Abstract
Abstract 117
There are few therapies for thrombocytopenia in MDS, which is found in ∼50% of pts with low/int-1 MDS and is associated with shortened survival.
IPSS low/int-1 MDS pts receiving supportive care, with platelets 1) ≤20×109/L or 2) ≤ 50×109/L with a history of bleeding, were randomized 2:1 to 750 μg romiplostim:PBO for 26 wk with MDS supportive care, and a 4-wk washout followed by bone marrow (BM) biopsy. Pts continued as randomized, with any MDS therapy, for 24 wk with another 4-wk washout followed by a BM biopsy. The 1° endpoint was the number of clinically significant bleeding events (CSBE, grade ≥2 per modified WHO scale); other endpoints included protocol-defined platelet transfusion events (PTE), platelet response per IWG 2006 (HI-P), survival, and safety, including progression to AML, defined conservatively as: 1) ≥20% blasts in the BM or peripheral blood after 4 wk off romiplostim, or 2) pathology consistent with leukemia (eg, chloroma or leukemia cutis), or 3) anti-leukemic treatment initiation.
Of 250 pts enrolled (romiplostim 167, PBO 83): 59% were male, median (Q1, Q3) age was 70 (61, 77) year, WHO classes RCMD (68%), RAEB-1 (13%), MDS-U (11%), RA (4.4%), RCMD-RS (2.4%), RARS (0.8%), RAEB-2 (1%), IPSS status low (25%), int-1 (71%), int-2 (0.4%), and IPSS cytogenetics good (78%), intermediate (18%), poor (1.6%). There were more pts who were RAEB-1 (14% vs 11%) and RAEB-2 (1% vs 0%) and fewer who were MDS-U (10% vs 15%) with romiplostim (all NS). Due to DMC concerns regarding the potential for transient increases in blast cell counts and the risk for progression to or treatment for AML, study drug was discontinued in Feb 2011, affecting 28% of pts. Also leading to withdrawal were consent withdrawn (romiplostim 13%, PBO 15%), adverse events (AE) (12%, 5%), and alternative therapy (7%, 11%). The mean number of CSBE/pt was romiplostim 1.47, PBO 1.94 (HR 0.83, 95% CI: 0.66, 1.05, p = 0.13); rates were romiplostim 18.6%, PBO 26.5%. The overall number of bleeding events was reduced with romiplostim (RR 0.92, 95% CI: 0.86, 0.99, p = 0.026). PTE rates/100 pt-year were romiplostim 748.9, PBO 1013.5 (RR 0.77, 95% CI: 0.66, 0.88, p<0.001). HI-P rates were romiplostim 36.5% (61 pts), PBO 3.6% (3 pts) (OR 15.6, 95% CI: 4.7, 51.8, p<0.001). From wk 4 on, median platelets with romiplostim were consistently higher than with PBO (p<0.001). The overall 1-year K-M survival was romiplostim 80%, PBO 78% (HR 1.03, 95% CI: 0.54, 1.95) (Figure 1), with 28 deaths (17%), none (0%) hemorrhagic and 5 (3%) from AML and MDS disease progression, with romiplostim and 14 deaths (17%), including 4 (4.8%) hemorrhagic and 3 (3.6%) from AML and MDS disease progression, with PBO.
Median time on romiplostim was 21.5 wk (range: 1, 50). SAE rates were romiplostim 40%, PBO 27%; those frequent (≥5%) SAE occurring ≥2x more with romiplostim were pneumonia, pyrexia, thrombocytopenia, and atrial fibrillation; those occurring ≥2x more with PBO were diarrhea, dyspnea, and cerebral hemorrhage. Peripheral blast increases are described below (Table).
| On treatment | End of study | |||
|---|---|---|---|---|
| Peripheral blasts >10% | Blasts ↓to ≤10% | Persistently elevated blasts confirmed | No follow-up available | |
| Romiplostim | 25 (15%) | 14 (8.4%) | 2 (1.2%) | 9 (5.4%) |
| PBO | 3 (3.6%) | 2 (2.4%) | 0 (0%) | 1 (1.2%) |
| On treatment | End of study | |||
|---|---|---|---|---|
| Peripheral blasts >10% | Blasts ↓to ≤10% | Persistently elevated blasts confirmed | No follow-up available | |
| Romiplostim | 25 (15%) | 14 (8.4%) | 2 (1.2%) | 9 (5.4%) |
| PBO | 3 (3.6%) | 2 (2.4%) | 0 (0%) | 1 (1.2%) |
AML rates through 58 wk were romiplostim 6.0%, PBO 2.4% (HR 2.51, 95% CI: 0.55, 11.47). Of the 13 AML cases, 9 (69%) were in pts who were initially RAEB-1 and 4 (31%) were diagnosed by anti-leukemic therapy initiation, which could include hypomethylating agents. Of pts who were RAEB-1 at baseline, 2/9 (22%) PBO pts developed AML vs 7/24 (29%) of romiplostim pts. AML-free survival rates were similar (HR 1.13, 95% CI: 0.60, 2.13).
Romiplostim treatment in low/int-1 MDS pts resulted in a 15-fold increase in achieving HI-P. Although there were more platelet transfusions with PBO (p <0.001), there still was a trend for more clinically significant bleeding events with PBO (p = 0.13) than romiplostim. The AE profile of romiplostim was generally comparable with PBO, with no hemorrhagic deaths with romiplostim. Increases in peripheral blasts >10% occurred more frequently with romiplostim but generally resolved after romiplostim discontinuation. AML was defined conservatively; cases are pending central pathology review. AML occurred primarily in pts who were initially RAEB-1 and in more pts with romiplostim. Overall and AML-free survival rates were similar.
Giagounidis:Amgen: Consultancy; GlaxoSmithKline: Consultancy. Off Label Use: This trial examined the use of romiplostim, which is indicated for use in ITP, in MDS. Mufti:Celgene: Consultancy, Research Funding. Kantarjian:Amgen: Research Funding. Fenaux:Celgene: Honoraria, Research Funding; Amgen: Honoraria; Roche: Research Funding; Janssen Cilag: Research Funding. Sekeres:Celgene Corp: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees. Kuendgen:Celgene: Honoraria. Platzbecker:Amgen: Honoraria; GSK: Honoraria. Gaidano:Amgen: Honoraria. Jedrzejczak:Amgen: Membership on an entity's Board of Directors or advisory committees, Research Funding. Hu:Amgen: Employment, Equity Ownership. Yang:Amgen: Employment, Equity Ownership. Jun:Amgen: Employment, Equity Ownership.
