Kasymjanova, Goulnar MD*; Correa, José A. PhD†; Kreisman, Harvey MD*; Dajczman, Esther RN, MSc*; Pepe, Carmela MD*‡; Dobson, Sarah MSc*; Lajeunesse, Lucie RN*; Sharma, Rajesh MD*; Small, David MD*‡
Chemotherapy in advanced non-small cell lung cancer (NSCLC) is known to improve both quality of life (QoL) and symptom control when compared with best supportive care.1 Despite these improvements, chemotherapy offers only a modest survival benefit for patients with inoperable NSCLC.2 Consequently, QoL and supportive care are the priority throughout the treatment. Physical and functional well-being are considered essential dimensions of overall QoL. It is therefore of utmost importance to consider the patient functional status during and after chemotherapy. Most research on functional status in cancer patients is done using physician-reported Eastern Cooperative Oncology Group-performance status (ECOG-PS) or patient-reported QoL questionnaires. The correlation between PS and prognosis is well known.3 However, the ECOG-PS is very subjective and in 50% of cases does not agree with patients perception of functional status.4 Other more objective measures of functional status/QoL and prognostic tools are needed to complement the existing ones.
The 6-minute walk (6MW) has been suggested as simple, safe, and inexpensive measure of functional status.5 It has been used in pulmonary and cardiovascular literature, and shown to predict survival, morbidity, and mortality.6,7 Recently, it was used in preoperative thoracic surgery patients and in patients with lung cancer undergoing radiation therapy.7,8 Although lung cancer is among the most common malignancies in North America,9–11 the impact of chemotherapy on the exercise capacity and subsequent outcome of patients with advanced NSCLC is not well understood. We prospectively evaluated physical function in patients with advanced NSCLC at diagnosis and after two cycles of chemotherapy. These evaluations were made as part of a larger longitudinal observational study designed to identify characteristics associated with change in function of patients receiving chemotherapy for advanced lung cancer. We hypothesized that patients would experience a clinically significant functional decline, characterized as a decrease in 6MW distance, after two cycles of chemotherapy. Following the reported predictive/prognostic value of 6MW in other diseases, we also hypothesized that 6MW may predict response to treatment and survival.
PATIENTS AND METHODS
Consecutive patients with newly diagnosed advanced NSCLC were identified from the outpatient pulmonary oncology clinic of the Jewish General Hospital. All patients were eligible to receive at least two cycles of chemotherapy. The exact regimen was single or double agent, and was prescribed at the discretion of the treating physician. Participants underwent an initial 6MW, and 1 to 3 days later underwent the prechemotherapy 6MW. On day 21 or 28 of cycle 2, depending on the cycle length, another 6MW was performed. The protocol was approved by the Institutional Review Board at the Jewish General Hospital.
Patients with NSCLC stage IIIA, IIIB, or IV, ECOG-PS 0 to 2 and with a life expectancy of at least 4 months were included. Those with uncontrolled cardiac or musculoskeletal disease, steroids for central nervous system metastases or on erythropoietic growth factors were excluded. Subjects were not permitted to participate in an exercise rehabilitation/training program during or for 2 months before the study.
Following a standardized protocol, after instructions to cover as much distance as possible, participants walked up and down a 30-m hallway for the allotted 6 minutes.12 Patients were instructed to walk at their own pace and were advised to slow down or stop as needed and resume walking as soon as they felt they were able to do so. At the end of 6 minutes, the distance covered was measured by the instructor. Dyspnea was measured by the Borg scale, oxygen saturation and pulse rate were assessed at the start and end of 6MW. Practice test, prechemo, and postchemo tests were performed. The rationale for the initial (practice) test was to account for a learning effect. It is recommended to do the practice test one day apart from pretreatment test.5 In our cohort, the initial test was performed in 1.5 (0.9–3.1) days before the prechemo 6MW.
