|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 15 (August 1), 2004: pp. 3099-3103 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.08.040 Differential Prognostic Impact of ComorbidityFrom the Division of Hematology/Oncology, University of California San Diego School of Medicine, La Jolla, CA; Department of Otolaryngology-Head and Neck Surgery and Clinical Outcomes Research Office, and Department of Medicine, Division of Medical Oncology, Washington University School of Medicine; and Department of Mathematics, Washington University, St Louis, MO Address reprint requests to Jay F. Piccirillo, MD, FACS, Department of Otolaryngology-Head and Neck Surgery, Washington University School of Medicine, 660 S Euclid Ave, Campus Box 8115, St Louis, MO 63110; e-mail: piccirij{at}msnotes.wustl.edu
PURPOSE: Cancer patients with concurrent comorbid conditions have worse outcomes than patients with no comorbidities. We hypothesized that the prognostic impact of comorbidities would be greatest for patients with cancers associated with a long natural history and least in patients with aggressive cancers. PATIENTS AND METHODS: Using the Barnes-Jewish Hospital Oncology Data Services cancer registry, we grouped 11,558 patients with breast, lung, colon, or prostate cancer by morphologic stage at diagnosis and then determined the 1-year overall survival rate for each group. Overall, severity of comorbidity was assessed from chart review and classified into one of four groups: none, mild, moderate, or severe. The relative prognostic impact of comorbidity was measured by the hazard ratio and adjusted for the prognostic impact of age, race, and sex. RESULTS: One-year overall survival rate ranged from 20% for 1,005 patients with distant spread of lung cancer to 98% for 3,325 patients with localized prostate cancer. Adjusted hazard ratio of moderate/severe comorbidity (relative to none/mild) ranged from 1.04 to 4.48. The correlation between overall survival rate and severity of comorbidity was statistically significant (r2 = 0.56; P < .001). The proportion of variance in outcome explained by comorbidity ranged from less than 1% to almost 9%, depending on tumor site and stage. CONCLUSION: Concurrent comorbidities had the greatest prognostic impact among groups with the highest survival rate and the least impact in groups with the lowest survival rate. These findings can be used to help determine the role comorbidity information should play in studies of cancer outcomes.
Cancer patients often have other diseases or medical conditions in addition to their cancer. These conditions are referred to as comorbidities. Multiple studies have shown that cancer patients with comorbid conditions have worse outcomes than patients without comorbid ailments.1 This relationship between severity of comorbidities and outcome has been found in several different cancer sites, including head and neck cancer,2-4 breast,5,6 prostate,7,8 and colon cancer,9 among others. The presence and severity of comorbidities is distinct from the overall performance status, another well-established prognostic variable.10,11 The above-cited studies establish that, in patients with these cancers, increasing severity of comorbidity is associated with increasing risk of death. However, the magnitude of the prognostic impact may vary among populations with different cancers. For example, in a recent analysis of 451 patients who underwent resection of tumor-node-metastasis system (TNM) stage I non–small-cell lung cancers (NSCLC), the presence of comorbid conditions adversely impacted overall survival.12 In contrast, the presence of comorbidities had no impact on the overall survival of 179 patients with TNM stage III NSCLC treated with chemoradiotherapy or 226 TNM stage IV patients treated with chemotherapy.13 This finding, along with clinical experience, led us to suspect that the prognostic impact of concurrent comorbidities is greater in cancers for which the associated overall mortality is lower. Thus we predicted that in cancers that are more indolent (such as prostate cancer), comorbidities would have greater impact on overall survival, whereas in aggressive, rapidly fatal cancers (such as TNM stage IV NSCLC), comorbidities would have less impact. The goal of this study was to assess the differential prognostic impact of comorbidity in four different cancers. To answer this question, we grouped cancer cases based on site and morphologic stage and then assessed the impact of overall comorbidity severity on survival.
Study Design This was a prospective cohort study.
Study Population
Comorbidity
Statistical Analysis The Washington University Human Studies Committee approved this research.
