Elsevier

Clinical Oncology

Volume 29, Issue 9, September 2017, Pages 576-584
Clinical Oncology

Original Article
Prediction of 90 Day and Overall Survival after Chemoradiotherapy for Lung Cancer: Role of Performance Status and Body Composition

https://doi.org/10.1016/j.clon.2017.06.005Get rights and content

Highlights

  • Low body mass index and poor performance status predict reduced 90 day survival.

  • Low muscle attenuation (myosteatosis) is prognostic for overall survival.

  • Few patients are underweight despite evidence of active cachexia.

  • Personalised clinical decision making should incorporate known prognostic factors.

Abstract

Aims

If appropriate patients are to be selected for lung cancer treatment, an understanding of who is most at risk of adverse outcomes after treatment is needed. The aim of the present study was to identify predictive factors for 30 and 90 day mortality after chemoradiotherapy (CRT), and factors that were prognostic for overall survival.

Materials and methods

A retrospective cohort study of 194 patients with lung cancer who had undergone CRT in South East Scotland from 2008 to 2010 was undertaken. Gender, age, cancer characteristics, weight loss, body mass index (BMI), performance status (Eastern Cooperative Oncology Group; ECOG) and computed tomography-derived body composition variables were examined for prognostic significance using Cox's proportional hazards model and logistic regression.

Results

The median overall survival was 19 months (95% confidence interval 16.3, 21.7). Four of 194 patients died within 30 days of treatment completion, for which there were no independent predictive variables; 22/194 (11%) died within 90 days of treatment completion. BMI < 20 and ECOG performance status ≥2 were independent predictors of death within 90 days of treatment completion (P = 0.001 and P = 0.004, respectively). Patients with either BMI < 20 or ECOG performance status ≥ 2 had an odds ratio of death within 90 days of 5.97 (95% confidence interval 2.20, 16.19), rising to an odds ratio of 13.27 (1.70, 103.47) for patients with both BMI < 20 and ECOG performance status ≥ 2. Patients with low muscle attenuation had significantly reduced overall survival (P = 0.004); individuals with low muscle attenuation had a median survival of 15.2 months (95% confidence interval 12.7, 17.7) compared with 23.0 months (95% confidence interval 18.3, 27.8) for those with high muscle attenuation, equating to a hazard ratio of death of 1.62 (95% confidence interval 1.17, 2.23, P = 0.003).

Conclusion

Poor performance status, low BMI and low muscle attenuation identify patients at increased risk of premature death after CRT. Risk factors for adverse outcomes should inform personalised discussions with patients about the potential harms as well as the intended benefits of treatment.

Introduction

Lung cancer is the most common cause of cancer death worldwide. In the UK, a 10 year survival rate of 4.9% remains disappointingly low [1]. Given that most patients with lung cancer are incurable at diagnosis, with survival typically measured in ‘months’ [2], targeting treatments at improved quality of life, extended life or both is the priority [3]. In general, more radical treatment options carry more significant side-effects and a greater patient burden. An informed understanding of the probable outcomes following radical treatments is therefore important for clinicians and patients alike, ahead of treatment plans being made.

Clinical decision making in lung cancer in the UK is informed by national guidelines [4], [5], which are underpinned by an extensive evidence base. However, the lung cancer population is typically elderly and multimorbid [2], [4], [5], [6], and such patients are under-represented in clinical trials. Indeed, a recent review of international lung cancer guideline concordance questioned the generalisability of clinical guidelines to ‘real world’ patients [7].

In addition to cancer treatment modality, a number of factors are known to influence prognosis in lung cancer, including disease stage [2], [6], [8], [9], age [2], [6], [8], gender [2], [6], performance status [2], [6], [9], [10], systemic inflammation [10], [11], [12], [13], comorbidity [2], [6], muscle wasting [11], [14], [15] and weight loss [8], [15], [16], [17], [18]. Cachexia, a complex syndrome encompassing many of these poor prognostic factors (namely weight loss, muscle wasting and systemic inflammation), is responsible for numerous negative patient outcomes in cancer [19]. However, it is not currently assessed or managed in any systematic or consistent way in the lung cancer clinic.

There is a lack of understanding about which predictive factors relate to early mortality (presumed to be a consequence of anti-cancer treatment) and which to reduced overall survival as a result of progressive cancer. The potential for severe adverse outcomes due to cancer treatment toxicity is well recognised [4], [5] and death within 30 days of systemic anti-cancer treatment is now a commonly used proxy for quality of care [20]. Death within the 90 day post-treatment period is of interest in surgical cancer populations [21] and in Scotland lung cancer mortality within 30 or 90 days of the receipt of anti-cancer treatment is a routinely collected quality performance indicator [22].

