 2: Chemoradiotherapy adherence in the intention-to-treat
population
(71 [20%] vs 34 [8%]), and anaemia (66 [18%] vs 34 [8%];
appendix pp 12–43). Serious adverse events occurred in
107 (30%) patients in the adjuvant chemotherapy group
21 with other or unknown cause in the versus 98 (22%) in the chemoradiotherapy only group,
chemoradiotherapy only group. The 5-year cumulative with infectious adverse events being the most
incidence of cervical cancer-specific death was also common. Established adverse events associated with
similar in in both groups (21% [95% CI 17–26] vs 24% carboplatin and paclitaxel, such as nausea, vomiting,
[20–28]; p=0·21). haematological toxicity, alopecia, fatigue, myalgia, and

Articles
100 Hazard ratio 0·90 (95% CI 0·70–1·17); p=0·81

90
80
70
Overall survival (%)
60
50
40
30 Events at 5-year overall

5 years survival
20
10 Chemoradiotherapy only group (n=456) 116 deaths 71% (95% CI 66–75)

Adjuvant chemotherapy group (n=463) 105 deaths 72% (95% CI 67–76)
0
0 12 24 36 48 60
Time since randomisation (months)
Number at risk
(cumulative number censored)
Adjuvant chemotherapy group 463 (0) 403 (38) 347 (59) 307 (80) 245 (127) 149 (209)
Chemoradiotherapy only group 456 (0) 417 (20) 343 (45) 306 (58) 244 (102) 164 (176)
Figure 2: Kaplan-Meier estimates of overall survival
100 3-year hazard ratio 0·86 (95% CI 0·69–1·08; p=0·17)

5-year hazard ratio 0·86 (95% CI 0·69–1·08; p=0·58)
90
80
Progression-free survival (%)
70
60
50
40 Events at 3 years Progression-free Events at 5 years Progression-free

survival at 3 years survival at 5 years
30
Chemoradiotherapy 150 events (14 deaths; 66% (95% CI 61–70) 163 events (17 deaths; 62% (95% CI 57–66)
20 only group (n=456) 136 progressions) 146 progressions)
10 Adjuvant chemotherapy 127 events (11 deaths; 70% (95% CI 65–74) 147 events (18 deaths; 63% (95% CI 58–68)
group (n=463) 116 progressions) 129 progressions)
0
0 12 24 36 48 60
Time since randomisation (months)
Number at risk
(cumulative number censored)
Adjuvant chemotherapy group 463 (0) 351 (35) 302 (53) 266 (70) 215 (111) 134 (182)
Chemoradiotherapy only group 456 (0) 335 (18) 275 (43) 255 (51) 210 (88) 137 (156)
Figure 3: Kaplan-Meier estimates of progression-free survival
peripheral neuro pathy, were more frequent in the 369 (91%) completed a questionnaire at baseline (appendix
adjuvant chemotherapy group and resulted in 84 serious p 70). Mean QLQ-C30 global health status and quality of
adverse events in 56 (16%) participants. Febrile life scores were worse in the adjuvant chemotherapy
neutropenia occurred in nine (2%) patients in both group versus the chemoradiotherapy only group for
treatment groups. There were no deaths attributed to 3–6 months after completion of chemoradiotherapy;
study treatment. however, mean status and score were similar between the
Long-term safety data were available for 316 (88%) of groups from months 12 to 36 (appendix pp 71, 74).
361 patients in the adjuvant chemotherapy group and 284 (61%) of 463 patients in the adjuvant chemo
377 (83%) of 453 patients in the chemoradiotherapy therapy group and 252 (55%) of 456 patients in
group. Grade 2 sensory peripheral neuropathy, peripheral the chemoradiotherapy only group had RECIST 1.1
motor neuropathy, pain in extremity, hyperglycaemia, measurable disease at baseline. 150 (53%) of 284 patients
and fever related to chemotherapy were all significantly in the adjuvant chemotherapy group had a complete
more common in the adjuvant chemotherapy group than response and 77 (27%) had a partial response. 124 (49%)
in the chemotherapy alone group (appendix pp 44–58). of 252 patients in the chemoradiotherapy only group had
406 (88%) patients in the adjuvant chemotherapy group a complete response and 68 (27%) had a partial response
and 401 (88%) patients in the chemoradiotherapy group (p=0·55). 244 (53%) of 463 patients in the adjuvant
participated in HRQoL collection, of whom 369 (92%) and chemotherapy group and 223 (49%) of 456 in the

Articles
A
Events/total 5–year overall survival Hazard ratio (95% CI) pinteraction
Chemoradiotherapy Adjuvant Chemoradiotherapy Adjuvant
only group chemotherapy group only group chemotherapy group
Pelvic or common iliac nodal involvement, or both 0·57

