The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: a prospective, blinded, multi-centre, observational, cohort study

Background: Neoadjuvant cancer treatment is associated with improved survival following major oesophagogastric cancer surgery. The impact of neoadjuvant chemo/chemoradiotherapy on physical fitness and operative outcomes is however unclear. This study aims to investigate the impact of neoadjuvant chemo/chemoradiotherapy on fitness and post-operative mortality. Methods: Patients with oesophagogastric cancer scheduled for chemo/chemoradiotherapy and surgery were recruited to a prospective, blinded, multi-centre, observational cohort study. Primary outcomes were changes in fitness with chemo/chemoradiotherapy, measured using cardiopulmonary exercise testing and its association with mortality one-year after surgery. Patients were followed up for re-admission at 30-days, in-hospital morbidity and quality of life (exploratory outcomes). Results: In total, 384 patients were screened, 217 met the inclusion criteria, 160 consented and 159 were included (72% male, mean age 65 years). A total of 132 patients (83%) underwent chemo/chemoradiotherapy, 109 (71%) underwent chemo/chemoradiotherapy and two exercise tests, 100 (63%) completed surgery and follow-up. A significant decline in oxygen uptake at anaerobic threshold and oxygen uptake peak was observed following chemo/chemoradiotherapy: -1.25ml.kg -1.min -1 (-1.80 to -0.69) and -3.02ml.kg -1.min -1 (-3.85 to -2.20); p<0.0001). Baseline chemo/chemoradiotherapy anaerobic threshold and peak were associated with one-year mortality (HR=0.72, 95%CI 0.59 to 0.88; p=0.001 and HR=0.85, 0.76 to 0.95; p=0.005). The change in physical fitness was not associated with one-year mortality. Conclusions: Chemo/chemoradiotherapy prior to oesophagogastric cancer surgery reduced physical fitness. Lower baseline fitness was associated with reduced overall survival at one-year. Careful consideration of fitness prior to chemo/chemoradiotherapy and surgery is urgently needed.

Patients with oesophagogastric cancer scheduled for chemo/chemoradiotherapy and surgery were recruited to a prospective, blinded, multi-centre, observational cohort study.Primary outcomes were changes in fitness with chemo/chemoradiotherapy, measured using cardiopulmonary exercise testing and its association with mortality one-year after surgery.Patients were followed up for re-admission at 30-days, inhospital morbidity and quality of life (exploratory outcomes).

Conclusions:
Chemo/chemoradiotherapy prior to oesophagogastric cancer surgery reduced physical fitness.Lower baseline fitness was associated with reduced overall survival at one-year.Careful consideration of fitness prior to chemo/chemoradiotherapy and surgery is urgently needed.

Plain language summary
Background:Cancer treatments such as chemotherapy and radiotherapy given to people with oesophageal and gastric cancer (also known as cancer of the food pipe/stomach) before surgery can improve survival.However, the impact such treatments have on fitness and recovery after surgery is unclear.The aim of this research was to understand the impact cancer treatments has on fitness and any complications after surgery.Methods: Patients with oesophageal and gastric cancer (also known as cancer of the food pipe/stomach) who were being treated by cancer treatment and surgery were recruited from different hospitals in the UK.All participants were asked to undertake an exercise test to measure fitness and fill out questionnaires to measure quality of life before and after cancer treatment.Complications patients experienced after surgery, the number of patients who had to be readmitted to hospital 30 days after surgery and one-year survival was recorded.

Plain language summary
Background: Cancer treatments such as chemotherapy and radiotherapy given to people with oesophageal and gastric cancer (also known as cancer of the food pipe/stomach) before surgery can improve survival.However, the impact such treatments have on fitness and recovery after surgery is unclear.The aim of this research was to understand the impact cancer treatments has on fitness and any complications after surgery.
Methods: Patients with oesophageal and gastric cancer (also known as cancer of the food pipe/stomach) who were being treated by cancer treatment and surgery were recruited from different hospitals in the UK.All participants were asked to undertake an exercise test to measure fitness and fill out questionnaires to measure quality of life before and after cancer treatment.Complications patients experienced after surgery, the number of patients who had to be readmitted to hospital 30 days after surgery and one-year survival was recorded.

