Special Feature
Ovarian Cancer: Can One Be Too Old for Cancer Care?
By Robert L. Coleman, MD, Associate Professor, University of Texas; M.D. Anderson Cancer Center, Houston, is Associate Editor for OB/GYN Clinical Alert.
It is now well appreciated that the combination of increasing life expectancy and declining fertility rates is producing a generation of older individuals—a trend that is anticipated to continue into the foreseeable future. In fact, both the number of septuagenarian and octogenarian cohorts in the US population are expected to more than double in number and will account for a significantly higher proportion of the population within the next 25 years according to the United States Census Bureau. These projections are impacting medicine in many ways from health care policy discussions to strategic resource allocation in preparation of this expanded legion. In regard to disease, many challenges will present themselves, not the least of which will be growth in those conditions which have incidence rates that do not peak until the 70’s and 80’s. A representative model in this regard is ovarian cancer where peak annual incidence rises until approximately age 75 before trending downward. The traditional explanation for this observation is not only the great time to be exposed to "events" leading to the malignant process but also a declining denominator—those women available from exposure.
Despite the incidence rates of this disease and the potential age migration over time, there is a clear bias as to the clinical approach for patients as they age. Frequently, co-morbidities are encountered which limit the degree and intensity of therapy. In other cases though, a significant variance is rooted in either anticipated morbidity or response to treatment. In this manner, clinicians appropriately make selective recommendations to care based on the overall big picture. Since the usual treatment plan for ovarian cancer patients involves a combination of cytoreductive surgery and adjuvant chemotherapy, the risk events involved are not only immediate but also extend through several months of treatment with the potential of cumulative effects. Nonetheless, the question arises as to whether this bias is soundly rooted. In addition, can one quantify this bias to establish a risk profile from which an evidence-based recommendation to care be made?
An article published recently by Uyar and colleagues1 attempted to characterize surgical and chemotherapeutic management of elderly patients by evaluating and comparing the patterns of care between patients 70-79 and older than 80 years of age. The retrospective study collected patients 70 years and older who underwent chemotherapy for primary disease at a single institution. Several variables were collected from a review of the medical records including an important (and validated) objective measure of co-morbidity—the Charlson Scale. This 19-item scale numerically quantifies the 12-month survival risk based on attendant co-existing medical ailments. Patients were excluded if they underwent surgery but did not get adjuvant chemotherapy. In all, 131 patients were identified; 90 (69%) were 70-79 years of age. The average age between the cohorts differed by approximately 9 years (74 vs 83). Stage, tumor type, and race were similar between the cohorts. Importantly, the degree of co-morbid illness as quantified by the Charlson Scores was similar between the cohorts. Balance of this confounding variable allows a less restrictive analysis of the outcomes but likely reflects selection bias in those presenting for care. The younger cohort was significantly more likely to undergo surgery and get standard platinum and taxane-based combination chemotherapy.
In regard to chemotherapy tolerance, no difference between the age cohorts was observed for treatment delays or discontinuation of therapy—although the latter was more frequent in those getting combination regimens as compared to monotherapy. Neither age nor the Charlson scores influenced these variables. The older cohort did have a higher degree of Grade 3 and 4 hematologic toxicity when administered combination chemotherapy regimens; however, no difference was observed for monotherapy regimens. Restricting the analysis to those patients in both cohorts who underwent surgery, the percentage of patients achieving optimal cytoreduction was similar by age class and, thus, overall survival between the cohorts was similar. In the multivariate model considering age, stage, type of chemotherapy and surgery, only the latter was independently significant to overall survival. These data indicate that the elderly (80 and older) are less likely to receive surgery and combination chemotherapy despite similar tolerance to their less elderly counterparts.
Data from this trial mirror those reported previously but not without exception.2,3 Age and performance status have routinely been pointed to as adverse outcome variables. Thigpen, reporting on behalf of the Gynecologic Oncology Group (GOG) demonstrated that these 2 factors along with volume of residual disease as adverse variables to outcome in a pooled analysis of more than 2000 patients treated on GOG protocols.4 While age (older than 69 years) negatively affected prognosis, there was no evidence that this age group could not tolerate dose-intensive chemotherapy. The authors advocated that this patient cohort be treated as their younger counterparts, particularly since they were at greater risk for poorer survival.
Others have countered that the age naturally is associated with other factors (such as cytoreducibility, tumor kinetics, stage and tolerance to chemotherapy), which distinguishes the senior cohort from younger patients. While on the surface this may be true, age, independently appears to be less of a predictive factor when other co-morbidities and performance status are considered. Complicating matters further, "elderly" has been variably defined making analyses of this cohort, in general, difficult to synthesize. Indeed, chronological age and functional age are poorly correlated.
The one unifying characteristic of the data available is the retrospective nature with which has been collected and analyzed. In this sense, the conclusions offer no greater inference than hypothesis-generating statements. Clearly, a prospective clinical trial with quality-of-life measures, standardization of therapy and careful stratification of important covariates is needed to tease through the important concepts of recommending a treatment program for "elderly patients." The GOG has recently endorsed a protocol to evaluate age and outcomes and will be a welcomed addition to the gynecologic oncology body of literature.
References
- Uyar D, et al. Treatment patterns by decade of life in elderly women (> or = 70 years of age) with ovarian cancer. Gynecol Oncol. 2005;98:403-408.
- Balducci L. Evidence-based management of cancer in the elderly. Cancer Control. 2000;7:368-376.
- Lawton FG, Hacker NF. Surgery for invasive gynecologic cancer in the elderly female population. Obstet Gynecol. 1990;76:287-289.
- Thigpen T, et al. Age as a prognostic factor in ovarian carcinoma. The Gynecologic Oncology Group experience. Cancer. 1993;71:606-614.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content