Yes 70/221 65/224 65 (57–71) 67 (59–73) 0·91 (0·65–1·27)
No 51/225 42/231 76 (70–82) 78 (71–83) 0·84 (0·56–1·26)
Unknown 2/10 2/8 90 (47–99) 40 (1–83) 4·32 (0·39–47·75)
Requirement for EFRT 0·17
No 99/397 95/404 74 (69–78) 72 (67–77) 0·98 (0·74–1·30)
Yes 24/59 14/59 53 (38–66) 69 (52–81) 0·60 (0·31–1·17)
FIGO 2008 stage 0·38
Stage IB1 and node positive, stage IB2, or stage IIA 31/152 33/154 79 (71–85) 75 (66–82) 1·09 (0·67–1·78)
Stage IIB 46/196 39/197 74 (66–80) 76 (68–82) 0·93 (0·61–1·43)
Stage IIIB or stage IVA 46/108 37/112 55 (45–65) 62 (51–71) 0·71 (0·46–1·09)
Age (years) 0·019
<60 109/390 86/393 70 (65–75) 74 (69–79) 0·79 (0·59–1·05)
≥60 14/66 23/70 77 (64–86) 60 (45–72) 1·89 (0·97–3·68)
Country 0·99
USA and Canada 102/366 88/373 70 (64–75) 71 (65–76) 0·88 (0·66–1·17)
Australia and New Zealand 21/84 19/81 75 (64–83) 77 (66–85) 0·91 (0·49–1·69)
Smoking 0·88
Never smoked 60/237 49/224 74 (67–79) 74 (67–80) 0·92 (0·63–1·34)
Current or former smoker, or unknown 63/219 60/239 69 (61–75) 70 (63–76) 0·88 (0·62–1·25)
Tumour histology (post hoc) 0·98
Squamous or adenosquamous cell carcinoma 102/377 91/395 71 (66–76) 72 (66–77) 0·90 (0·68–1·20)
Adenocarcinoma 21/79 18/68 71 (58–81) 72 (59–82) 0·87 (0·46–1·64)
Overall 123/456 109/463 71 (66–75) 72 (67–76) 0·90 (0·70–1·17)
B Events/total 5-year progression-free survival Hazard ratio (95% CI) pinteraction

Chemoradiotherapy Adjuvant Chemoradiotherapy Adjuvant
only group chemotherapy group only group chemotherapy group
Pelvic or common iliac nodal involvement, or both 0·90

Yes 93/221 84/224 56 (49–63) 58 (50–65) 0·83 (0·62–1·12)
No 72/225 65/231 66 (59–73) 70 (63–76) 0·88 (0·63–1·23)
Unknown 3/10 2/8 70 (33–89) 42 (1–84) 1·09 (0·17–6·81)
Requirement for EFRT 0·32
No 138/397 130/404 64 (59–69) 64 (59–69) 0·91 (0·71–1·15)
Yes 30/59 21/59 45 (31–58) 57 (41–70) 0·68 (0·39–1·19)
FIGO 2008 stage 0·87
Stage IB1 and node positive, stage IB2, or stage IIA 46/152 40/154 70 (61–76) 70 (61–77) 0·80 (0·53–1·23)
Stage IIB 68/196 60/197 64 (57–71) 65 (57–72) 0·92 (0·65–1·31)
Stage IIIB or stage IVA 54/108 51/112 45 (35–55) 51 (40–60) 0·84 (0·57–1·23)
Age (years) 0·015
<60 148/390 121/393 61 (56–66) 65 (60–70) 0·77 (0·60–0·98)
≥60 20/66 30/70 66 (51–77) 53 (39–65) 1·67 (0·95–2·95)
Country 0·96
USA and Canada 142/366 127/373 59 (53–64) 61 (55–66) 0·86 (0·68–1·10)
Australia and New Zealand 26/84 22/81 68 (57–77) 74 (63–82) 0·79 (0·45–1·40)
Smoking 0·77
Never smoked 82/237 71/224 64 (57–70) 64 (56–70) 0·90 (0·65–1·23)
Current or former smoker, or unknown 86/219 80/239 59 (51–65) 63 (56–69) 0·83 (0·61–1·13)
Tumour histology (post hoc) 0·94
Squamous or adenosquamous cell carcinoma 135/377 122/395 63 (58–68) 65 (60–70) 0·86 (0·68–1·10)
Adenocarcinoma 33/79 29/68 53 (40–65) 54 (41–66) 0·90 (0·55–1·48)
Overall 168/456 151/463 62 (57–66) 63 (58–68) 0·86 (0·69–1·08)
0·01 0·10 1·00 10·00 100·00
Favours adjuvant chemotherapy Favours chemoradiotherapy alone