Introduction
For patients with locally advanced oesophagogastric (OG) cancer, multimodal therapy incorporating surgery and neoadjuvant chemotherapy or combined chemoradiotherapy (referred to as chemo/chemoradiotherapy herein) offers improved survival over surgical therapy alone 1-4 .However, improvement in overall survival may come at the cost of increased treatment toxicity and mortality in some patients 5,6 .
Physical fitness assessed objectively using Cardiopulmonary Exercise Testing (CPET) may be the best functional predictor of complications following major surgery 7 and is increasingly being adopted in the perioperative setting to guide perioperative care and decision-making 8,9 .Preliminary data from our group and others suggest that chemo/chemoradiotherapy before OG cancer surgery result in a clinically important reduction in physical fitness (oxygen update (VO 2 ) at anaerobic threshold (AT) and VO 2 peak) [10][11][12][13] .In a small, single-centre, pilot, unblinded study, we previously reported that low baseline physical fitness (VO 2 at AT and VO 2 peak) was associated with reduced one-year survival in patients completing chemotherapy and surgery, but not in patients who did not complete chemotherapy 10 .
The aims of this study were to investigate the impact of chemo/ chemoradiotherapy on fitness (VO 2 at AT and VO 2 peak), mortality (at one-year after surgery) and post-operative outcomes (Post-Operative Morbidity Survey and EQ-5D-5L).In this prospective, multi-centre, blinded study, we set out to validate the hypothesis that chemo/chemoradiotherapy was associated with reduced physical fitness and that this change in physical fitness (relative change and change in risk stratification category) would be associated with all-cause mortality at one-year.Further, we explored the hypotheses that reduced fitness following chemo/chemoradiotherapy was associated with increased post-operative morbidity and worse patient reported outcomes.

Study design
This prospective study involved participants with OG cancers scheduled for chemo/chemoradiotherapy followed by elective resection with curative intent.The study protocol, methods and statistical analysis plan are available in open access format 14 .This study was funded by the National Institute for Health (NIHR), Research for Patient Benefit Programme (PB-PG-0609-18262).The research protocol was registered with clinicaltrials.gov(NCT01325883 -30th March 2011) and approved by the Dyfed Powys Research Ethics Committee (11/WA/0072).Written informed consent was obtained from all subjects.The study is described according to the STROBE statement 15 .

Recruiting hospitals
The study was conducted in four NHS hospitals in England: University Hospital Southampton (Southampton), University Hospital Aintree (Aintree), Lancashire Teaching Hospital (Preston) and South Tees Hospital (South Tees).A study management board and independent data and safety monitoring committee oversaw the project.

Eligibility criteria
Briefly, patients with a histologically confirmed, potentially curable (able to undergo chemo/chemoradiotherapy followed by curative elective resection) adenocarcinoma, squamous, or mucinous/ undifferentiated carcinoma of the oesophagus, oesophagogastric junction (i.e.tumours involving both the cardia and the oesophagus on endoscopy) or stomach were eligible for inclusion.Eligible patients were ≥18 years of age, had a World Health Organization (WHO) performance status score of ≤2, and had adequate hematologic, renal, hepatic and pulmonary function, as well as no history of other cancer or previous chemo/ chemoradiotherapy.The study excluded patients who were unable to give informed consent, had non-resectable disease, were unable to perform CPET due to known contra-indication (e.g.lower limb dysfunction), or who declined planned surgery or neoadjuvant cancer treatments.Eligible patients were staged according to a pre-determined protocol.All patients underwent pre-treatment staging based on a pre-determined protocol.This included a medical history, physical examination, pulmonary function tests, routine hematologic and biochemical test, esophago-gastric endoscopy with histologic biopsy +/-endoscopic ultrasound, computer tomography of the neck, chest and abdomen, 18F-fluorodeoxyglucose positronemission tomography and in special circumstances external ultrasonography of the neck, with fine-needle aspiration of lymph nodes when cancer was suspected.Re-staging was undertaken using computer tomography of the chest and abdomen and 18F-fluorodeoxyglucose positron-emission tomography +/-laparoscopy in selected cases.Radiological responses post-chemo/ chemoradiotherapy were based on definitions outlined by the RECIST version 1.1 criteria for solid target lesions 16 .

