- © 2007 by American Society of Clinical Oncology
Adjuvant Therapy in the Elderly: Making the Right Decision
- From the University of Vermont and Vermont Cancer Center, Burlington, VT; Department of Biostatistics, Mayo Clinic College of Medicine, Rochester, MN; Sandro Pitigliani Medical Oncology Unit, Department of Oncology, Hospital of Prato, Prato, Italy; and the Multidisciplinary Oncology Institute, Genolier, Switzerland
- Address reprint requests to Hyman B. Muss, MD, University of Vermont and Vermont Cancer Center, 1 S Prospect St, UHC Campus, St Joseph 3400, Burlington, VT 05401; e-mail: hyman.muss{at}uvm.edu
Abstract
Adjuvant chemotherapy has led to improvements in relapse-free and overall survival in patients with breast, colon, and non–small-cell lung cancer, yet many older patients are not offered these potentially life-saving treatments. Moreover, older patients have been either excluded or under-represented in most adjuvant trials, limiting the generalizability of these treatments to older populations. Limited data in elders suggest that older patients derive significant benefits from adjuvant therapies provided they have life expectancies exceeding 5 years. Making treatment decisions in elders is challenging. Many have major comorbidities that may substantially limit life expectancy and minimize or negate the benefits of adjuvant chemotherapy. In this review, we discuss the potential benefits of adjuvant treatment in older patients with solid tumors with a focus on general principles involved in the selection of adjuvant therapy for patients with breast, colon, and non–small-cell lung cancer. In addition, we discuss the role of comorbidity and how it factors in treatment decisions. Finally, we discuss future research directions and funding for elders with cancer.
INTRODUCTION
The use of adjuvant chemotherapy in patients with breast, colorectal, and non–small-cell lung cancer has led to improvements in relapse-free and overall survival. These gains have been impressive, and yet, many older patients are not offered these treatments. Many of the trials that have established the role of adjuvant chemotherapy have excluded elders, limiting the generalizability of these treatments to older populations. In addition, many elders have substantial comorbidities that may substantially limit life expectancy and the effectiveness of adjuvant chemotherapy. In this review, we discuss the potential benefits of adjuvant treatment in patients with solid tumors with a focus on the potential benefits of such treatment in elders. In addition, we discuss the role of comorbidity and how to account for it in treatment decisions. Finally, we discuss future research directions and funding opportunities in this area.
ADJUVANT THERAPY IN BREAST, COLORECTAL, AND NON–SMALL-CELL LUNG CANCER
Breast Cancer
The median age at diagnosis of breast cancer in North America and Europe is approximately 65 years, and the incidence increases with increasing age.1 In patients with endocrine-sensitive tumors, the administration of 5 years of tamoxifen reduces the annual recurrence and breast cancer mortality rates by 39% and 31%, respectively, irrespective of age.2 These impressive risk reductions are noted in women 70 years and older as well as in younger cohorts. More recent data suggest that postmenopausal patients treated with an aromatase inhibitor have even further improvements in event-free survival that average approximately 3% to 5%; data are also emerging from these trials showing small improvements in overall survival.3-7 Most older patients have hormone receptor–positive tumors and are good candidates for endocrine therapy.
The added value of chemotherapy to endocrine therapy in elders with hormone receptor–positive tumors or its use alone in elders with hormone receptor–negative tumors is an area of great debate. The most recent Oxford overview of the Early Breast Cancer Trialists' Collaborative Group showed that the benefit of polychemotherapy (combination chemotherapy) progressively decreased with increasing age.2 Polychemotherapy reduced the annual recurrence and breast cancer mortality rates by more than 35% and approximately 30%, respectively, in patients younger than 50 years, but these rates were reduced by only 23% and 9%, respectively, in patients aged 60 to 69 years. In women aged ≥ 70 years, the hazard ratios for recurrence and breast cancer mortality are 0.88 and 0.87, respectively, suggesting a benefit from chemotherapy in this group of patients as well; however, the large CIs surrounding these reductions in this subgroup reflect the small number of older patients accrued in these randomized trials. Moreover, there was a significant advantage for anthracycline- versus nonanthracycline-containing chemotherapy for improving both relapse-free and overall survival, with benefits seen in women younger than 50 years and older than 50 years. Of note, at 15 years, the proportional benefit of chemotherapy was similar for patients with estrogen receptor (ER) –positive and –negative tumors. This is important because, if we assume the same benefits of chemotherapy for women 70 years and older as women 50 to 69 years old, then the added value of chemotherapy to endocrine therapy would be most likely seen in elders with life expectancies of 10 years or more. A recent analysis of chemotherapy effectiveness in node-positive patients showed that adjuvant chemotherapy was more effective in patients with ER-negative disease.8 The International Breast Cancer Study Group Trial IX showed that the added benefit of three cycles of cyclophosphamide, methotrexate, and fluorouracil to tamoxifen for patients with lymph node–negative disease was dependent on ER status and that, for patients with ER-positive tumors, there was no benefit from cyclophosphamide, methotrexate, and fluorouracil plus tamoxifen compared with tamoxifen alone.9 Similar results were found in the National Surgical Adjuvant Breast and Bowel Project B-20 trial for patients 60 years and older.10 In a gene assay from tumor samples of patients