Articles
chemoradiotherapy only group were assessable for PET improved progression-free survival, but the abridged
response according to PERCIST version 1.1. 138 (57%) of follow-up time of 3 years prohibited any robust
244 patients in the adjuvant chemotherapy group conclusion about the effect on overall survival. Our
versus 111 (50%) of 223 in the chemoradiotherapy findings did not show a similar improvement in
only group had a complete metabolic response on PET progression-free survival or overall survival to those
or PET and CT 4 months or longer after completing reported by Dueñas-González and colleagues,9 despite a
chemoradiotherapy (p=0·14). A complete metabolic larger sample size and 5 years of follow-up. Our results
response on PET or PET and CT was associated with are consistent with a 2022 meta-analysis,22 which
improved 5-year overall survival (HR 0·22 [95% CI included two randomised trials and eight matched case-
0·14–0·33]; p<0·0001) and 5-year progression-free control and retrospective studies. Both this meta-
survival (0·24 [0·17–0·33]; p<0·0001). analysis and one done earlier did not find a survival
benefit but showed increased toxicity with the addition
Discussion of adjuvant chemotherapy.22,23 Our results are also
The planned addition of four cycles of paclitaxel plus consistent with the similarly designed ACTLACC trial,24
carboplatin following chemoradiotherapy with cisplatin which used three cycles of adjuvant carboplatin and
for locally advanced cervical cancer did not improve paclitaxel, but closed after interim analysis due to
overall survival or progression-free survival, but did futility. Although we used carboplatin plus paclitaxel
increase the number of adverse events and diminish adjuvant chemotherapy, rather than cisplatin plus
patient-reported health and quality of life during and up gemcitabine, this difference probably does not
to 6 months after treatment. The survival results are account for the different outcomes given that platinum
somewhat surprising given the suggestive evidence for plus paclitaxel is recommended as the most active
adjuvant cytotoxic therapy in this same patient popu regimen to treat metastatic disease.25 The study by
lation in previous trials.6,10,23 This academic study, with its Dueñas-González and colleagues9 also added additional
large sample size, mature follow-up, and multi-national gemcitabine chemotherapy to cisplatin during the
accrual, provides opportunities to generate hypotheses standard chemo radiotherapy. However, a subsequent
for the results seen. trial testing the effect of adding gemcitabine only during
Despite the use of standard cisplatin-based chemo chemoradiotherapy was closed early due to futility.16
radiotherapy, treatment failures still occur. This led to the The OUTBACK trial had some limitations. First is the
design of trials aiming to improve on chemoradiotherapy, poor compliance with planned adjuvant chemotherapy,
which were largely focused on adding cytotoxic or novel with 102 (22%) of 463 patients in the adjuvant
drugs to chemoradiotherapy, and, since 2010, immune chemotherapy group not receiving any planned adjuvant
checkpoint inhibitors.14–18 Of the trials with novels drugs treatment. This is a substantial enough fraction to effect
reported to date, tirapazamine moved into a randomised outcomes; as a result the sample was increased after the
phase 3 trial, but did not improve progression-free trial started to account for this. A similar non-compliance
survival or overall survival.19 Additionally, a presentation rate also occurred in previous trials of adjuvant
at the 2022 IGCS Annual Global Meeting indicated that chemotherapy (eg, 14% in the study by Dueñas-González
adding adjuvant durvalumab to chemoradiotherapy did and colleagues,9 32% in the RTOG 90-01 trial,5 23% in
not improve progression-free survival.20 the ACTLACC trial,24 and 29% in the GOG109 trial21 .
The rationale for adjuvant chemotherapy following Patient preference was the most common reason for not
chemoradiotherapy has its proof of concept in two of commencing planned adjuvant chemotherapy, which
the original positive trials that lead to the US National probably resulted from residual adverse events after
Cancer Institute alert. RTOG 90-015 used one additional standard chemoradiotherapy, particularly fatigue. Before
cycle of cisplatin and fluorouracil following radiotherapy, the primary analysis, we identified an association
and the Gynecological Oncology Group (GOG) protocol between older age (>60 years), non-White race, and not
10921 used two additional cycles of cisplatin and completing the initial standard chemoradiotherapy, with
fluorouracil following radiotherapy after radical not proceeding to any adjuvant therapy.13 This association
hysterectomy.5,21 Subsequently, the trial by Dueñas- suggests a subset of patients in this trial with additional
González and colleagues9 evaluating the addition of susceptibilities that affected their ability to commence
gemcitabine to chemoradiotherapy followed by two assigned adjuvant therapy, which might have included
additional cycles of cisplatin and gemcitabine showed relatively young patients facing financial and social
pressures, and those with concerns about alopecia. The
poor compliance with adjuvant therapy in multiple
trials, including OUTBACK, also raises the hypothesis
of whether future trials of adjuvant systemic therapy
Figure 4: Subgroup analyses of 5-year overall survival (A) and progression-
free survival (B) should randomly assign patients to treatment after
EFRT=extended field radiotherapy. FIGO=International Federation of completion of standard chemoradiotherapy. However,
Gynecology and Obstetrics. HR=hazard ratio. our sensitivity analysis did not find clinically meaningful

Articles
Chemoradiotherapy only group (n=453) Adjuvant chemotherapy group (n=361) p value*

