Follow-up of Real Life Treated Multiple Myeloma Patients: Response, Disease Progression and Overall Survival,

Abstract 3937

Treatment results in clinical practice and in real life can be different from each other. Therefore, evaluating real life outcomes in Multiple Myeloma (MM) patients in order to understand treatment outcome in clinical practice is very important.

All patients diagnosed with MM and treated between Jan 2000 and Jul 2010 at Karolinska University Hospital, Huddinge and Södersjukhuset were included. Furthermore, all diagnosed patients between Jan 2005 and Jul 2010 at Karlolinska University Hopital, Solna were included. At diagnosis data regarding age, gender, type of myeloma and skeleton destruction as well as Ca, Hb, β2-microglobelin, albumin and creatinine were collected. For each treatment line, drugs given and specific start and stop dates for each drug were collected. Serum and urine M-protein was collected at baseline and each time the measurement differed from the previous. Near complete response (nCR) was defined as an immeasurable M-protein in the blood and urine by standard electrophoresis but not always confirmed by immunofixation since it was not required in clinical practice. Partial response (PR) was defined as a 50% reduction in M-protein. No response (NR) was defined as less than 50% reduction of M-protein. Progress was considered when M-protein increased with ≥ 25%. Novel agents were defined as bortezomib, lenalidomide and thalidomide.

In the total population n=674, 89 (13%) patients were excluded due to non treatment demanding disease or plasmacytoma without MM diagnosis setting the study population to n=585. The median follow up of the study population was 815 days. The median age in the study population was 68 years and 45% were women. High dose treatment (HDT) was given to 214 (37%) patients. The median age in the HDT population was 58 years and 34% were women. The median age in the non-HDT population was 75 years and 51% was women. The median number of given treatment lines was 2. The proportion of patients receiving more treatment lines were declining rapidly with the number of received lines of therapy. The response distribution for the entire population nCR/PR/NR was 27/41/32 in 1st line, 18/34/48 in 2nd line. In the HDT population the same response distribution was 50/38/12 in 1st line and 28/38/34 in 2nd line. In the non-HDT population the response distribution was 14/43/43 in 1st line and 12/33/55 in 2nd line. The depth of the responses was significantly dependent on the line of therapy. Responses were significantly better in the HDT population vs. the non-HDT population in the first two treatment lines. There seems to be a correlation between the responses in 1st and 2nd line in the entire population. Improving the responses in 2nd line compared to 1st line seems not very likely. Statistical analysis shows that the non-HDT patients receiving novel agents in 1st line had a significantly higher probability of achieving nCR, 25% vs 9%.The same statistical difference was seen in the HDT patients with nCR rates of 65% vs 47%. The median TTP/TTNT for the study population was 309/381 days in the 1st line and 182/210 in 2nd line. In the HDT population the TTP/TTNT was 538/639 days in 1st line and 199/257 in 2nd line. In the non-HDT population the TTP/TTNT was 244/295 days in 1st line and 177/193 in 2nd line. In the HDT population both the TTP and the TTNT in the 1st line were significantly better than in the non-HDT population with no difference in the 2nd line. There was a significant trend of increasing TTP/TTNT in 1st line depending on the increased depth of the response. Statistical analysis showed that the use of novel agents in 1st line predicted a longer TTP for the non-HDT population. Median OS was 4.3 years 95% CI [3.9;5.0] with 53% censored. HDT patients had a significant longer median OS 6.3 years 95% CI [5.4;8.5] than the non-HDT patients OS 3.0 years 95% CI [2.2;3.6]. The median OS for non-HDT patients with 1st line best response nCR was 4.9 years, PR 3.3 years and NR 1.5 years. Kaplan-Mayer analysis shows that use of novel agents in the 1st line predicted a longer OS, median 5.1 years vs. 4.1 years.

Patients receiving 2nd and 3rd line treatment were declining rapidly. To get a good response in 1st line increases the likelihood of having a good response in 2nd line. Receiving an nCR in 1st line seems to be very important and this study confirms the prognostic impact of achieving at least nCR. The use of novel agents improves the outcome for patients in clinical practice.