Psychosocial predictors of survival in metastatic melanoma

Purpose: Research interest in psychosocial predictors tars Of the onset and course of cancer has been active since the 1950s, However, results have be en contradictory and the literature is noted for methodologic weaknesses, I n this prospective study, we aimed to systematically obtain data on psychos ocial factors associated with human response to illness. Patients and Methods: One hundred twenty-five patients with metastatic mela noma completed questionnaires measuring cognitive appraisal of threat, copi ng, psychologic adjustment, perceived aim of treatment, social support, and quality of life (QOL). Questionnaires were completed, where possible, ever y 3 months for 2 years after diagnosis. Survival was measured from date of study entry to date of death or was censored at the date of last follow-up for surviving patients. Results: In a multivariate Cox regression analysis of baseline data, which controlled for demographic and disease predictors, the psychologic variable s of perceived aim of treatment (P < .001), minimization (P < .05), and ang er (P < .05) were independently predictive of survival, Patients who were m arried (P < .01) and who reported a better QOL (P < .05) also survived long er. Conclusion: The prognostic significance of psychologic and QOL scores remai ned after allowance for conventional prognostic factors. If these associati ons reflect an early perception by the patient or doctor of disease progres sion, then measures are at least valuable early indicators of such progress ion. If psychologic processes have a more direct influence on the course of the underlying illness, then it may be! possible to manipulate them for th erapeutic effect. We are now conducting a randomized controlled trial of a psychologic intervention to further elucidate these issues. (C) 1999 by Ame rican Society of Clinical Oncology.