Grade 1–2 Grade 3 Grade 4 Grade 5 Grade 1–2 Grade 3 Grade 4 Grade 5
Abdominal pain 179 (40%) 16 (4%) 0 0 175 (48%) 19 (5%) 0 0 0·0086
Alanine aminotransferase increased 77 (17%) 3 (1%) 1 (<1%) 0 98 (27%) 2 (1%) 0 0 0·0020
Alopecia 40 (9%) 0 0 0 284 (79%) 1 (<1%) 0 0 <0·0001
Anaemia 259 (57%) 34 (8%) 1 (<1%) 0 238 (66%) 66 (18%) 0 0 <0·0001
Anorexia 138 (30%) 5 (1%) 0 0 129 (36%) 2 (1%) 0 0 0·21
Anxiety 62 (14%) 0 1 (<1%) 0 76 (21%) 1 (<1%) 0 0 0·020
Arthralgia 40 (9%) 1 (<1%) 0 0 79 (22%) 1 (<1%) 0 0 <0·0001
Aspartate aminotransferase increased 46 (10%) 2 (<1%) 1 (<1%) 0 84 (23%) 2 (1%) 0 0 <0·0001
Back pain 111 (25%) 5 (1%) 0 0 96 (27%) 4 (1%) 0 0 0·79
Constipation 204 (45%) 1 (<1%) 1 (<1%) 0 192 (53%) 1 (<1%) 0 0 0·067
Creatinine increased 57 (13%) 4 (1%) 1 (<1%) 0 59 (16%) 3 (1%) 0 0 0·30
Cystitis non-infective 102 (23%) 6 (1%) 0 0 95 (26%) 6 (2%) 0 0 0·40
Dehydration 40 (9%) 14 (3%) 0 0 50 (14%) 9 (2%) 0 0 0·071
Depression 57 (13%) 1 (<1%) 0 0 70 (19%) 2 (1%) 1 (<1%) 0 0·012
Dermatitis radiation 64 (14%) 0 1 (<1%) 0 64 (18%) 1 (<1%) 0 0 0·37
Diarrhoea 323 (71%) 21 (5%) 0 0 277 (77%) 21 (6%) 0 0 0·064
Dizziness 53 (12%) 1 (<1%) 0 0 65 (18%) 2 (1%) 0 0 0·028
Dysgeusia 58 (13%) 0 0 0 60 (17%) 0 0 0 0·12
Dyspnoea 58 (13%) 3 (1%) 0 1 (<1%) 78 (22%) 3 (1%) 0 0 0·0037
Dysuria 54 (12%) 0 0 0 58 (16%) 0 0 0 0·088
Oedema limbs 49 (11%) 2 (<1%) 0 0 52 (14%) 0 0 0 0·14
Fatigue 361 (80%) 8 (2%) 0 0 327 (91%) 9 (2%) 0 0 <0·0001
Febrile neutropenia 0 8 (2%) 1 (<1%) 0 0 9 (2%) 0 0 0·63
Female genital tract fistula 10 (2%) 6 (1%) 2 (<1%) 0 6 (2%) 5 (1%) 0 0 0·78
Fever 32 (7%) 0 0 0 54 (15%) 3 (1%) 0 0 0·00017
Headache 94 (21%) 1 (<1%) 0 0 103 (29%) 0 0 0 0·025
Hearing impaired 47 (10%) 0 0 0 51 (14%) 4 (1%) 0 0 0·019
Haemorrhage bladder 76 (17%) 7 (2%) 0 0 54 (15%) 5 (1%) 0 0 0·76
Haemorrhage rectum 62 (14%) 2 (<1%) 0 0 65 (18%) 1 (<1%) 0 0 0·23
Hot flashes 106 (23%) 0 0 0 115 (32%) 2 (1%) 0 0 0·0066
Hyperglycaemia 41 (9%) 8 (2%) 3 (1%) 0 58 (16%) 7 (2%) 3 (1%) 0 0·0087
Hypertension 28 (6%) 17 (4%) 0 0 45 (12%) 12 (3%) 0 0 0·0077
Hypoalbuminaemia 54 (12%) 1 (<1%) 0 0 67 (19%) 2 (1%) 0 0 0·021
Hypocalcaemia 56 (12%) 3 (1%) 0 0 58 (16%) 1 (<1%) 0 0 0·24
Hypokalaemia 69 (15%) 11 (2%) 4 (1%) 0 65 (18%) 17 (5%) 0 0 0·30
Hypomagnesaemia 100 (22%) 2 (<1%) 2 (<1%) 0 111 (31%) 6 (2%) 0 0 0·0096
Hyponatraemia 62 (14%) 3 (1%) 0 0 46 (13%) 12 (3%) 0 0 0·019
Insomnia 96 (21%) 1 (<1%) 0 0 104 (29%) 0 0 0 0·030
Kidney infection 1 (<1%) 5 (1%) 0 0 0 3 (1%) 0 0 0·62
Lymphocyte count decreased 114 (25%) 167 (37%) 41 (9%) 0 77 (21%) 179 (50%) 32 (9%) 0 0·0012
Menopause 8 (2%) 12 (3%) 0 0 5 (1%) 16 (4%) 0 0 0·35
Myalgia 52 (11%) 0 0 0 141 (39%) 3 (1%) 0 0 <0·0001
Nausea 335 (74%) 14 (3%) 0 0 296 (82%) 11 (3%) 0 0 0·016
Neutrophil count decreased 84 (19%) 27 (6%) 7 (2%) 0 117 (32%) 61 (17%) 10 (3%) 0 <0·0001
Pain 41 (9%) 3 (1%) 0 0 44 (12%) 3 (1%) 0 0 0·33
Pain in extremity 94 (21%) 3 (1%) 0 0 123 (34%) 2 (1%) 0 0 0·00011
Pelvic pain 146 (32%) 11 (2%) 0 0 138 (38%) 8 (2%) 0 0 0·20
Peripheral motor neuropathy 19 (4%) 0 0 0 72 (20%) 4 (1%) 0 0 <0·0001
Peripheral sensory neuropathy 130 (29%) 1 (<1%) 0 0 271 (75%) 16 (4%) 0 0 <0·0001
Platelet count decreased 140 (31%) 3 (1%) 2 (<1%) 0 192 (53%) 14 (4%) 2 (1%) 0 <0·0001
(Table 3 continues on next page)

Articles
Chemoradiotherapy only group (n=453) Adjuvant chemotherapy group (n=361) p value*