Where evidence does exist around prognostic and predictive factors, assimilating this into clinical decision making can be a challenge. Furthermore, there can be a lack of understanding by some patients that cancer treatment has the potential to do more harm than good. Evidence is needed to inform discussions between lung cancer clinicians and their real world patients about the probable benefits and the potential harms of treatment. It is only with this evidence that cancer treatments can be targeted most appropriately.

The aim of this retrospective cohort study was to identify predictive factors for 30 and 90 day mortality after chemoradiotherapy (CRT) for lung cancer, and factors that were prognostic for overall survival.

Section snippets

Materials and Methods

All patients diagnosed with lung cancer between 2008 and 2010 in South East Scotland given CRT were included in the retrospective cohort study. National Caldicott approval was granted for the collation of existing patient data.

Patient Characteristics at Diagnosis

Demographic and clinical characteristics at diagnosis for all 194 patients who underwent CRT for lung cancer diagnosed in South East Scotland between 2008 and 2010 are outlined in Table 1, Table 2. Most patients had either stage III NSCLC or limited stage SCLC, and were treated with concurrent CRT. Ninety per cent of patients had an ECOG performance status of 0/1. The mean BMI for both men and women was in the overweight (BMI > 25) category, with over half of women and nearly two-thirds of men

Discussion

Treatment with CRT for lung cancer is lengthy and often burdensome. The expectation for most patients is of significantly extended survival. Current criteria for starting treatment include disease stage and performance status. In the present study, 98% and 89% of patients, respectively, survived more than 30 and 90 days after the completion of CRT, thereby suggesting that, for most, these criteria are fit for purpose. For the 22/194 (11.3%) patients who died within 90 days, BMI < 20 and ECOG

Conclusions

The present study has identified three clinical variables (low performance status, low BMI and low muscle attenuation) associated with significantly reduced survival (early and late) after CRT for lung cancer. Further work is needed to validate these findings and to examine the relationship between comorbid illness, frailty and cachexia. Understanding more about actual causes of early death, including the relationship with treatment toxicity, should also be a priority. Patient selection for

Acknowledgements

This work is dedicated to the memory of Ken Fearon, Professor of Colorectal Surgery and world leader in cachexia research. His academic expertise and mentorship were so greatly valued and he is very sadly missed. The authors also wish to acknowledge the Melville Charitable Trust (funded J. Bowden research fellowship), Professor Gerry Humpris and Dr Damien Williams, University of St Andrews (MD supervisors of J. Bowden), SCAN lung cancer team.

References (46)

  • E.B. Haverkort et al.

    Self-reporting of height and weight: valid and reliable identification of malnutrition in preoperative patients

    Am J Surg

    (2012)
  • V.E. Baracos et al.

    Body composition in patients with non-small cell lung cancer: a contemporary view of cancer cachexia with the use of computed tomography image analysis

    Am J Clin Nutr

    (2010)
  • K. Huang et al.

    Follow-up of patients after stereotactic radiation for lung cancer: a primer for the nonradiation oncologist

    J Thorac Oncol

    (2015)
  • G. Buccheri et al.

    Importance of weight loss definition in the prognostic evaluation of non-small-cell lung cancer

    Lung Cancer

    (2001)
  • K.J. Sanders et al.

    Early weight loss during chemoradiotherapy has a detrimental impact on outcome in NSCLC

    J Thorac Oncol

    (2016)
  • A. Warner et al.

    Factors associated with early mortality in patients treated with concurrent chemoradiation therapy for locally advanced non-small cell lung cancer

    Int J Radiat Oncol Biol Phys

    (2016)
  • Cancer Research UK

    Cancer research statistics for the UK

    (2010–2011)
  • G. Silvestri et al.

    Preferences for chemotherapy in patients with advanced non-small cell lung cancer: descriptive study based on scripted interviews

    Br Med J

    (1998)
  • National Institute of Health and Clinical Excellence

    Lung cancer: diagnosis and management

    (2011)
  • Scottish Intercollegiate Guidelines Network

    Management of lung cancer. A clinical guideline

    (2014)
  • A.L. Rich et al.

    Inequalities in outcomes for non-small cell lung cancer: the influence of clinical characteristics and features of the local lung cancer service

    Thorax

    (2011)
  • S.-I. Go et al.

    Sarcopenia and inflammation are independent predictors of survival in male patients newly diagnosed with small cell lung cancer

    Supportive Care Cancer

    (2016)
  • L. Zhu et al.

    Glasgow prognostic score predicts prognosis of non-small cell lung cancer: a meta-analysis

    SpringerPlus

    (2016)
  • Cited by (0)

    Died September 2016.

    View full text