Abstract

Up to now no systematic analysis on the impact of different age categories on survival in patients with multiple myeloma has been reported. Information on possible correlations of host and tumor related prognostic factors with different age categories are lacking. We studied these parameters in a large cohort of patients with multiple myeloma (n=10.750) submitted by participating institutions and groups in the international staging system (ISS) project. Prognostic factors were recorded and age was calculated at start of initial chemotherapy. Patients were grouped into 6 age cohorts (<40, 40−<50, 50−<60, 60−<70, 70−<80 and ≥80 years). P values were calculated with the Jonckheere-Terpstra test and Spearman’s correlation coefficient was used where appropriate.

The sequential median survivals constantly decreased by decade from 61 months to 60, 53, 40, 32 and 24 months in the 6 patient cohorts from age < 40 years to age >80 years examined, respectively, with a median value of 44 months (p<0.0001). The distribution of prognostic factors by age revealed a highly significant correlation between high serum ß2 microglobulin ( Sß2M, ≥3.5mg/dl) and age, ranging from 45% in patients in the youngest to 75% of patients in the oldest age cohort (r=0.17 (0.15–0.19), p<0.0001). A similar correlation was seen between low serum albumin (<3.5g/dl) and age: The proportion of patients with low serum albumin levels increased from 32% in patients at age < 40 years to 54% in patients > 80 years (r= −0.11(−0.13, −0.09), p<0.0001). Consequently, as Sß2M and serum albumin constitute the prognostic parameters of the ISS, a close correlation between ISS stage and age was found (p< 0.0001). The proportion of patients with ISS stage I (Sß2M < 3.5mg/dl and serum albumin ≥ 3.5 mg/dL) was 40% in patients aged <40 years and only 12% in those aged ≥80 years. In contrast, 44% of patients of the oldest and 31% of the youngest age cohort presented with ISS stage III. In addition, a similar, albeit lesser trend was noted for decreasing hemoglobin with age (r=−0.08 (−0.10, −0.07, p<0.0001) and increasing serum creatinine with age (r=0.08 (0.06, 0.10), p <0.0001).

The parameters reflecting the biology of the myeloma clone did not vary between different age cohorts. Bone marrow plasma cell infiltration (BMPC) ≥33%, CRP levels ≥0.8 (mg/dL) and normal LDH was seen in similar frequencies in the different age categories. Similarly, no age dependent variation in cytogenetically defined prognostic variables was seen. The proportion of patients with Del 13 and of those with t (11; 14), t (4; 14) did not differ between the different age categories; these data were obtained in a limited number of patients only (616, 544 and 418 patients, respectively).

In conclusion, age was identified as important prognostic factor in the six different age cohorts examined. Poorer survival with higher age is closely linked to higher ISS stage. In addition, creatinine and low hemoglobin correlate, albeit to a lesser degree, with increasing age, but not parameters reflecting adverse biologic features of the tumor clone (LDH, BMPC, CRP, del 13, t(11;14) t(4;14). Hence, an ailing host and not a more aggressive tumor clone seems to account for the inverse correlation between survival and age.

Author notes

Corresponding author