Grade 1–2 Grade 3 Grade 4 Grade 5 Grade 1–2 Grade 3 Grade 4 Grade 5
(Continued from previous page)
Premature menopause ·· 11 (2%) 0 0 1 (<1%) 17 (5%) 0 0 0·11
Proctitis 49 (11%) 2 (<1%) 0 0 34 (9%) 0 0 0 0·36
Sepsis 0 1 (<1%) 8 (2%) 0 0 1 (<1%) 3 (1%) 0 0·32
Small intestinal obstruction 1 (<1%) 5 (1%) 0 0 1 (<1%) 8 (2%) 1 (<1%) 0 0·31
Syncope 2 (<1%) 5 (1%) 0 0 4 (1%) 5 (1%) 0 0 0·51
Thrombosis, thrombus, or embolism 19 (4%) 7 (2%) 0 0 18 (5%) 5 (1%) 2 (1%) 0 0·78
Tinnitus 84 (19%) 0 0 0 87 (24%) 1 (<1%) 0 0 0·079
Urinary frequency 92 (20%) 0 0 0 84 (23%) 0 0 0 0·31
Urinary incontinence 69 (15%) 0 0 0 68 (19%) 1 (<1%) 0 0 0·20
Urinary tract infection 63 (14%) 17 (4%) 1 (<1%) 0 51 (14%) 17 (5%) 0 0 0·87
Urinary tract obstruction 1 (<1%) 10 (2%) 0 0 1 (<1%) 8 (2%) 0 0 0·99
Urinary tract pain 45 (10%) 1 (<1%) 0 0 56 (16%) 0 0 0 0·039
Urinary urgency 40 (9%) 0 0 0 44 (12%) 0 0 0 0·12
Vaginal discharge 167 (37%) 0 0 0 147 (41%) 0 0 0 0·26
Vaginal dryness 56 (12%) 0 0 0 53 (15%) 1 (<1%) 0 0 0·33
Vaginal haemorrhage 164 (36%) 8 (2%) 1 (<1%) 0 133 (37%) 7 (2%) 1 (<1%) 0 0·95
Vaginal pain 65 (14%) 3 (1%) 0 0 47 (13%) 2 (1%) 0 0 0·84
Vaginal stricture 57 (13%) 10 (2%) 0 0 39 (11%) 5 (1%) 0 0 0·49
Vomiting 165 (36%) 11 (2%) 0 0 166 (46%) 15 (4%) 0 0 0·0042
Weight loss 46 (10%) 6 (1%) 0 0 52 (14%) 6 (2%) 0 0 0·16
White blood cell decreased 80 (18%) 16 (4%) 3 (1%) 0 74 (20%) 31 (9%) 7 (2%) 0 0·00066
Data are n (%). Data are for adverse events grade 1–2 occurring in at least 10% of patients and grade 3–5 events in at least 1% of patients. *χ² test comparing none versus mild
(grade 1–2) versus severe (grade 3–5) between the treatment groups. Data are for adverse events grade 1–2 occurring in at least 10% of patients and all grade 3–5 events are
in the appendix (pp 59–68).
Table 3: Summary of adverse events, safety population
improvements in survival, even in the subset of patients By comparison, in the study by Dueñas-González and
who completed initial chemoradiotherapy, which was colleagues,9 the mean external beam dose was 50·4 Gy
the group most likely to complete assigned adjuvant with a median duration of overall radiation of 49 days
therapy. The ongoing INTERLACE trial (NCT01566240) in the adjuvant chemotherapy group and 45 days in
aims to answer the question about whether giving a the control group. Because adherence to radiotherapy
short course of once-a-week chemotherapy before was similar between the adjuvant chemotherapy and
chemoradiotherapy could improve patient outcomes and chemoradiotherapy only group these differences are
be more deliverable unlikely to have affected the primary endpoint of the
Second, we used four-field radiotherapy and OUTBACK trial. Additionally, 3-year progression-free
brachytherapy to help generalise radiation availability survival rates were the same in the chemoradiotherapy
for this global trial. Athough clinical practise in some only group of our study and in the control group of the
high-income countries is now moving to more study by Dueñas-González and colleagues9 suggesting
sophisticated radiotherapy planning, such as intensity- that the differences in radiation adherence do not
modulated radiation therapy (IMRT), the type of explain the discordant findings between the two trials.
radiotherapy used in the OUTBACK trial is still widely Emerging data suggest that improved imaging and
used and accepted standard treatment in the radiotherapy delivery might identify tumours at increased
2023 National Comprehensive Cancer Network risk for treatment failure and that might benefit from
guidelines. Only 77% of patients in this trial completed additional therapy. The EMBRACE I multicentre,
planned chemoradiotherapy (all five doses of cisplatin), prospective cohort study26 confirmed the benefit of
irrespective of allocated treatment. Radio therapy was image-guided brachytherapy for locally advanced cervical
completed within 8 weeks in approximately 65% of cancer, which was given to approximately a third
patients with a mean dose of 45·6 Gy, indicating room of patients in the OUTBACK trial. In subset analysis of
for improvement. Impressively, brachytherapy was given EMBRACE I, clinical and pathological factors associated
to 95% of patients in the chemoradiotherapy only group with local treatment failure, included having any positive
and 94% in the adjuvant chemotherapy group. common iliac lymph nodes (but negative para-aortic