The description of the study population is shown in Table 1. The number of patients per cancer site ranged from 1,878 with colorectal cancer to 4,230 with prostate cancer. The majority of patients with lung and colorectal cancer were 65 years or age and older, whereas the majority of patients with prostate and breast cancer were younger than 65 years. The overwhelming majority of patients were white, although there were a significant number of black patients with cancers of the lung, breast, and colorectum. The highest levels of comorbidity (moderate and severe) were found among the lung cancer patients (38.9%), whereas the least amount was among prostate (13.3%) and breast (18.3%) cancer patients. For prostate cancer, the large number of patients who were younger than 65 years of age and low levels of comorbidity likely reflects Barnes-Jewish Hospital's participation in clinical trials of screening for prostate cancer.
The 1-year overall survival rate for each cancer site and Surveillance, Epidemiology, and End Results (SEER) stage, adjusted hazard ratio for each SEER stage (as compared with localized), and the adjusted hazard ratio attributable to moderate/severe comorbidity (as compared with none/mild) is shown in Table 2. For each cancer site, there was an inverse relationship between the adjusted hazard ratio for SEER stage and adjusted hazard ratio for moderate/severe comorbidity. For example, as the adjusted hazard ratio for patients with lung cancer across SEER stage went from 1.0, 1.48, 1.89, 2.88, and 5.88, the hazard ratio for moderate/severe comorbidity went from 1.78, 1.94, 1.21, 1.29, and 1.06. The hazard ratio for patients with prostate cancer across SEER stage went from 1.0, 1.2, and 9.13, whereas the hazard ratio for moderate/severe comorbidity went from 4.48, 3.48, and 1.86. This relationship between the prognostic impact of comorbidity and 1-year overall survival rate for the different combinations of cancer site and SEER stage is shown in Figure 1. The log hazard ratio associated with moderate/severe comorbidity is graphed against the 1-year overall survival rate for each cancer site and morphologic stage. At the one end of the overall survival range is NSCLC with distant metastases, where the 1-year overall survival rate is 20.5% and the hazard ratio of moderate/severe comorbidity is 1.06. At the other end of the survival range is localized prostate cancer, where the 1-year overall survival rate is 98% and the hazard ratio of moderate/severe comorbidity is 4.48. As can be seen in Figure 1, there is a strong association between 1-year overall survival rate and the log adjusted hazard ratio of comorbidity (r2 value of 0.56; P = .001). The proportion of explained variation by comorbidity within each cancer site/stage group indicates a similar relationship of increasing importance of comorbidity with decreasing cancer lethality.
In this study, we found that the prognostic importance of overall comorbidity for patients with breast, lung, colon, or prostate cancer is relative to the mortality burden of the index cancer. Although this is an intuitive concept, the data we present provide quantitative evidence and support. Gradual processes, like chronic comorbid conditions, tend to take time to manifest their effects, and these effects might not be seen if a more rapid process, like an aggressive cancer, supervenes. Therefore, comorbidity is prognostically most important in situations where the prognostic impact of the tumor is small. Conversely, in the situation where the tumor is advanced or aggressive and the prognosis is poor, comorbidity information is less important. Although several previous studies have shown comorbidities to have prognostic significance for the survival of patients, we know of no other studies that provide a rule determining the relative impact of comorbidities across different groups of cancer patients. Our findings can be used to predict the prognostic importance of comorbidity and to assess the potential value of including comorbidity in clinical studies in which overall survival is relevant. For example, comorbidity assessment should be included in prognostic research for conditions in which survival is relatively good, such as prostate cancer and localized breast cancer. Because cancer-specific and overall mortality is low in these populations, concurrent comorbidities can have a significant influence on survival. If these comorbid conditions are important factors determining overall survival, they may mask the overall survival impact of cancer