Conditional Melanoma Cancer Survival in the United States.

Beyond relative survival, which indicates the likelihood that patients will not die from causes associated with their cancer, conditional relative survival probabilities provide further useful prognostic information to cancer patients, tailored to the time already survived from diagnosis. This study presents conditional relative survival for melanoma patients in the United States, diagnosed during 2000-2008 and followed through 2012. Analyses are based on 62,803 male and 50,261 female cases in population-based cancer registries in the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute. Five-year relative survival estimates are presented for melanoma patients who have already survived one, two, three, four, or five years after the initial diagnosis. Five- and ten-year relative survival decreases with age, stage at diagnosis, and is lower among males, Blacks, and Hispanics. Five-year conditional relative survival improves with each year already survived. The potential for improvement in five-year conditional relative survival is greatest for older age, males, Blacks, Hispanics, and in later staged cases. For local disease, five-year conditional relative survival was significantly lower in ages greater than 65 years and in Blacks. It was significantly higher in females, non-Hispanics, and married individuals. Age had a greater inverse relationship with five-year survival in later staged disease. A similar result occurred for females and married individuals. In contrast, non-Hispanics had better five-year survival if diagnosed with local or regional disease, but not distant disease.