Articles
nodes). Use of elective para-aortic radiation in this higher FIGO 2008 to FIGO 2018 staging transition. For example,
risk group decreased para-aortic treatment failures.26 The in a previous OUTBACK analysis the 5-year progression-
ongoing EMBRACE II study (NCT03617133) is evaluating free survival rate for FIGO 2008 stage III and IVA disease
IMRT and volumetric modulated arc therapy (VMAT) for was 48%, but the 5-year overall survival rate for FIGO 2018
elective para-aortic radiotherapy and integrating radiation stage III and IVA disease was 56%.27 The OUTBACK
boost for cervical cancer to improve local control and re-analysis provides helpful information for benchmarking
decrease distant failures. An important question is control group expectations for future studies using the
whether there is more to be gained by adding more drugs 2018 staging system.
to chemoradiotherapy, or instead ensuring that excellent In conclusion, adjuvant chemotherapy with four cycles
quality primary chemoradiotherapy is available to all of carboplatin and paclitaxel did not improve overall
women with cervical cancer. Furthermore, new less-toxic survival or progression-free survival after chemoradio
radiation techniques could make testing the benefit of therapy with once-a-week cisplatin for locally advanced
giving additional systemic therapies more feasible. cervical cancer. Use of adjuvant carboplatin and paclitaxel
Like our study, the EMBRACE I trial26 recruited a following chemoradiotherapy should not be used in this
population of patients that was felt to be at high risk for disease setting. Future studies should select participants
recurrence, but ultimately was not found to be particularly with high-risk disease and overcome barriers to adherence
high risk. A 2022 exploratory analysis of the OUTBACK with treatment.
trial27 re-evaluated outcomes when the tumours were Contributors
restaged on the basis of FIGO 2018 staging criteria, LRM, KNM, VG, KN, MTK, DR, MQ, MRS, and BJM conceptualised
which now incorporates nodal status. In this analysis, the the study. LRM, EHB, VG, KN, MTK, NB, YCL, and KD curated and
verified the study data. EHB and VG did the formal analysis. LRM,
FIGO 2008 stage III and IV (T3 and T4 lesions) had a KNM, KN, DR, MQ, VG, and MRS acquired the funding. LRM and
5-year progression-free survival rate of 48% and 5-year KNM wrote the original draft of the manuscript. LRM, KNM, MQ, DR,
overall survival rate of 58%.27 These more advanced MRS, BJM, DKG, WSJ, VG, and EHB refined the methods. LRM,
tumours do appear to be at the highest risk of recurrence, KNM, NB, and KD were project administrators. EHB and VG did the
statistical analysis. LRM, KNM, MQ, DR, MRS, BJM, KN, MTK, and
and patients with these tumours might benefit from any VG supervised the study. LRM, KNM, MRS, BJM, EHB, VG, KN, and
addition to chemoradiotherapy. However, in our study SVD validated the data. LRM, EHB, YCL, and KD visualised the data.
and others,24 those with higher stage tumours or node- LRM, KNM, EHB, VG, KN, MTK, NB, YCL, KD, AWF, WSJ, DKG, PK,
positive disease received no benefit from the addition of SB, JST, WKH, CAM, MB, AS, TEL, IJB, CHH, SVD, MQ, DR, BJM,
and MRS participated in the investigation, collected resources, and
adjuvant chemotherapy. reviewed and edited the manuscript. All authors had full access to all
Finally, in the past, when no effective interventions the data in the study and accept responsibility for the decision to
were available after systemic chemotherapy with or submit for publication.
without bevacizumab for advanced or recurrent disease, Declaration of interests
overall survival was the best endpoint because no LRM reports grant from National Health and Medical Research Council
interventions were expected to confound the results.28 (NHMRC) Australia for support for the present manuscript.
KNM reports research funding to institution outside of submitted work
With the approval of immune checkpoint inhibitors used from PTC Therapeutics, Lilly, Merck, and GSK/Tesaro; participation on
concomitantly with systemic chemotherapy with or an Advisory Board for AstraZeneca, Aravive, Alkemeres, Addi,
without bevacizumab for advanced or metastatic disease Blueprint Medicines, Eisai, EMD Serono, Elevar Therapeutics, GSK/
Tesaro, Genentench/Roche, Hengrui, Immunogen, INXmed, IMab,
and approvals for monotherapy use for second-line
Lilly, Mereo, Mersana Therapeutics, Merck, Myriad, Novartis,
treatment of metastatic disease, the expected overall Novocure, OncXerna, Onconova Therapeutics, Tarveda Therapeutics,
survival has improved and these drugs could confound VBL Therapeutics, and Verastem Oncology; and leadership role as
outcomes depending on use and availability.29,30 With new Associate Director of Gynecological Oncology Group Partners and on
the Board of Directors for GOG Foundation. WSJ reports honoraria for
drugs in the pipeline, one has to consider whether
invited talks from Carl Zeiss; payments made for participation on a
progression-free survival might be an appropriate data and safety monitoring board from Novocure; and Co-Chair of
primary endpoint in lieu of or in combination with Gynaecological Committee until June 2020 with NRG Oncology. DKG
overall survival for future studies. However, these newer reports grants and contracts from Elekta; payment or honoraria from
ANZGOG, payment for expert testimony from in a Montana court case,
therapies are unlikely to have confounded overall survival
participating on a DSMB or Advisory Board for Merck; and leadership
in OUTBACK because they were not available to patients role in NRG Oncology as Co-Chair of the Gyn Committee. PK reports
in study sites during the study timeframe. funding outside of this submitted work from ANZGOG Fund for New
Future studies should focus on participants with high Research and Medical Research Future Fund Reproductive Cancers
Grant; is the Deputy Chair of Treatment Pillar for the National Cervical
risk disease who have the most to gain from additional Cancer Elimination Strategy; is on the Board Director for ANZGOG
treatment, and avoid over-treating patients who do well and AMA Victoria. WKH reports consulting fee payments from
with standard of care treatments. Subsequent translational AstraZeneca and SeaGen and payment for participation on a DSMB
research from the OUTBACK trial will be important from Syneos and Inovio. CAM reports support for the present
manuscript from National Cancer Institute, and funding to their
to improve understanding of which groups of patients institution from Syros, Deciphera, Astellas Pharma GlaxoSmithKline/
are most at risk of disease recurrence. Furthermore, Tesaro, Seagen, Genmab, EMD Serono, Merck, Regeneron, Moderna,
benchmarks for studies moving forward will have to Astra Zeneca, and Genentech, outside of the submitted work.
account for the stage migration that occurred with the TEL reports support for the present manuscript from the National