therapy, which effectively improves cancer-specific survival. A logical extension of this concept is to the study of the impact of cancer screening on survival. Cancer-specific mortality is much lower among patients being screened for cancer (eg, most women who receive mammograms do not die of breast cancer). The importance of all-cause mortality in such studies was recently emphasized in an editorial by Black et al, 20 in which they make the point that the results of screening may prompt interventions that ultimately lead to the death of the patient, countering the disease-specific survival benefit of screening. In studies (such as screening or adjuvant treatment) in which disease-associated mortality is relatively low, a disproportionate number of patients with concurrent comorbidities in the treatment arm may be more susceptible to treatment-related mortality or mortality caused by their comorbid conditions. This could mask any survival impact of treatment. Prospective comorbidity assessment, along with knowledge of the impact of comorbidity in that disease state, might allow investigators to control for this. The finding that comorbidities have little impact on the survival of patients with aggressive tumors is in accord with previous studies that showed age, a generally recognized demographic prognostic factor, does not impact on the survival of patients with TNM stage IV NSCLC.21 An example of this is provided by Langer et al22 in their subanalysis of a phase III trial of combination chemotherapy for patients with metastatic NSCLC. They found that although older patients had significantly higher comorbidities, their response rate, quality of life, and survival did not differ from that of younger patients. On the other hand, our findings that comorbidity was not as significant a prognostic factor in more extensive and aggressive cancers is at odds with findings from Firat et al.11 They found that comorbidity was an important prognostic factor in stage III NSCLC and recommended that comorbidity should be included in protocols studying advanced-stage disease and used for stratification. Frasci et al23 also found that comorbidity, as defined by the Charlson score,24 was a significant prognostic factor. Severe comorbidity was significantly related to early termination of treatment for older patients with advanced NSCLC enrolled onto a clinical chemotherapy trial. Our findings suggest that patients with concurrent comorbidities could be included in clinical trials for many aggressive tumors without major concern about masking any survival benefit owing to the intervention. In these groups, the cancer, rather than other concurrent diseases, primarily determines survival. Therefore, improvements in disease-specific survival will translate into improvements in overall survival regardless of coexisting disease. Comorbidity information, even if not prognostically important on its own, can help investigators understand the relationships between other variables (ie, age, race, and survival). In addition, a priori assessment of comorbidities can help evaluate treatment-exacerbated deaths (those deaths caused by interactions between the intervention and preexisting medical conditions).25,26 Such interactions will become more important as the population ages and the burden of comorbidity increases. The four cancer types in our analysis (colon, breast, lung, and prostate) were chosen because they are the most commonly diagnosed cancers in the United States. We have begun to analyze groups of patients with less common cancers and have found that the rule relating the impact of comorbidity to mortality burden continues to hold true. For example, the presence of moderate to severe comorbidities is prognostically important among patients with localized head and neck cancer, a curable disease. In contrast, coexisting comorbidities had little impact on the survival of patients with pancreatic cancer, a notoriously aggressive disease. As the size of the Barnes-Jewish Hospital cancer registry database increases, it will be possible to determine the impact of comorbidities on other less common cancers. In addition, it may be possible to break down the overall comorbidity score into the component comorbid conditions, and then separately determine how each comorbid condition contributes to patient outcome. Detailed analyses of this sort are only possible if cancer registrars collect comorbidity information at the time of medical record review and chart abstraction.
The authors indicated no potential conflicts of interest.