Study recruitment
All potentially eligible patients were identified at multidisciplinary meetings and approached with written information.All patients provided written informed consent.

Outcome measures
Fitness.CPET was used to assess physical fitness before and following completion of chemo/chemoradiotherapy (approximately four weeks following completion of chemo/ chemoradiotherapy, immediately before planned surgery) and CPET was conducted according to a published protocol 17,18 .All CPETs were performed using identical software and hardware at each recruitment site using an electromagnetically braked cycle ergometer (Ergoline 2000), a 12-lead ECG, non-invasive blood pressure measurement and pulse oximetry, and a metabolic cart (Geratherm Respiratory GmbH, Love Medical Ltd).CPET allowed for the derivation of anaerobic threshold (AT) using the modified V-Slope method.The modified V-Slope method identifies the anaerobic threshold as the tangential breakpoint in the rate of change of VCO 2 relative to VO 2 (oxygen uptake -carbon dioxide output) from the line of unity ('line of one') during the incremental stage of the exercise test.CPET was independently reported by two independent experienced observers (SJ, DL) blinded to CPET timepoint and clinical outcomes, with a third adjudicator (MAW) if >5% variance in VO 2 at AT was observed.All cancer multidisciplinary team members including the treating surgeon, anaesthetist, oncologist and peri-operative teams were blind to all CPET data.

Health-related quality of life (HRQol).
HRQoL was measured using patient reported outcome measure (EQ-5D-5L 19 ) questionnaire before and following completion of chemo/ chemoradiotherapy (approximately four weeks following completion of chemo/chemoradiotherapy, immediately before planned surgery).
Post-operative outcome.All patients were followed-up (by staff blinded to CPET results), using the Post-Operative Morbidity Survey (POMS) 20 at day 3, 5, 8 and 15 post-operatively.Length of hospital stay and critical care length of stay was calculated by subtracting the discharge date from the admission date.The Revised Cardiac Risk Index (RCRI) 21 was calculated preoperatively and the O-POSSUM score was completed postoperatively 22 .

Neoadjuvant chemo/chemoradiotherapy
We did not attempt to standardise chemo/chemoradiotherapy regimes.Chemotherapy regimens included: Epirubicin, Oxaliplatin, Capecitabine (EOX); Epirubicin, Cisplatin, Capecitabine (ECX) 23 ; Epirubicin, Cisplatin, 5-Fluorouracil (ECF) 24 , chemotherapy as part of the STO3 trial -ECX or ECX + Bevacizumab 25 , chemotherapy as part of the OEO5 trial -ECX or Cisplatin and 5-Fluorouracil 26 ; chemoradiotherapy as part of the CROSS trial -Carboplatin, Paclitaxel with concurrent radiotherapy 27 ; chemoradiotherapy as part of the NEOSCOPE trial -Oxaliplatin and Capecitabine or Carboplatin and Paclitaxel with concurrent radiotherapy and induction Oxaliplatin and Capecitabine chemotherapy 28 ; Herceptin, Cisplatin and Capecitabine; Capecitabine alone; and Cisplatin alone.CROSS style radiotherapy was administered at a total radiation dose of 41.4Gy given in 23 fractions of 1.8Gy each, with 5 fractions administered per week, starting on the first date of the first chemotherapy cycle.NEOSCOPE style radiotherapy was administered at a total radiation dose of 45Gy given in 25 fractions of 1.8Gy each, with 5 fractions administered per week, starting on the first date of the first chemotherapy cycle.All patients were treated with external beam radiation.Patients were closely monitored for toxic effects of chemo/chemoradiotherapy using the National Cancer Institute's Common Terminology Criteria for Adverse Events, version 3.0 29 .

Surgery
All patients underwent surgery within 4-6 weeks of chemo/ chemoradiotherapy.Open, hybrid or fully minimal access approaches were used depending on patient characteristics and surgeon preference.A thoracoscopic assisted three-stage esophagectomy or an Ivor-Lewis esophagectomy based on tumour location was undertaken.A transthoracic approach was performed for tumours extending proximally to the tracheal bifurcation.For tumours involving the oesophagogastric junction, an Ivor-Lewis oesophagectomy resection was performed.Gastric tube reconstruction was the preferred technique for restoring intestinal continuity.Gastric surgery consisted of a radical resection of the primary tumour and at least a D1+ lymph node dissection.

