Decreased Survival Rates of Older-Aged Patients with Melanoma: Biological Differences or Undertreatment?

Abstract

One of the most interesting unanswered questions in melanoma research is the paradoxical observation that melanomas among older patients have a different natural history and survival outcome compared with middle-aged and younger melanoma patients. Since 1978, a number of studies have published studies demonstrating that patient age is an independent predictive factor of melanoma survival, and that patient age independently predicts the incidence of sentinel node metastases among clinically node-negative patients. In this issue of the Annals of Surgical Oncology are two interesting articles analyzing patient age and the incidence of nodal metastases among melanoma patients. Using the Surveillance, Epidemiology, and End Results Program (SEER) population-based database of 47,577 melanoma patients, Cavanaugh-Hussey and colleagues found that older melanoma patients had a higher mortality rate and a lower incidence of sentinel node metastases compared with younger patients. This fascinating inverse relationship between sentinel node metastasis rate and patient age was first described in 2004 by Chao et al. and Sondak et al. We recently analyzed outcomes of 7756 patients from the American Joint Committee on Cancer (AJCC) Melanoma Staging database and found that primary melanomas became more advanced with increasing age by decade— tumors were thicker, exhibited higher mitotic rates, and were more likely to be ulcerated—and that patient age was an independent predictor of sentinel node metastases in a multifactorial analysis (p 0.00012). However, survival results from both our studies and that of CavanaughHussey found, paradoxically, a significant decrease in the incidence of sentinel node metastasis as patient age increased, and an increased mortality compared with younger patients. Do these conclusions mean that we should be more conservative in our recommendations to older patients? Not according to the second study published in this month’s issue of Annals of Surgical Oncology by Sabel and colleagues who analyzed the outcomes of 952 melanoma patients 75 years of age or older from the University of Michigan Melanoma Database. The incidence of sentinel node metastases was 25 % among patients selected for surgery in the Michigan study, whereas the incidence of sentinel node metastases in the AJCC study was lower at 15.5 % for patients C80 years of age. How can we reconcile these differences? Both the SEER and AJCC melanoma databases are more representative of national patient populations, whereas the University of Michigan database may reflect ‘referral bias’ that is inherent to patients referred to a regional academic referral center. Thus, patients who were more frail or debilitated might not have been able to make the journey to a regional referral center. When reporting our results, those of us in academic referral centers need to be cognizant that our patient population is not necessarily reflective of that in a community-based practice. In addition, there is the physician judgment and experience that greatly influences our surgical recommendations as to whether older patients might benefit from this staging procedure and undergo an operation safely. This ‘physician expert’ bias is not necessarily reflected in current measurements of patient frailty. Thus, the Charleston Frailty Index in the study by Sabel et al. did not influence the decision to conduct a sentinel lymph node biopsy, most likely because this frailty index does not represent all the issues of co-morbidity that we, as surgeons, take into account when making recommendations regarding surgery. Society of Surgical Oncology 2015