A difference in walking distance of 54 m (95% confidence interval [CI] 37–71 m) had been associated with a noticeable clinical difference in the performance of patients with chronic obstructive pulmonary disease (COPD) and was considered to be clinically significant in our study. Those patients with less than a 54-m decline were categorized as improved/unchanged, whereas those with ≥54 m decline were considered as having a clinically significant decline. The predicted norms for 6MW were calculated using Enright’s equation.13 Patients were categorized into one of two different levels (≥400 m and <400 m) according to their performance on the initial 6MW, using a threshold of 400 m, as recommended in lung cancer patients receiving radiation therapy.14
This 400 m cutoff is, in keeping with other 6MW thresholds of 300 to 400 m used, in various other populations such as heart failure,7,15 COPD,16–18 and pulmonary hypertension.19,20
C-reactive protein (CRP) and hemoglobin (Hgb) levels were determined using standard techniques.
To detect a clinically significant difference of 54 m (95% CI; 37–71) in the 6MW and to accommodate an estimated dropout rate of 25%, sample size was calculated to be 63 patients. Participants were classified as dropouts (died before any follow-up evaluations, too disabled to complete functional assessments, or refused to perform the 6MW) or completers. To determine the association with dropout, various patient characteristics and initial 6MW were evaluated in completers and dropouts. Contingency tables and the χ2 test were used for categorical variables and t tests were used for continuous variables.
Evidence that the posttherapy 6MW data are missing at random, a preliminary step for the application of multiple imputation (MI) methods,21 was examined with a logistic regression model using the binary variable (dropout/completer) as a dependent variable (results are not shown in this paper). The initial 6MW test, together with the covariates of interest: age, sex, stage, ECOG-PS, body mass index (BMI), CRP, Hgb level, coexisting COPD, and type of chemotherapy, were used as additional independent variables. Results showed that in our cohort the probability of a subject missing post-therapy 6MW depended on initial 6MW only. This indicates that data are missing at random and, therefore, MI methods may be employed.
To investigate risk factors for functional decline, multiple linear regression modeling was used with the difference between prechemotherapy and postchemotherapy 6MW as the dependent variable. The covariates included in the model were age, sex, stage, BMI, CRP level, Hgb level, and coexisting COPD. After the advice of Allison,22 who recommends that only available data be used when data for the primary outcome are missing but data on the independent variables of interest are complete, we performed the regression using only the data for the 45 completers. Nevertheless, as a test of robustness, we also applied MI methods to substitute the missing 6MW values within subjects using the SAS procedures: Proc MI, for imputation of the data with up to 100 complete data sets and Proc MIANALYZE, for combining the results.
In completers, the prechemotherapy 6MW was compared with postchemotherapy using a paired t test. To investigate the prevalence of clinically significant decline, patients were categorized into two groups: improved/unchanged (6MW may have increased but did not decrease >54 m) versus declined (>54 m decline in 6MW after chemotherapy). The baseline characteristics of these groups were compared using t tests and χ2 tests where appropriate.
Association between clinically significant changes in 6MW and freedom from progression was investigated by χ2 test. Freedom from progression was defined as time from diagnosis to documented disease progression and included patients with complete response, partial response or stable disease according to RECIST criteria.
Overall survival was defined as the time elapsed from the date of diagnosis to the date of death, or the date of the last follow-up for patients alive at the time of analysis. Overall survival curves were drawn by the Kaplan-Meier method. Multiple-covariate survival analysis was done using the Cox model and all known or potential prognostic factors. As it is an exploratory, hypothesis-building, study more than four variables were used in this model (age, sex, stage, CRP level, Hgb level, weight loss, ECOG-PS, and initial 6MW).23 The initial 6MW was chosen for Cox regression analysis because it allowed us to include data from all 64 patients. The likelihood ratio test was used to test the contribution of each variable to the model when added last, that is after adjustment for all of the other covariates.
All statistical analyses were carried out using the SAS software, version 9.1 (SAS Institute, Cary, NC) and the SPSS software, version 14.0 (SPSS, Chicago, IL). All statistical tests were two tailed. p values of 0.05 or less were considered to indicate significance.
The mean age of the 64 patients who consented was 62 years (SD 10.8). Fifty-four patients (84%) had a PS score of 0 or 1, and 10 (16%) had a PS of 2. Seventy-two percent were found to have adenocarcinomas and 56 patients (86%) were diagnosed with distant metastases (IIIB pleural effusion or IV). Sites of metastases included lung, bone, liver, and adrenal. Sixty-two patients received chemotherapy; 53 received double agent platinum-based chemotherapy, and 9 patients were given single-agent chemotherapy. Two patients withdrew consent and never received chemotherapy.