Follow-up
During the first year after surgery, patients were followed up for re-admission at 30-days post-operatively, all-cause mortality at 30-days and one-year post-operatively.Additional allcause mortality follow-up was completed at five-year postoperatively.Patients completed an EQ-5D-5L questionnaire at 30-days and one-year post-operatively.

Aims and objectives
Pre-defined primary aims briefly include: 1) observing changes in physical fitness (VO 2 at AT and VO 2 peak) following chemo/chemoradiotherapy, measured using CPET; and 2) interrogating the association of change in physical fitness following chemo/chemoradiotherapy and mortality one-year after surgery.This was evaluated in two ways: A) by evaluating the relative decrease in physical fitness associated with chemo/ chemoradiotherapy and its association with mortality 1-year after surgery; and B) by evaluating whether a change in the risk stratification category (low risk VO 2 at AT >14 ml.kg. - min -1 , medium risk VO 2 at AT 11.0-14.0ml.kg -1 .min - , high-risk VO 2 at AT 8.0-10.9ml.kg -1 .min - , highest risk VO 2 at AT <8.0 ml.kg -1 .min - ) following chemo/chemoradiotherapy would be associated with an increased one-year mortality following surgery when compared with those that do not change their risk stratification category.Similar interrogations were undertaken for mortality at 5-years after surgery.Risk categories were defined a priori 14 .
The trial had several exploratory end points that are detailed in the study protocol 14 .Briefly we explored: 1) the relationships between baseline/relative decrease in physical fitness (VO 2 at AT and VO 2 at Peak) following chemo/ chemoradiotherapy and post-operative in-hospital morbidity (measured by the Post-Operative Morbidity Survey) and patient reported quality of life (measured by EQ-5D-5L); 2) the ability of less fit (VO 2 at AT and VO 2 at Peak) patients to tolerate chemo/chemoradiotherapy compared to patients with a higher fitness; 3) the relationship between patients who tolerate chemo/ chemoradiotherapy poorly and an increase in post-operative outcomes (POMS, one-year mortality, EQ-5D); and 4) if patients undergoing chemoradiotherapy exhibit a greater decline in physical fitness (VO 2 at AT and VO 2 at Peak) compared to patients undergoing chemotherapy.

Sample size calculation
Based on our previously published data 10 demonstrating a standard deviation of the differences in VO 2 at AT values of 3.8 ml.kg .-1.min - , we calculated that 152 patients were needed to detect a difference of 1.0 ml.kg .-1.min - of VO 2 at AT using a paired t-test at the 5% significance level with 90% power (114 patients with 80% power), assuming the standard deviation of these differences is 3.8 ml.kg .-1.min - .To detect a difference in one-year mortality rates of 23% (34% versus 11% -based on published pilot data 10 ) between the two VO 2 at AT change groups [no change/deteriorate], we calculated that 104 patients were required using a chi-squared test at the 5% significance level with 80% power, assuming equal numbers of patients in both groups.

Statistical analysis
All data was inputted by double-data entry and validation was done according to procedures set out in the study data management and data validation plan overseen by the study management group.Detailed statistical methods are described elsewhere 14 .The primary analysis was a comparison of physical fitness (VO 2 at AT) before and after chemo/chemoradiotherapy using a paired t-test.Distributional assumptions were assessed using a normal plot.Other fitness comparisons between independent patient groups were made using the two-sample t-test.The Cox proportional hazards model was used to investigate the relationship between 'change in fitness' and mortality within 5-year and the Kaplan-Meier plot was used to illustrate the survival of different patient groups.Multiple regression with backward elimination (at the 5% level) was used to investigate the relationship between postchemo/chemoradiotherapy fitness and the various pre-chemo/ chemoradiotherapy fitness variables.The predictive ability of both primary aim models to ascertain how prognostic the relative decrease in physical fitness was evaluated.Data was adjusted for baseline fitness and "penalized" regression models which contains more factors, even when the "Rule of 10" is not, met were used 30 .Logistic regression was used to investigate the relationship between POMS morbidity and fitness.All analyses were performed with the statistical software Stata 14.0.