entered onto the National Surgical Adjuvant Breast and Bowel Project B-14 and B-20 trials, Paik et al11,12 showed that only patients with tumors that had high risk of metastases at 10 years had a large absolute benefit of chemotherapy if they were treated with tamoxifen. In these studies, older patients were found to have tumors with a more favorable profile for endocrine responsiveness compared with younger patients. However, a retrospective analysis of intensive versus less intensive therapies in node-positive patients showed that patients aged 65 years and older derive similar proportional improvement in relapse-free and overall survival as younger patients but with a higher rate of treatment-related mortality.13 The dose of chemotherapy may also be important. In one detailed analysis, older patients were less likely to receive standard doses of chemotherapy compared with younger patients.14 Whether these lower doses affected outcome is uncertain, but elders should be dosed similar to younger patients and also according to the published dose modifications described for standard regimens. The challenge of selecting adjuvant therapy for seniors is discussed in detail elsewhere in this monograph. Current data suggest that chemotherapy is of greatest value in seniors with node-positive, ER- and progesterone receptor–negative breast cancer.15 New molecular and genetic assays will help to refine these decision criteria and to select patients most likely to benefit from adjuvant systemic therapy.
Colon Cancer
The median age of patients diagnosed with colon cancer is 72 years, and the incidence of the disease increases with increasing age.16 Until surveillance and prevention strategies dramatically impact the incidence of colon and rectal cancers and as the life expectancy of the world's population increases, physicians will continue to see increasing numbers of older patients with colon cancer. Chemotherapy has been proven to improve outcomes in patients with node-positive (stage III) disease, with benefit suggested in at least selected patients with node-negative (stage II) disease. In these patients, chemotherapy can make the difference between recurrence and cure. However, chemotherapy is not curative in all patients and is accompanied by considerable toxicity. Is it reasonable to recommend a potentially toxic and expensive 6-month course of treatment to a 75-, 80-, or 85-year-old person?
Adjuvant chemotherapy is considered to be standard of care for patients with stage III colon cancer based on an improvement in overall 5-year survival rate from approximately 50% (with no adjuvant therapy) to 65% (with fluorouracil-based therapy) after surgical removal of the primary tumor.17 Recently, the standard of care for stage III patients has evolved to add a third agent, oxaliplatin, to the formerly standard regimen of fluorouracil plus leucovorin therapy.18,19
Population-based data derived from the Surveillance, Epidemiology, and End Results registry indicate that the likelihood of receiving adjuvant chemotherapy after resection of stage III colon cancer declines with age. In the late 1990s, 78% of patients aged 65 to 69 years with stage III cancer received adjuvant chemotherapy compared with just 34% of patients aged 80 to 84 years.20 In addition, once adjuvant therapy is initiated, elders are more likely to have it discontinued before completion, possibly decreasing its effectiveness.21 This underscores the importance of trying to minimize treatment-related toxicity in elders. However, several recent analyses of clinical trials that compare the outcomes of patients aged 70 years and older with those of patients younger than age 70 years suggest that older patients have similar benefits and similar patterns and severity of adverse events compared with their younger counterparts. An absolute survival advantage of 7% was found in both patients groups.22 Similar benefits for elders and younger patients have been noted for the regimen of infusional fluorouracil, leucovorin, and oxaliplatin.23 Additionally, the dose-intensity of infusional fluorouracil, leucovorin, and oxaliplatin as adjuvant therapy was comparable regardless of age. However, increasing age will almost always be associated with poorer overall survival because the probability of dying of noncancer causes increases as one ages.24 For these reasons, age is not a predictive factor for the benefits of chemotherapy in colon cancer, but it will always be a prognostic factor for overall survival.
Older patients entered onto clinical trials are a selected group who generally do not suffer from limiting comorbid conditions. Population-based registries provide an additional data source for assessing the efficacy of adjuvant therapy for colon cancer. In a Surveillance, Epidemiology, and End Results–based population study, Sundararajan et al25 reported a hazard ratio of 0.66 (95% CI, 0.60 to 0.73) in favor of fluorouracil-based treatment compared with surgery alone in patients with stage III disease aged 65 years or older. Similar data were reported by Iwashyna and Lamont.26 Population-based studies do not feature random assignment; thus, firm casual conclusions cannot be drawn, but taken together with the clinical trial data previously discussed, the consistency of these data strongly suggests a benefit for chemotherapy in elderly patients deemed sufficiently fit to receive such therapy. Encouragingly, there are some data to suggest that these findings are being translated into clinical practice. In 1990 to 1991, approximately 22% of patients aged 80 years and older received treatment, whereas a decade later, nearly 40% of patients in this age group received treatment.27
Recent studies indicate that 80% of patients who will develop recurrent colon cancer do so within the first 3 years, and nearly all have manifest recurrent disease within 5 years of diagnosis.28 Because the fit elderly individual has a life expectancy exceeding 5 years in most cases, consideration should be given to adjuvant therapy for fit elders regardless of age.
