Evaluation of the Revised International Staging System (R-ISS) in an Independent Cohort of Unselected Patients with Multiple Myeloma

The International Staging System (ISS), which is based on beta2-microblobulin (β2M) and serum albumin, has been widely used for the risk stratification of multiple myeloma (MM) patients, since 2003. Chromosomal abnormalities (CA) detected by iFISH have been also recognized as strong prognostic factors, while elevated serum lactate dehydrogenase (LDH) has been consistently associated with poor prognosis. In order to improve the prognostic power of ISS, IMWG has revised ISS (R-ISS) by adding high risk cytogenetics by iFISH and serum LDH: R-ISS-1 includes patients with ISS-1 (serum β2M level <3.5 mg/L and serum albumin level ≥3.5 g/dL), no high-risk CA [del(17p) and/or t(4;14) and/or t(14;16)] and normal LDH levels (below the upper limit of normal (UNL)); R-ISS-3 includes patients with ISS-3 (serum β2M level >5.5 mg/L) and either high-risk CA or high LDH level; and R-ISS-2 includes all the other possible combinations. R-ISS was based on the data of 3,060 patients who had participated in clinical trials. However, it has not been validated in an independent cohort of unselected patients. Our aim was to evaluate R-ISS in consecutive, unselected patients, treated in a single center.

Our study included 475 patients with available data for cytogenetics [t(4;14) and del17p by iFISH], LDH and ISS. Median age was 67 years (range 27-91); 53% were >65 years and 25% >75 years of age. Only 8.6% did not receive new drugs as primary therapy; 42% received IMiDs (19% thalidomide-based, 23% lenalidomide-based) and 49% bortezomib-based primary therapy, while 36% received ASCT. In the IMWG cohort, 65% had received ASCT, 6% no new drugs, 44% proteasome inhibitors and 66% IMiDs.  Per ISS, 24% were ISS-1, 34% ISS-2 and 42% ISS-3. High risk cytogenetics (either t(4;14) or del17p) were present in 23.5% and elevated LDH in 15%. In the IMWG cohort used for the formulation of R-ISS, 38% were ISS-1, 38% ISS-2 and 24% ISS-3; thus, our patients had more often ISS-3 and less often ISS-1 disease. High risk CAs and elevated LDH were not different compared to our cohort of patients (24% and 13%, respectively). The difference in ISS disposition between the two cohorts probably reflects the unselected nature of our population, which also included patients with severe renal impairment who often are excluded from clinical trials.

Per R-ISS, 85 (18%) patients had R-ISS-1, 83 (17.5%) had R-ISS-3 and 306 (64.5%) had R-ISS-2. The disposition in the original cohort was 28% for R-ISS-1, 62% for R-ISS-2 and 10% for R-ISS-3. This difference was due to the higher proportion of patients with ISS-3 disease in our cohort. The R-ISS disposition in those ≤65 years, was 21%, 60% and 19% for R-ISS-1, -2 and -3; among patients 66-75 years it was 19%, 63% and 18%, and among those >75 years it was 11%, 74% and 15%, respectively (p=0.128). The median follow up was 40 months; 57% of patients progressed or died and 63% have remained alive. Median PFS was 27 months and estimated median OS was 63 months. Median PFS for R-ISS-1,-2 and -3 was 34, 28 and 17 months, respectively (p<0.001). According to R-ISS, the probability of 3-year OS was 83%, 69% and 45% and of 5-year OS 77%, 53% and 19% for R-ISS-1, -2 and -3, respectively (p<0.001). In patients treated with ASCT, the probability for 5-year OS per R-ISS stage was 93%, 77% and 32%, respectively (p<0.001), while for patients not treated with HDM it was 64%, 41% and 13% (p<0.001). The probability for 5-year OS for patients treated with bortezomib was 95%, 69% and 18% for R-ISS-1, -2 and -3 (p<0.001) and for those treated with IMiDs, it was 68%, 41% and 23%, respectively (p=0.002).

We evaluated the performance of R-ISS in patients ≤65, 66-75 and >75 years. In patients ≤65 years, the probability for 5-year OS was 84%, 71% and 29%, for R-ISS-1, -2 and -3 (p<0.001); for patients 66-75 years, it was 73%, 43% and 18% (p=0.001), while in patients >75 years, the median OS was >5 years, 35 and 29 month, respectively (p=0.122). Thus, R-ISS identified a group of patients >75 with favorable prognosis, although there was no significant difference in the OS for R-ISS-2 vs -3, probably due to the impact of other comorbidities and performance status of the very elderly.

In conclusion, our data in consecutive, unselected patients, with differences in the characteristics and treatment approaches compared to the original IMWG cohort, verified that R-ISS provides significant prognostic information and it allows the identification of three different patient groups with clearly different outcome.