Patient characteristics
From September 2011 to September 2016, we enrolled 160 patients (one-year mortality follow-up to September 2017).One withdrew consent and was not included in the analysis (Figure 1).Patient characteristics for the whole group are shown in Table 1.Cancer regime, tumour characteristics and radiological responses to chemo/chemoradiotherapy using RECIST v1.1 criteria are presented in Table 2. Patients who completed a pre-chemo/chemoradiotherapy CPET only (i.e.patients who did not progress to surgery due to a serious adverse event, a palliative diagnosis on restaging, death during chemo/ chemoradiotherapy or progressing to surgery after a serious adverse event or no CPET; n=23) had lower rates of chemo/chemoradiotherapy completion (7/23 (30.4%) vs. 86/109 (78.9%)), cycles undertaken (1.9 vs. 2.9) and were found to have distant disease on restaging (6/23 (50%) vs. 6/109 (5.5%)).No adverse events during CPET were recorded.
Considering clinical risk groups, pre-chemo/chemoradiotherapy VO 2 at AT risk group (low/medium risk vs. high/highest  Abbreviations: NAC (neoadjuvant chemotherapy), CRT (neoadjuvant chemoradiotherapy), VO 2 at AT (oxygen uptake at anaerobic threshold), VO 2 at Peak (oxygen uptake at peak exercise), WR at AT (work rate at anaerobic threshold), WR at Peak (work rate at peak exercise), V E /VCO 2 at AT (ventilatory equivalent for carbon dioxide at the anaerobic threshold), V E /VCO 2 at Peak (ventilatory equivalent for carbon dioxide at peak exercise), V E /VO 2 at AT (ventilatory equivalent for oxygen at the anaerobic threshold), V E /VO 2 at Peak (ventilatory equivalent for oxygen at peak exercise), PETCO 2 at AT (end tidal carbon dioxide at the anaerobic threshold), PETCO 2 at Peak (end tidal carbon dioxide at peak exercise), VO 2 /HR at AT (oxygen pulse at the anaerobic threshold), VO 2 /HR at Peak (oxygen pulse at peak exercise risk) was associated with mortality at one-year (HR=7.06;95%CI: 2.04 to 24.40; p=0.002) whilst there was insufficient evidence for mortality at five-year follow (HR = 1.39 (0.80 to 2.41; p=0.237)) (Figure 6).Post-chemo/chemoradiotherapy showed weaker evidence for an association at one year (HR=3.29;95%CI: 0.95 to 11.36; p=0.06) and there was insufficient evidence for five-year mortality (HR = 0.79 (0.46 to 1.39; p = 0.426) (Figure 7).A weak relationship between change in fitness risk group and survival at one year was noted (Better HR=0.78, 95%CI: 0.17 to 3.48; Worse HR=0.53, 95%CI: 0.17 to 1.64;      p=0.550) whilst a stronger relationship was noted at 5 years (Better HR = 1.09, 95%CI: 0.48 to 2.48; Worse HR = 0.52, 95%CI: 0.27 to 1.01; p=0.097).
Pre-chemo/chemoradiotherapy VO 2 at Peak was also independently associated with 1-year mortality (HR=0.85,95%CI: 0.76 to 0.95, p=0.005).There was no evidence of an association between VO 2 at Peak post-chemo/chemoradiotherapy or relative change in VO 2 at Peak from pre-to post-chemo/ chemoradiotherapy.Following adjustment for post-chemo/ chemoradiotherapy VO 2 at Peak, the relative change in VO 2 at Peak following chemo/chemoradiotherapy was associated with 1-year mortality (HR=0.85,95%CI: 0.74 to 0.98, p=0.023).

