Multiple myeloma in younger patients: the role of age as prognostic factor.

The presenting features of 356 previously untreated multiple myeloma (MM) patients grouped according to age were analyzed in order (a) to elucidate the possible differences in initial clinical and laboratory features between patients younger than 50 years and the older ones and (b) to statistically assess the prognostic value of the parameters considered, with particular emphasis on the prognostic impact of age. Patients were divided into two groups: group I included 61 patients aged less than 50, group II comprised 295 patients aged 50 or more. No significant differences were found between the two groups in terms of either clinical or laboratory initial characteristics. The treatments adopted and the response to therapy did not differ in the two groups. The prognostic value of presenting features was evaluated for the whole cohort by univariate and multivariate analysis, considering both the observed survival rates and survival rates corrected for the effect of other independent causes of death, using a Poisson model. In both models, calcium level (RR 2.33), performance status (RR 1.83), and creatinine (RR 1.69) maintained their independent negative prognostic value. In contrast, the impact of age was different in the two models. In fact, patients younger than 50 seem to have a better prognosis when the observed survival rates are considered, but they show an increased risk of death when the model takes into account the expected mortality of the underlying population. In conclusion, this study shows that the younger cohort of MM patients has no distinctive initial characteristics with respect to older patients. In multivariate analysis, creatinine levels, calcemia, and performance status show a relevant negative independent prognostic value. Regarding the prognostic impact of age, survival seems to be better among patients younger than 50 than in older patients when the observed survival rates are considered but is significantly worse when the mortality of the corresponding general population, is taken into account.