Articles
Cancer Institute Community Research Program and support for 8 Narayan K, Fisher RJ, Bernshaw D, Shakher R, Hicks RJ. Patterns
attending meetings and travel from SHCC MU NCORP Grant. of failure and prognostic factor analyses in locally advanced
CHH reports support for the present manuscript from GOG and NRG, cervical cancer patients staged by positron emission tomography
and leadership role in the Executive Board, Western Association of and treated with curative intent. Int J Gynecol Cancer 2009;
Gynecologic Oncology, and the 2023 Program Committee Chair, 19: 912–18.
Western Association of Gynecologic Oncology. MQ reports personal 9 Dueñas-González A, Zarbá JJ, Patel F, et al. Phase III, open-label,
payment as Chair of an independent data and safety monitoring randomized study comparing concurrent gemcitabine plus cisplatin
and radiation followed by adjuvant gemcitabine and cisplatin versus
committee for the Garnet Study and Tesaro Study from GSK, and Chair
concurrent cisplatin and radiation in patients with stage IIB to IVA
of DSMB for the COMPASS Study. DR reports trial funding all to carcinoma of the cervix. J Clin Oncol 2011; 29: 1678–85.
institution outside of submitted work from MSD, GSK, Regeneron,
10 Nasioudis D, Smith AJ, Latif N. Adjuvant chemotherapy for patients
Roche, Bristol Myers Squibb, Kura, ALX Oncology and participation on with locally advanced cervical carcinoma receiving definitive
a DSMB for MSD, Regeneron, Sanofi. BJM reports consulting fees chemoradiation: utilization and outcomes before the OUTBACK
from Acrivon, Adaptimune, Agenus, Akeso Bio, Amgen, Aravive, trial. 2022 Society of Gynaecological Oncology Annual Meeting;
AstraZeneca, Bayer, Clovis, Easai, Elevar, EMD Merck, Genmab/ March 18–21, 2022 (abstr 361).
Seagen, GOG Foundation, Gradalis, Heng Rui, ImmunoGen, 11 Aaronson NK, Ahmedzai S, Bergman B, et al. The European
Karyopharm, Iovance, Laekna, Macrogenics, Merck, Mersana, Myriad, Organization for Research and Treatment of Cancer QLQ-C30:
Novartis, Novocure, OncoC4, Panavance, Pieris, Pfizer, Puma, a quality-of-life instrument for use in international clinical trials in
Regeneron, Roche/Genentech, Sorrento, TESARO/GSK, US Oncology oncology. J Natl Cancer Inst 1993; 85: 365–76.
Research, and VBL Therapeutics, Verastem, Zentalis and payment or 12 Lin DY, Wei LJ, Ying Z. Checking the Cox model with
honoraria from AstraZeneca, Merck, Myriad, TESARO/GSK, Roche/ cumulative sums of Martingale-based residuals. Biometrika 1993;
Genentech, Clovis, and Easai. MRS reports grant to institution from 80: 557–72.
NHMRC Australia for support for the present manuscript, and grants 13 Mileshkin L, Barnes E, Moore K, et al. Disparities starting adjuvant
to their institution from Amgen, Astellas, AstraZeneca, Bayer, chemotherapy for locally advanced cervix cancer in the
BeiGene, Bristol Myers Squibb, Celgene, Medivation, Merck Sharp & international, academic, randomised, phase 3 OUTBACK trial
Dohme, Novartis, Pfizer, Roche, Sanofi, and Tilray. All other authors (ANZGOG 0902, RTOG 1174, NRG 0274). ESMO Congress;
Sept 16–21, 2021 (abstr 1923).
declare no competing interests.
14 Duska LR, Scalici JM, Temkin SM, et al. Results of an early safety
Data sharing analysis of a study of the combination of pembrolizumab and pelvic
Deidentified study data are available for sharing. To request access to the chemoradiation in locally advanced cervical cancer. Cancer 2020;
deidentified study data, please contact the corresponding author. 126: 4948–56.
Requests will be reviewed by the Trial Management Committee and 15 Rose PG, Sill MW, McMeekin DS, et al. A phase I study of
written applications from investigators with the academic capability and concurrent weekly topotecan and cisplatin chemotherapy with
credibility to undertake the work proposed will be considered. The whole pelvic radiation therapy in locally advanced cervical cancer:
scientific merit of the proposal, including the appropriate methods, a gynecologic oncology group study. Gynecol Oncol 2012;
analysis, and publication plan will be assessed. Consideration will be 125: 158–62.
taken of any overlap with analyses already undertaken or planned to be 16 Wang CC, Chou HH, Yang LY, et al. A randomized trial comparing
undertaken by the study team. If a proposal is approved, a signed data concurrent chemoradiotherapy with single-agent cisplatin versus
cisplatin plus gemcitabine in patients with advanced cervical cancer:
transfer agreement will be required before data sharing.
an Asian Gynecologic Oncology Group study. Gynecol Oncol 2015;
Acknowledgments 137: 462–67.
We thank the women who participated and their families; staff at sites 17 Rischin D, Narayan K, Oza AM, et al. Phase 1 study of tirapazamine
in the USA, Australia, New Zealand, Canada, China, Singapore, and in combination with radiation and weekly cisplatin in patients with
Saudi Arabia; staff at the NHMRC Clinical Trials Centre, The University locally advanced cervical cancer. Int J Gynecol Cancer 2010; 20: 827–33.
of Sydney, Australia New Zealand Gynaecological Oncology Group, and 18 Kunos CA, Sherertz TM. Long-term disease control with triapine
NRG; Australian NHMRC (APP1044349) and US National Cancer based radiochemotherapy for patients with stage IB2-IIIB cervical
Institute for financial support; Hospira for providing paclitaxel to the cancer. Front Oncol 2014; 4: 184.
Australia and New Zealand sites; and the OUTBACK IDSMC members. 19 DiSilvestro PA, Ali S, Craighead PS, et al. Phase III randomized
trial of weekly cisplatin and irradiation versus cisplatin and
References tirapazamine and irradiation in stages IB2, IIA, IIB, IIIB, and IVA
1 Kamangar F, Dores GM, Anderson WF. Patterns of cancer cervical carcinoma limited to the pelvis: a Gynecologic Oncology
incidence, mortality, and prevalence across five continents: defining Group study. J Clin Oncol 2014; 32: 458–64.
priorities to reduce cancer disparities in different geographic 20 Monk BJ, Toita T, Wu X, Carlos J. Durvalumab, in combination with
regions of the world. J Clin Oncol 2006; 24: 2137–50. and following chemoradiotherapy, in locally advanced cervical
2 Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: cancer: results from the phase 3 international, randomized, double-
GLOBOCAN estimates of incidence and mortality worldwide for blind, placebo-controlled CALLA trial. IGCS Annual Global
36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209–49. Meeting; Sept 29 to Oct 1, 2022 (abstr Plenary 01).
3 Spayne J, Ackerman I, Milosevic M, Seidenfeld A, Covens A, 21 Peters WA 3rd, Liu PY, Barrett RJ 2nd, et al. Concurrent
Paszat L. Invasive cervical cancer: a failure of screening. chemotherapy and pelvic radiation therapy compared with pelvic
Eur J Public Health 2008; 18: 162–65. radiation therapy alone as adjuvant therapy after radical surgery in
4 Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and high-risk early-stage cancer of the cervix. J Clin Oncol 2000;
adjuvant hysterectomy compared with radiation and adjuvant 18: 1606–13.
hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 22 Horeweg N, Mittal P, Gradowska PL, Boere I, Nout RA, Chopra S.
1999; 340: 1154–61. A systematic review and meta-analysis of adjuvant chemotherapy
5 Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent after chemoradiation for locally advanced cervical cancer.
chemotherapy compared with pelvic and para-aortic radiation for Crit Rev Oncol Hematol 2022; 172: 103638.
high-risk cervical cancer. N Engl J Med 1999; 340: 1137–43. 23 Tangjitgamol S, Katanyoo K, Laopaiboon M, Lumbiganon P,
6 Rose PG, Bundy BN, Watkins EB, et al. Concurrent cisplatin-based Manusirivithaya S, Supawattanabodee B. Adjuvant chemotherapy
radiotherapy and chemotherapy for locally advanced cervical cancer. after concurrent chemoradiation for locally advanced cervical
N Engl J Med 1999; 340: 1144–53. cancer. Cochrane Database Syst Rev 2014; 12: CD010401.
7 Chemoradiotherapy for Cervical Cancer Meta-Analysis 24 Tangjitgamol S, Tharavichitkul E, Tovanabutra C, et al.
Collaboration. Reducing uncertainties about the effects of A randomized controlled trial comparing concurrent
chemoradiotherapy for cervical cancer: a systematic review and chemoradiation versus concurrent chemoradiation followed by
meta-analysis of individual patient data from 18 randomized trials. adjuvant chemotherapy in locally advanced cervical cancer patients:
J Clin Oncol 2008; 26: 5802–12. ACTLACC trial. J Gynecol Oncol 2019; 30: e82.