Fitness and post-operative outcomes
Of the 109 patients who complete chemo/chemoradiotherapy, 108 underwent surgery (100 patients had complete pre-and post-chemo/chemoradiotherapy CPET data as well as surgical outcomes).Surgery and histopathological outcomes are summarised in Table 4.A total 6% of patients underwent a palliative bypass, 57% of patients underwent a laparoscopic operation with only one patient having a conversion to open surgery.An R0 resection was achieved in 75% of patients.A complete pathological response (TRG1) was achieved in 14%.Table 5 summarizes the post-operative outcomes as defined by POMS at Day3, 5, 8 and 15.Of note, on Day 8, 24% of patients had a complication in the low-risk group, whilst 46% of patients still had a complication in the highest risk.Median length of stay was 12 days for the low-risk group and 13.5 days in the highest risk group.Seven patients in the low/medium risk groups died postoperatively at 1-year (11.6%) whilst 18 patients died in the high/highest risk groups (36.7%).Thirty-one patients in the low/medium risk groups died post-operatively at five-years (51.6 %) whilst 28 patients died in the high/highest risk groups (57.1 %).There was insufficient evidence of an association found between POMS at Day 3, 5, 8 and 15 and pre-chemo/ chemoradiotherapy VO 2 at AT (OR=1.0495%CI: 0.85 to 1.27; p=0.730).Further, insufficient evidence of relationships was found with either post-chemo/chemoradiotherapy fitness or change in fitness.Similar results were seen for VO 2 at Peak.Of note 46% of highest risk patients were discharged back to their own home post-operatively (compared with 95% in the lowrisk group), and 46% of patients in the high-risk group were re-admitted within 30-days (compared with 5% in the low-risk group).

Fitness and quality of life
We found no evidence of an association between changes in EQ-5D-5L and the relative change in VO 2 at AT or VO 2 peak following chemo/chemoradiotherapy at 30-days and one-year post-operatively (Spearman correlation 30-days r=0.062, p=0.611 and r=0.10, p=0.393; 1-year r=0.162, p=0.191 and r=0.141, p=0.253 respectively).
Chemo/chemoradiotherapy adverse effects and tolerability A patient's ability to complete the planned full cycles of chemo/ chemoradiotherapy (tolerability) as a function of their baseline fitness was assessed.Adverse events encountered during chemotherapy and chemoradiotherapy are presented in Table 6.
No association between VO 2 at AT or VO 2 peak and tolerability was found (p=0.161 and p=0.057).One-year survival was no better for patients who tolerated chemo/chemoradiotherapy (HR = 0.42, 95%CI: 0.17 to 1.06; p=0.067).Additionally, there was no evidence that either morbidity at day 5 or EQ-5D at 30-days post-operatively was associated with tolerance (p=0.149 and p=0.132).