Non–Small-Cell Lung Cancer
More than half of all patients diagnosed with lung cancer are aged 65 years and older, and incidence increases with increasing age. Lung cancer is the leading cause of cancer and cancer-related mortality worldwide, with the vast majority of patients diagnosed at an advanced, noncurable stage. Early diagnosis is uncommon but may be improved by new technologies such as spiral computed tomography scanning.29,30 Although older adjuvant chemotherapy trials were generally negative, more recent randomized trials that include platinum- or fluorouracil-based chemotherapy have shown statistically significant improvements averaging approximately 5% in both relapse-free and overall survival.31,32 Although there are concerns about the potential toxicities of platinum-based regimens,33 a retrospective analysis of almost 600 patients entered onto randomized platinum-based trials of the Eastern Cooperative Oncology Group, of whom 15% were aged 70 years and older, showed similar response rates, survival, and toxicity in fit elderly patients compared with younger patients; however, leukopenia and neuropsychiatric toxicities were higher in elderly patients.34 Specific trials focused on older patients exploring less toxic regimens should be considered in this setting. In addition, recent genetic tests that more accurately predict recurrence than clinical characteristics alone35 or that predict the potential value of platinum-based regimens36 would be ideal to study in fit elderly patients with non–small-cell lung cancer.
ROLE OF COMORBIDITY
Aging is associated with an increased incidence of comorbidities and loss of function, which can decrease survival or lead to a need to adapt cancer therapies.37 Comorbidities as competing causes of mortality can assume a large role in determining survival in older cancer patients38-41 and can minimize or negate the benefit of adjuvant treatment.42 In one study, the presence of three or more comorbid conditions was a strong predictor of survival, independent of tumor stage (Table 1). 43
Influence of Comorbidity on 3-Year Survival of Women With Breast Cancer
Prognostic indexes that include both comorbidities and function have been developed for older patients, and both comorbidity and function are always investigated in a comprehensive geriatric assessment (CGA). The CGA “is a multidisciplinary evaluation in which the multiple problems of older persons are uncovered and described, and in which the recourses and strengths of the person are catalogued, the need for service assessed, and a coordinated care plan developed, to focus interventions on the person's problems.”44 The domains evaluated in the CGA are listed in Table 2.
Key Domains Evaluated in a Comprehensive Geriatric Assessment
An expert task force of the International Society of Geriatric Oncology concluded that a CGA should be used in older cancer patients to detect unaddressed problems and to recommend interventions to improve functional status and possibly improve survival.45 On the basis of the evidence that patients seen in the oncology setting tend to be healthier and less disabled than traditional geriatric patients,46 frail and vulnerable patients seem to be the ideal candidates for a CGA-based approach. The best method of geriatric assessment for cancer patients remains to be defined. As a practical example, a combination of tools used frequently for CGA includes scales for assessing the activities of daily living and instrumental activities of daily living, the Folstein Mini-Mental State Examination, and the Geriatric Depression Scale. Several reliable indexes are also available to measure comorbidity in cancer patients.47 The Charlson Comorbidity Index is probably the most widely used index and is valid for predicting mortality risk up to 10 years in a variety of conditions. Because little is still known about the effect of individual comorbidities on the impact of cancer treatment, assessing comorbidity in clinical trials is essential. The Breast International Group has added the Charlson Comorbidity Index to an ongoing adjuvant chemotherapy trial targeting elderly breast cancer patients. In addition, some of the Breast International Group trials will also prospectively evaluate the use of the Vulnerable Elders Survey 13 scale to see whether it can predict which patients are at high risk of death from non–breast cancer causes.48 A short, mostly self-administered CGA instrument will also be prospectively tested by the Cancer and Leukemia Group B in several disease sites.49
SELECTING ELDERLY PATIENTS FOR ADJUVANT THERAPY AND CLINICAL TRIALS
Selecting patients for adjuvant therapy remains a great challenge. The use of web-based tools such as Adjuvant! (www.adjuvantonline.com) can be most helpful in predicting the benefits of adjuvant therapy in older patients with breast and colon cancer and incorporates life expectancy data based on age.50 An example of how this model can help estimate the benefits of adjuvant therapy is shown in Figure 1. Irrespective of the potential benefits of treatment, a careful discussion of adverse effects is mandatory. Other key issues related to treatment selection are listed in Table 3.