All 64 patients completed the initial 6MW, 59 (92%) completed the prechemotherapy 6MW and 45 (70%) completed all three tests. There was significant difference between the initial and prechemotherapy 6MW distances: 424 m (SD 94) versus 447 m (SD 101), respectively (p = 0.005). All patients received chemotherapy: 53 patients received double agent chemotherapy with 6 of them as part of dual modality therapy (4 concomitant and 2 sequential chemo-radiotherapy), 9 patients received single-agent chemotherapy. Of the 19 dropouts, 6 patients died before the end of the study, 9 were considered too disabled to complete the 6MW because of severe symptoms, 1 participant was missed because of protocol violation, and 3 others withdrew consent. The characteristics of dropouts and completers are shown in Table 1. Those who were dropped out were more likely to be female or have a higher CRP level before chemotherapy. The mean distance of the initial 6MW of drop outs (19 patients) was 361 m (SD 99) compared with 445 m (SD 85) for the 45 patients who completed the study (p = 0.001). The difference was also significant (p = 0.01) when distance was expressed as a percentage of predicted value (Table 1).
Of the 45 completers, 40 patients received double agent chemotherapy (all 6 patients receiving combination therapy had radiation only after completion of the initial two cycles of chemotherapy) and 5 patients received single-agent chemotherapy. On the prechemotherapy 6MW, all 45 completers walked more than 50% of the predicted distance with 71% (32 of 45) of them performing above 75% of predicted. The 6MW was repeated with a mean of 22 (SD 6) days after the start of the second cycle. The mean 6MW in completers decreased from 462 m (SD 100) before chemotherapy to 422 m (SD 111) after two cycles (p = 0.01), there was no correlation between the decline in 6MW and type of chemotherapy (p = 0.07). Of 45, 27 patients performed more poorly in 6MW after two cycles of chemotherapy, when compared with prechemo 6MW; 13/27 patients had a clinically significant (>54 m) decline in 6MW. The multiple linear regression showed that patients with less advanced stage and higher BMI were more likely to improve in 6MW after two cycles of chemotherapy (Table 2). However, sex, age, Hgb level, and CRP level or concomitant COPD did not affect the changes in 6MW. Results of MI methods were similar to those with completers (n = 45). The rate of disease progression was 7 of 32 (22%) in those who maintained/improved and 3 of 13 (23%) in those who have declined in 6MW, and was not statistically different between these groups, p = 0.48 (Figure 1). The overall median survival of all patients entered into this study was 11.1 months; in completers it was 14.4 months and in dropouts 4.1 months. Survival of those who had a clinically significant decline in 6MW was not statistically different from that of those whose 6MW improved/unchanged (15.1 months versus 12.8 months; p = 0.32).
The mean of the initial 6MW for the 64 patients was 420 m (SD 97) or 84% (SD 22) of predicted distance. Twenty-nine patients were below 400 m and 35 patients were above 400 m on the initial 6MW. The difference between patients with initial 6MW of <400 and ≥400 m is shown in Table 3. Those patients with 6MW <400 m were more likely to progress. Their median survival (6.7 months, 95% CI; 2.6–10.8) was also significantly shorter than that of patients with 6MW ≥400 m (13.9 months, 95% CI; 10.0–17.8) (p = 0.01). Kaplan-Meier survival curves for these two groups are shown in Figure 2. Univariate analysis of prognostic factors showed that younger age (p = 0.03) and earlier stage (p < 0.0001) were significantly associated with survival. A smaller BMI (p < 0.06) and better PS (p < 0.07) were of borderline significance. Sex, smoking history, weight loss, CRP level, Hgb level, and use of doublet chemotherapy were not associated with survival. After adjusting for known covariates of interest in a Cox regression model, patients with initial 6MW ≥400 m had a significantly greater survival time than those with initial 6MW <400 m; hazard ratio 0.44; 95% CI; 0.23–0.83 (p = 0.001) (Table 4).
This prospective study demonstrated that the ability to perform the 6MW declines substantially with time. Even a few days after the initial 6MW, patients dropped out and others were unable to complete the 6MW after two cycles of chemotherapy. This drop-out rate occurred despite a relatively normal distribution of initial 6MW compared with predicted values. There was a statistically significant decline in 6MW even in the select group of 45 patients who completed all three 6MW. In 29% of completers, the decline in 6MW was clinically significant.