Discussion
This prospective, observer blinded, multi-centre, observational, cohort study in patients undergoing multimodal neoadjuvant cancer therapies for locally advanced OG cancers showed a significant decline in objectively measured fitness (VO 2 at AT and VO 2 peak) following chemo/chemoradiotherapy.Importantly, baseline fitness (VO 2 at AT and VO 2 peak) was strongly associated with mortality at one-year.Finally, we observed a higher incidence of post-operative morbidity and a longer length of hospital stay in the high/highest risk groups, as well as a greater decline in VO 2 at AT following chemoradiotherapy compared to chemotherapy treatment.
Multimodal neoadjuvant therapies for OG cancer patients are associated with marginal gains in overall survival 6,31 .However, these treatments might be accompanied with a significant cost to patient's fitness.Although overall survival has improved for oesophageal patients, a network meta-analyses showed that chemoradiotherapy increased the risk of post-operative mortality when compared to chemotherapy or surgery alone, but improved tumour regression was found 32 .Similarly, overall survival was moderately improved in gastric cancer patients undergoing platinum based triplet regimens 6,33 , however such patients experience significant peri-chemotherapy morbidity with increased lymphocytopenia and hemoglobinopathy.In some patients, pre-operative treatment might have no meaningful benefit and may even cause harm 34 .
The reliability and association of selected CPET variables with post-operative outcome has been established for major abdominal surgery 9 , however this relationship has not been adequately investigated in OG patients, where surgery is frequently in two body cavities.At present, there is little evidence supporting the use of pre-operative CPET to aid shared decision making and guide perioperative care prior to OG cancer surgery.Conclusive evidence regarding changes in fitness after neoadjuvant treatment for OG cancer and any possible relationship with operative outcomes does not exist.So far small, unblinded, single-centre, observational studies report similar significant declines in fitness with chemo/ chemoradiotherapy, associations between low fitness and higher post-operative cardio-respiratory complications and sustained reductions in fitness between the end of neoadjuvant chemotherapy (NAC) and surgery [10][11][12][13] .
This study reports clinically important reductions in fitness associated with chemo/chemoradiotherapy, validating our initial pilot and single-centre observations [10][11][12][13] .This may be attributed to changes in metabolic health such as changes in body composition (sarco-cachexia) due to neoadjuvant treatments, recently recognised in OG cancer cohorts [35][36][37] .Sarcopenia and sarcopenic obesity are associated with early termination of neoadjuvant treatments, dose limiting toxicity, operative morbidity, poor oncological outcomes, including poor survival 38 .Sarcopenia, like poor fitness is a modifiable risk factor.Multimodal prehabilitation interventions might rescue the decline in fitness and body composition seen with neoadjuvant treatments [39][40][41] , however, improving baseline fitness (associated with one-year mortality) is challenging and a broader public health issue.Whilst mechanisms of reduced fitness are unclear, a reduction in muscle mass is also known to reduce ventilatory efficiency.This study demonstrates a reduction in ventilatory equivalent for carbon dioxide (V E /VCO 2 ) at AT and peak.Moreover, mitochondrial dysfunction and sarcopenia are attributed to toxicity from neoadjuvant platinum-based compounds.Mitochondrial DNA, cell cycle arrest 42 , sustained activation of degradative proteasome and autophagy systems 43 , and altered NF-kB signalling 44 are linked to platinum based chemotherapy Neutropenia is defined as absolute neutrophil count <1.0 ×10 9 /L; Creatinine >1.5 upper limit for normal Of 132 patients receiving NAC/CRT; 86% underwent chemotherapy alone (38 patients (34%) incomplete treatment) and 14% underwent chemoradiotherapy (1 patient (5%) incomplete treatment).Acute kidney injury with a rise in serum creatinine more than 1.5 times the upper limit of normal was the most common adverse event encountered.Five deaths were recorded during NAC/CRT.
toxicity.Ultimately, these changes induce mitochondrial and cellular protein damage, leading to autologous destruction -mytophagy and muscle wasting 45 seen in the present study as a reduction in oxygen utilisation and power output at AT and peak exercise.
Objective fitness assessment prior to chemo/chemoradiotherapy provides useful information to guide personalized shared decision-making around operative risk and survival for the multi-disciplinary team and the patient.It may also guide neoadjuvant chemo/chemoradiotherapy choices, by directing more aggressive chemo/chemoradiotherapy choices to fitter individuals or even guide delays in the cancer treatment pathway to institute an intervention before neoadjuvant treatment in selected individuals.Informing the selection of risk-reducing neoadjuvant treatment options when baseline fitness is already compromised (e.g.selection of NAC over chemoradiotherapy for low baseline fitness patients), or even guide shared-decision making around palliative options might be clinically desirable.On the contrary, giving higher doses of chemotherapy or chemoradiation, or even more aggressive FLOT-based chemotherapy 32 in patients who are objectively in low-risk fitness categories might improve outcomes for selected patients.Objective risk-stratification might also direct higher levels of peri/postoperative care in unfit individuals, thereby improve utilisation of scarce critical care resources.The more information we are able to obtain on the patient's physiology, metabolic health and tumour status, the more we are able to effectively inform patients and their relatives in their decision-making process and consent.These data might also suggest the potential utility of tailored prehabilitation interventions during neoadjuvant treatments to optimise metabolic health and respiratory function, however this needs further evaluation as maintaining/ improving fitness during the whole cancer pathway might impart some long-term benefits 40,46,47 .
There are several strengths to this study.Firstly, it validates the findings of other smaller cohort studies utilising rigorous, prospective, observer-blinded methodology and providing multi-centre generalisability.It describes a novel, strongly independent association between baseline physical fitness (pre-chemo/chemoradiotherapy) and one-year overall survival, but unfortunately not between post-treatment fitness and survival.These observations emphasise the need for physiological staging, including evaluation of physical fitness, to be undertaken prior to chemo/chemoradiotherapy, and that fitness assessment after chemo/chemoradiotherapy might not play a role in predicting long-term outcomes.Further work is needed to establish this.Methodological rigor around dual-expert, blind CPET reporting and a low risk of confounding by indication is also a strength.
Limitations of this study included, a lack of accurate operative definitions, the heterogeneity of the cancer type and cancer treatments.This, however, is a pragmatic study and restrictions based around operative technique and care treatment were felt to be overly restrictive and not generalisable.Further, inability to successfully interrogate the association between fitness, morbidity and patient reported quality of life was due to a lack of statistical power.Furthermore, neither changes in physical fitness nor fitness-risk group following chemo/ chemoradiotherapy were independently associated with one-year survival due to a lack of statistical power.
In conclusion, neoadjuvant treatment prior to OG cancer surgery significantly reduces physical fitness, with patients who are unfit at baseline having lower survival at one-year post-operatively.Reviewer Expertise: Oesophagogastric cancer outcomes with regard to preoperative fitness, rehabilitation and muscle mass.
We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Figure 1 .
Figure 1.Study enrolment and follow-up.Of the 160 patients who were enrolled, 159 had pre-neoadjuvant chemo/chemoradiotherapy (NAC/CRT) cardiopulmonary exercise testing (CPET), 27 did not undergo NAC/CRT of which 5 went straight to surgery and were excluded.132 patients underwent NAC/CRT of which 8* went straight to surgery and a further 15 did not complete post-NAC/CRT CPET.109 patients underwent post-NAC/CRT CPET and 108 underwent surgery.* denotes patients who went straight to surgery after either a serious adverse event during NAC/CRT or patients who did not undergo CPET after NAC/CRT.