Benefits of specific adjuvant therapies at 10 years for a 66-year-old woman with a 2-cm, estrogen receptor–positive, moderately differentiated infiltrating ductal carcinoma with two positive lymph nodes. T, tamoxifen; AC, cyclophosphamide and doxorubicin; A > 2 = an anthracycline regimen containing more than two agents; CMF, cyclophosphamide, methotrexate, and fluorouracil. From Adjuvant! (www.adjuvantonline.com).
Key Issues Related to Treatment Selection
In many clinical trials, elderly patients have been explicitly excluded by eligibility criteria, perhaps for fear of excessive adverse events. When elders have not been excluded, the proportion of elderly patients enrolled onto trials significantly under-represents the relative disease prevalence in the population.28,51-54 This under-representation is more pronounced for early-stage cancer trials compared with trials in metastatic cancer.54 These factors have led some to advocate age-specific clinical trials, with enrollment limited to only patients exceeding an age cutoff. What would be of great value to clinicians considering elders for clinical trials would be a simple, rapid measure of assessing noncancer-related functional loss and mortality risk before entry. Recently, Lee et al55 described a simple validated prognostic index for community-dwelling elders that included age, self-reported comorbidity, and several functional measures and that accurately predicted 4-year mortality. Markers of inflammation might also be used to predict functional loss and mortality.56,57 Future trials that involve elders should consider collecting such data to see whether this information can predict functional loss and mortality risk.
Age group–specific trials have considerable appeal. Regimens that are considered to have relatively lower toxicity or fewer contraindications as a result of various comorbidity-related factors are obvious targets for such age group–specific trials. Such trials have some history of successful accrual34,58,59 and are clearly appropriate if a therapy has been established as safe and effective in a non–age-selected population. Trials designed to test for the noninferiority of a less toxic regimen compared with an established standard are also appropriate and may be the only mechanism to accrue the oldest of the old.
Elderly-specific trials must be considered cautiously. Regimens that are designed to be less toxic and, therefore, more suitable for the elderly may also be less efficacious. A single-arm trial of such a regimen, without a randomized comparison to an established standard, cannot be used to make such a determination. Likewise, randomized trials in which neither arm is considered standard of care suffer the same difficulty. Given these considerations, whenever possible, a preferred solution is to include elderly patients in a non–age-selected clinical trial, with the eligibility criteria designed to be as flexible as possible.54,60 Age may be considered as a stratification factor for the random assignment, and a prospectively specified subgroup analysis based on age may be included. Additionally, an age-specific adverse event monitoring plan may be included in the protocol to cease enrollment of the elderly if emerging data convincingly demonstrate that age is associated with an excessive risk of experiencing an adverse event.
Including elderly patients in non–age-specific clinical trials will not likely allow age-specific efficacy comparisons within an individual trial because of lack of statistical power. However, including these patients does allow for the exploration of secondary, age-related hypotheses, which may include a comparison of toxicity, or translational research projects. The inclusion of elderly patients in clinical trials allows data from multiple trials to be pooled for meta-analyses, which may allow definitive answers to age-specific questions, as has been demonstrated for breast cancer13 and colon cancer.22
FUTURE DIRECTIONS
Future research directions in adjuvant therapy should focus on integrating information on comorbidity and function with outcome. These data could then be used to develop appropriate exclusion criteria for clinical trials. By integrating a standardized geriatric assessment tool into a trial, toxicity and outcome can be analyzed in conjunction with geriatric assessment. Such a tool could then be used to define populations of elders who might benefit from specific treatments or those in whom treatment poses greater risk than benefit. Tools designed for this purpose are currently being evaluated. In addition, prospective medical data on elders, such as hemoglobin levels, albumin levels, and creatinine clearance, and information on potentially harmful drug interactions should be obtained and correlated with treatment outcome and toxicity.
Stronger efforts must be made to include elders in all clinical trials. Collaboration among clinical trial cooperative groups among all nations should be a key goal. To achieve these goals, elders and their physicians must be willing to become vocal advocates for increased funding. Finally, we must train cancer researchers in both gerontology as well as oncology. Without new investigators, progress will continue to be slow. All of us involved in the care of such patients must be willing to advocate for increased funding for research focused on decreasing the burden of cancer in elders.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The authors indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Hyman B. Muss, Laura Biganzoli, Daniel J. Sargent, Matti S. Aapro
Manuscript writing: Hyman B. Muss, Laura Biganzoli, Daniel J. Sargent, Matti S. Aapro
Final approval of manuscript: Hyman B. Muss, Laura Biganzoli, Daniel J. Sargent, Matti S. Aapro
Footnotes
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received December 19, 2006.
- Accepted February 7, 2007.