Presented at the 38th Annual Meeting of the American Society of Clinical Oncology, Orlando, FL, May 18-21, 2002. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Piccirillo JF, Feinstein AR: Clinical symptoms and comorbidity: Significance for the prognostic classification of cancer. Cancer 77:834-842, 1996[CrossRef][Medline] 2. Piccirillo JF: Importance of comorbidity in head and neck cancer. Laryngoscope 110:593-602, 2000[CrossRef][Medline]
3. Ribeiro KC, Kowalski LP, Latorre MR: Impact of comorbidity, symptoms and patient's characteristics on the prognosis of oral carcinomas. Arch Otolaryngol Head Neck Surg 126:1079-1085, 2000 4. Hall SF, Rochon PA, Streiner DL, et al: Measuring comorbidity in patients with head and neck cancer. Laryngoscope 112:1988-1996, 2002[CrossRef][Medline]
5. Yancik R, Wesley MN, Ries LA, et al: Effect of age and comorbidity in postmenopausal breast cancer patients aged 55 years and older. JAMA 285:885-892, 2001 6. Satariano WA : Comorbidity and functional status in older women with breast cancer: Implications for screening, treatment, and prognosis. J Gerontol 47:24-31, 1992 7. Clemens JD, Feinstein AR, Holabird N, et al: A new clinical-anatomic staging system for evaluating prognosis and treatment of prostatic cancer. J Chron Dis 39:913-928, 1986[CrossRef][Medline] 8. Albertsen PC, Fryback DG, Storer BE, et al: The impact of co-morbidity on life expectancy among men with localized prostate cancer. J Urol 156:127-132, 1996[CrossRef][Medline] 9. Yancik R, Wesley MN, Ries LG, et al: Comorbidity and age as predictors of risk for early mortality of male and female colon carcinoma patients: A population-based study. Cancer 82:2123-2134, 1998[CrossRef][Medline]
10. Extermann M, Overcash J, Lyman GH, et al: Comorbidity and functional status are independent in older cancer patients. J Clin Oncol 16:1582-1587, 1998 11. Firat S, Byhardt RW, Gore E: Comorbidity and Karnofksy performance score are independent prognostic factors in stage III non-small-cell lung cancer: An institutional analysis of patients treated on four RTOG studies—Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 54:357-364, 2001
12. Battafarano RJ, Piccirillo JF, Meyers BF, et al: Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 123:280-287, 2002 13. Tierney R, Read W, Page N, et al: The differential prognostic impact of comorbidity in lung cancer. Proc Am Soc Clin Oncol 22:253a, 2002 (abstr 1008) 14. American College of Surgeons: Registry Operations and Data Standards (ROADS): Volume II. Commission on Cancer. Chicago, IL, American College of Surgeons, Standards of the Commission on Cancer, 1999 15. Johnston AS, Piccirillo JF, Creech CM, et al: Validation of a comorbidity education program. J Regist Manage 28:125-131, 2001 16. Piccirillo JF, Costas I, Claybour P, et al: The measurement of comorbidity by cancer registries. J Regist Manage 30:8-14, 2003 17. Young JJ, Roffers S, Gloeckler Ries L, et al: SEER Summary Staging Manual 2000: Codes and Coding Instructions. Bethesda, MD, NIH Publication No. 01-4969, 2000 18. Cox DR: Regression methods and life tables (with discussion). J R Stat Soc B 34:187-220, 1972 19. Schemper M, Henderson R: Predictive accuracy and explained variation in Cox regression. Biometrics 56:249-255, 2000[CrossRef][Medline]
20. Black WC, Haggstrom DA, Welch HG: All-cause mortality in randomized trials of cancer screening. J Natl Cancer Inst 94:167-173, 2002 21. Skarin AT, Herbst RS, Leong TL, et al: Lung cancer in patients under age 40. Lung Cancer 32:255-264, 2001[CrossRef][Medline]
22. Langer CJ, Manola J, Bernardo P, et al: Cisplatin-based therapy for elderly patients with advanced non-small-cell lung cancer: Implications of Eastern Cooperative Oncology Group 5592, a randomized trial. J Natl Cancer Inst 94:173-181, 2002
23. Frasci G, Lorusso V, Panza N, et al: Gemcitabine plus vinorelbine versus vinorelbine alone in elderly patients with advanced non-small-cell lung cancer. J Clin Oncol 18:2529-2536, 2000 24. Charlson ME, Pompei P, Ales KL, et al: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chron Dis 40:373-383, 1987[CrossRef][Medline]
25. Welch HG, Black WC: Are deaths within 1 month of cancer-directed surgery attributed to cancer? J Natl Cancer Inst 94:1066-1070, 2002
26. Rothenberg ML, Meropol NJ, Poplin EA, et al: Mortality associated with irinotecan plus bolus fluorouracil/leucovorin: Summary findings of an independent panel. J Clin Oncol 19:3801-3807, 2001 Submitted August 6, 2003; accepted April 29, 2004.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