Figure 3 .
Figure 3. Kaplan-Meier plot of the estimated overall 5-year survival among patients with oesophago-gastric cancer who underwent CPET pre-and post-NAC/CRT followed by surgery, based on their change in VO 2 at AT risk stratification category following chemo/chemoradiotherapy (NAC/CRT).

Table 2 . Cancer regime, tumour characteristics and radiological responses of patients who underwent neoadjuvant chemotherapy (NAC) or chemoradiotherapy (CRT).
Data are n (%) or mean (SD) unless otherwise stated.an=121, b n=129.$Radiological responses post-NAC/CRT were based on definitions outlined by the RECIST version 1.1 criteria for solid target lesions.P-value for comparison between pre-and post NAC group vs. pre-NAC group.Statistical significance was taken <0.05 and highlighted in bold.Note: Complete cancer therapy is defined as successfully completion of all planned cycles of NAC/CRT.Incomplete cancer therapy is defined as unsuccessful completion of all planned NAC/CRT cycles for any reason.* T-stage, N-stage, M-stage both pre and post-NAC/CRT was assessed by means of computer tomography (CT) and/or endoscopic ultrasound and/or 18 F-fluorodeoxyglucose positron-emission tomography, classified according to TNM v7.

Table 3 . Cardiopulmonary exercise testing (CPET) variables before and after neoadjuvant chemotherapy (NAC) and chemoradiotherapy (CRT).
Data are mean (SD) unless otherwise stated.Statistical significance was taken as <0.05 and highlighted in bold.

Table 4 . Surgery and Histopathological Outcomes.
n=101, b n=100, c n=102.$ Tumor Regression Grade (TRG) was based on definitions outlined by the Mandard score.EMVI -Extra mural venous invasion.RO -complete resection with no tumor within 1mm of the resection margin.

Table 6 . Adverse events during chemotherapy and chemoradiotherapy -events of grade ≥2 during chemotherapy or chemoradiotherapy.
Adverse events were graded according to the National Cancer Institute's Common Terminology Criteria for Adverse Events, version 3.0.

the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes
61rvard Dataverse: Replication Data for Op Details: The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/LOB8PG56.Harvard Dataverse: Replication Data for Histology: The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/DSC-SOI57.[Harv ard] Dataverse: Replication Data for EQ-5D-5L 30 days post surgery: The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/ZDR7C158.Harvard Dataverse: Replication Data for EQ-5D-5L 1 year post surgery: The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/AGC3IL59.Harvard Dataverse: Replication Data for CPET Variables: The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/IM11CG60.guidelinesHarvardDataverse: STROBE checklist for 'The effects of cancer therapies on physical fitness before oesophagogastric cancer surgery: A prospective, blinded, multi-centre, observational, cohort study'.https://doi.org/10.7910/DVN/NUV1OI61.Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).
Competing Interests: No competing interests were disclosed.

Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.