Background: There is a need for reliable, easily measurable laboratory markers that may help dermatologists to predict the course of mycosis fungoides (MF) when they first evaluate their patients. Objectives: Our objective was to identify clinical, haematological or immunological parameters as predictors of mortality in patients with MF. Methods: We conducted a retrospective study on a prevalent cohort of 124 patients with MF hospitalized at IDI-IRCCS, Rome, Italy, from 1983 to 2001. We calculated the proportion of patients surviving (Kaplan-Meier product-limit estimates) 5 and 10 years after first hospital admission, and hazard ratios (HR) from the Cox proportional hazards model. Results: Patients’survival was linked to age and staging (lower survival in older patients and in patients with staging IIB-IV). Higher numbers of white blood cells (WBC) and neutrophils, lower numbers of CD8+lymphocytes, low haematocrit and lower levels of albumin were significantly associated with a lower survival probability. When simultaneously accounting for age and staging, CD8+[HR = 3.02, 95%confidence interval (CI) 1.01-9.07 for CD8+< 250 vs. ≥600 cells μL-1] and WBC (HR = 2.59, 95%CI 0.96-6.96 for WBC ≥9000 vs. < 6000 cells μL-1) were associated with survival. In addition, we observed an exceedingly high risk of death (HR = 12.40, 95%CI 3.11-49.43) for patients with a combination of WBC ≥9000 and CD8+< 600 cells μL-1 vs. WBC < 9000 and CD8+≥600 cells μL-1). Conclusions: The measurement of CD8+cells and WBC in MF seems to be a promising criterion to predict survival, and possibly to support treatment decisions and inclusion of patients in randomized controlled trials. Background: There is a need for reliable, easily measurable laboratory markers that may help dermatologists to predict the course of mycosis fungoides (MF) when they first evaluate their patients. Objectives: Our objective was to identify clinical, haematological or immunological parameters as predictors of mortality in patients with MF. Methods: We conducted a retrospective study on a prevalent cohort of 124 patients with MF hospitalized at IDI-IRCCS, Rome, Italy, from 1983 to 2001. We calculated the proportion of patients surviving (Kaplan-Meier product- limit estimates 5 and 10 years after first hospital admission, and hazard ratios (HR) from the Cox proportional hazards model. Results: Patients’ survival was linked to age and staging (lower survival in older patients and in patients with staging IIB-IV Higher numbers of white blood cells (WBC) and neutrophils, lower numbers of CD8 + lymphocytes, low haematocrit and lower levels of albumin were significantly associated with a lower s urvival probability. When simultaneously accounting for age and staging, CD8 + [HR = 3.02, 95% confidence interval (CI) 1.01-9.07 for CD8 + <250 vs.> 600 cells μL- 0.96-6.96 for WBC ≥9000 vs. <6000 cells μL-1) were associated with survival. In addition, we observed an exceedingly high risk of death (HR = 12.40, 95% CI 3.11-49.43) for patients with a combination of WBC ≥9000 and CD8 + <600 cells μL-1 vs. WBC <9000 and CD8 + ≥600 cells μL-1). Conclusions: The measurement of CD8 + cells and WBC in MF seems to be promising criterion to predict survival, and possibly to support treatment decisions and inclusion of patients in randomized controlled trials.