Until recently, the strongest indication for performing the 6MW test was to measure the response to medical intervention in patients with moderate to severe COPD, and less commonly in congestive heart failure and pulmonary hypertension. Whereas in these patient populations, the 6MW is a direct outcome measurement for response to treatment, in lung cancer the change in 6MW may be confounded by treatment toxicity.
We observed a statistically significant decline in 6MW after two cycles of chemotherapy. The deterioration of a patients functional status was more likely to be seen in patients with decreased BMI and more advanced stage. However, 31 and 30% of our population performed above their upper normal limit on initial 6MW and prechemo 6MW, respectively. Although this is not frequent enough to be responsible for a ceiling effect, it might partially diminish the negative effect of chemotherapy on 6MW changes.24
The prognostic value of the 6MW has been reported in COPD, heart failure, and pulmonary hypertension patients.6,8,16–18 Rostagno et al.7 showed that a 6-MW distance of <300 m is a simple and useful prognostic marker for morbidity and mortality in patients with congestive heart failure. Cote et al.17 reported that in COPD patients, a single 6MW, using a threshold of 350 m, is as good a predictor of mortality as VO2 (peak oxygen consumption during exercise). In a series of lung cancer patients undergoing surgical resection, those who were able to walk >400 m were more likely to have a successful outcome.18 A 6-MW ≥400 m identified those lung cancer patients with less toxicity after radiation therapy using univariate or multiple regression models.14 However, the predictive value of 6MW for progression after chemotherapy treatment was never reported in lung cancer patients. Our results suggest that baseline 6MW might be an independent predictor of response.
In particular, progression rate was significantly lower in patients who covered ≥400 m on initial 6MW. We believe that this identifies patients who are more physically fit and more likely to benefit from chemotherapy. The mechanism of the superior response in patients with better performance in 6MW remains yet unclear and further study is required. In addition, the initial 6MW ≥400 m also identified a group with lower rates of dropout, a better survival than those patients with an initial 6MW <400 m.
Multivariate analysis for response and survival has identified several important predictive and prognostic factors in advanced NSCLC.3 These include stage, ECOG-PS, sex, age, and use of chemotherapy doublets.3,25,26 Second line therapy may also be of importance.27 In our study, age and stage were significantly associated with survival, whereas BMI and PS approached significance in univariate analysis. Patients receiving platinum-based doublets did not have a superior survival but this may have been related to the small sample size and other prognostic factors.
Identification of patients with advanced NSCLC who should be treated aggressively is of primary concern to the practitioner. Of the prognostic factors influencing this decision, PS and stage are the most significant ones, and physicians routinely rely on them. Despite this well-documented association, PS is a subjective evaluation made by the treating physician.4,28,29 The added prognostic value of initial walking distance may further determine patient suitability for treatment and was of prognostic value in our population of patients with inoperable NSCLC.
The 6MW is easily administered in the outpatient clinical setting. The only required tools are a stopwatch and a quiet corridor of predetermined length, usually >30 m. The supervising technician instructs the subject to walk as far as possible during the 6 minutes (during which the subject is permitted to rest) and encourages the subject in a standardized fashion. Although more precise data on exercise capacity might have been obtained using treadmill-based exercise testing, our pilot study showed that this method of testing was not feasible in our patients.
In conclusion, the 6MW declines significantly after two cycles of chemotherapy. This decline may have been even greater. Similar to previous studies in COPD, pulmonary hypertension, heart failure, and radiotherapy for lung cancer, in this study an initial 6MW <400 m identified patients with advanced NSCLC at high risk. Overall, the initial 6MW holds promise because of its simplicity, ease of performance, and its possible prognostic and predictive value. This will need to be confirmed in a larger prospective study.
Supported by The Mona Zavalkoff Fund for Pulmonary Oncology and an unrestricted grant from Sanofi-Aventis.