Articles
25 Kitagawa R, Katsumata N, Shibata T, et al. Paclitaxel plus 28 Tewari KS, Sill MW, Penson RT, et al. Bevacizumab for advanced
carboplatin versus paclitaxel plus cisplatin in metastatic or cervical cancer: final overall survival and adverse event analysis of a
recurrent cervical cancer: the open-label randomized phase III trial randomised, controlled, open-label, phase 3 trial (Gynecologic
JCOG0505. J Clin Oncol 2015; 33: 2129–35. Oncology Group 240). Lancet 2017; 390: 1654–63.
26 Peters M, de Leeuw AAC, Nomden CN, et al. Risk factors for nodal 29 Colombo N, Dubot C, Lorusso D, et al. Pembrolizumab for
failure after radiochemotherapy and image guided brachytherapy in persistent, recurrent, or metastatic cervical cancer. N Engl J Med
locally advanced cervical cancer: an EMBRACE analysis. 2021; 385: 1856–67.
Radiother Oncol 2021; 163: 150–58. 30 Tewari KS, Monk BJ, Vergote I, et al. Survival with cemiplimab in
27 Mileshkin L, Moore KN, Barnes EH, et al. Staging locally advanced recurrent cervical cancer. N Engl J Med 2022; 386: 544–55.
cervical cancer with FIGO 2018 versus FIGO 2008: impact on
overall survival and progression-free survival in the OUTBACK trial
(ANZGOG 0902, RTOG 1174, NRG 0724). Proc Am Soc Clin Oncol
2022; 40: 5531 (abstr).


Articles: Background

Uploaded by

Copyright:

Available Formats

Articles: Background

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Articles: Background

Uploaded by

Copyright:

Available Formats

Articles

Adjuvant chemotherapy following chemoradiotherapy as

468 www.thelancet.com/oncology Vol 24 May 2023

www.thelancet.com/oncology Vol 24 May 2023 469

470 www.thelancet.com/oncology Vol 24 May 2023

www.thelancet.com/oncology Vol 24 May 2023 471

472 www.thelancet.com/oncology Vol 24 May 2023

ClinicalTrials.gov, NCT01414608, and the Australia New

www.thelancet.com/oncology Vol 24 May 2023 473

Forest plots and p values for interactions showed no

474 www.thelancet.com/oncology Vol 24 May 2023

100 Hazard ratio 0·90 (95% CI 0·70–1·17); p=0·81

30 Events at 5-year overall

10 Chemoradiotherapy only group (n=456) 116 deaths 71% (95% CI 66–75)

Figure 2: Kaplan-Meier estimates of overall survival

100 3-year hazard ratio 0·86 (95% CI 0·69–1·08; p=0·17)

40 Events at 3 years Progression-free Events at 5 years Progression-free

Figure 3: Kaplan-Meier estimates of progression-free survival

www.thelancet.com/oncology Vol 24 May 2023 475

Pelvic or common iliac nodal involvement, or both 0·57

B Events/total 5-year progression-free survival Hazard ratio (95% CI) pinteraction

Pelvic or common iliac nodal involvement, or both 0·90

0·01 0·10 1·00 10·00 100·00

Favours adjuvant chemotherapy Favours chemoradiotherapy alone

476 www.thelancet.com/oncology Vol 24 May 2023

www.thelancet.com/oncology Vol 24 May 2023 477

Chemoradiotherapy only group (n=453) Adjuvant chemotherapy group (n=361) p value*

478 www.thelancet.com/oncology Vol 24 May 2023

Chemoradiotherapy only group (n=453) Adjuvant chemotherapy group (n=361) p value*

Table 3: Summary of adverse events, safety population

www.thelancet.com/oncology Vol 24 May 2023 479

480 www.thelancet.com/oncology Vol 24 May 2023

www.thelancet.com/oncology Vol 24 May 2023 481

482 www.thelancet.com/oncology Vol 24 May 2023

You might also like