2. Spiro SG, Rudd RM, Souhami RL, et al. Chemotherapy versus supportive care in advanced non-small cell lung cancer: improved survival without detriment to quality of life. Thorax
3. Sculier JP, Chansky MSK, Crowley JJ, et al. The Impact of additional prognostic factors on survival and their relationship with the anatomical extent of disease expressed by the 6th edition of the TNM classification of malignant tumors and the proposals for the 7th edition. J Thorac Oncol
4. Dajczman E, Kasymjanova G, Swinton N, et al. Should patient-rated performance status affect treatment decisions in advanced lung cancer? J Thorac Oncol
5. ATS Statement. Guidelines for the six-minute walk test. Official statement of the American Thoracic Society. Am J Respir Crit Care Med
6. Lucas C, Stevenson LW, Johnson W, et al. The 6-min walk and peak oxygen consumption in advanced heart failure: aerobic capacity and survival. Am Heart J
7. Rostagno C, Olivo G, Comeglio M, et al. Prognostic value of 6-minute walk corridor test in patients with mild to moderate heart failure: comparison with other methods of functional evaluation. Eur J Heart Fail
8. Pinto-Plata VM, Cote C, Cabral H, et al. The 6-min walk distance: change over time and value as a predictor of survival in severe COPD. Eur Respir J
9. ASCO. Cancer Facts and Figures. Atlanta: American Cancer Society, 2007, http://www.cancer.org
. Accessed May 2007.
10. Canadian Cancer Society. Canadian Cancer Statistics 2007, http://www.cancer.ca
. Accessed May 2007.
11. Given B, Given C, Azzouz F, Stommel M. Physical functioning of elderly cancer patients prior to diagnosis and following initial treatment. Nurs Res
12. Nici L, Donner G, Wouters E, et al. American Thoracic Society/European Respiratory Society statement on pulmonary rehabilitation. Am J Respir Crit Care Med
13. Enright PL, Sherill DL. Reference equations for the six-minute walk in healthy adults. Am J Respir Crit Care Med
14. Miller KL, Kocak Z, Kahn D, et al. Preliminary report of the 6-minute walk test as a predictor of radiation-induced pulmonary toxicity. Int J Rad Oncol Biol Phys
15. Passantino A, Lagioia R, Mastropasqua F, et al. Short-term change in distance walked in 6 min is an indicator of outcome in patients with chronic heart failure in clinical practice. J Am Coll Cardiol
16. Carter R, Holiday DB, Nwasuruba C, et al. 6-Minute walk work for assessment of functional capacity in patients with COPD. Chest
17. Cote CG, Pinto-Plata V, Krasprzyk K, et al. The 6-min walk distance, peak oxygen uptake, and mortality in COPD. Chest
18. Holden DA, Rice TW, Stelmach K, et al. Exercise testing, 6-min walk, and stair climb in the evaluation of patients at high risk for pulmonary resection. Chest
19. Barst RJ, McGoon M, Torbicki A, et al. Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol
20. Miyamoto S, Nagaya N, Satoh T, et al. Clinical correlates and prognostic significance of six-minute walk test in patients with primary pulmonary hypertension. Comparison with cardiopulmonary exercise testing. Am J Respir Crit Care Med
21. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York: Wiley, 1987.
22. Allison PD. Missing Data. New York: Wiley/Sage, 2002.
23. Harrell FE. Regression Modeling Strategies with Applications to Linear Models, Logistic Regression, and Survival Analysis. New York: Springer, 2001.
24. Barber-Westin SD, Noyes FR, McCloskey JW. Rigorous statistical reliability, validity, and responsiveness testing of the Cincinnati knee rating system in 350 subjects with uninjured, injured, or anterior cruciate ligament-reconstructed knees. Am J Sport Med
25. Bernstein ED, Herbert SM, Hanna NH. Chemotherapy and radiotherapy in the treatment of resectable non-small-cell lung cancer. Ann Surg Oncol
26. Delbaldo C, Michiels S, Rolland E, et al. Second or third additional chemotherapy drug for non-small cell lung cancer in patients with advanced disease. Cochrane Database Syst Rev
27. Fossella FV, Lynch T, Shepherd FA. Second line chemotherapy for NSCLC: establishing a gold standard. Lung Cancer
28. Ando M, Ando Y, Hasegawa Y, et al. Prognostic value of performance status assessed by patients themselves, nurses, and oncologists in advanced non-small cell lung cancer. Br J Cancer
29. Blagden SP, Charman SC, Sharples LD, et al. Performance status score: do patients and their oncologists agree? Br J Cancer