Isolated Extramedullary (EMD) Blastic Transformation of Chronic Myeloid Leukemia (CML-BP) in the TKI Era: Characteristics, Treatments and Outcomes

Introduction: CML-BP is defined by an increase of blasts in peripheral blood (PB) or bone marrow (BM) (>20-30%) or extramedullary disease (EMD). Occasionally, EMD presents without evidence of transformation in BM. Information on the characteristics, distribution, treatment and outcomes of pts with isolated EMD CML-BP is sparse. We report our experience in 50 pts with isolated EMD CML-BP.

Methods: We identified 128 pts diagnosed with EMD CML-BP who were treated with a TKI at some point in the course of CML. These patients were diagnosed with EMD between 2/2000 and 6/2015. Isolated EMD CML-BP (i.e., no increase blasts in PB or BM) was observed in 50 (39%) pts and 78 (61%) pts had both EMD and BM involvement (EMD+BM). All pts were enrolled in prospective therapeutic clinical trials with different treatment modalities. Clinical characteristics were reviewed at the time of initial diagnosis of EMD. Failure free survival (FFS; failure defined as disease progression, death, non-responder, discontinuation of first line therapy for any reason) and overall survival (OS) were defined from the time of initial diagnosis of EMD to the time of treatment failure/death, respectively.

Results: The median age in 50 pts with isolated EMD was 49 years (range, 22 to 75); 74% were male and 62% were Caucasian. Of the 50 pts with isolated EMD, 16 (32%) presented as CML-BP at the initial diagnosis of CML (11 with isolated EMD CML-BP and 5 with BM CML-BP; these 5 pts later relapsed with isolated EMD CML-BP after a median of 19.4 mo). The other 34 pts had been diagnosed with CML-CP (n=30) or CML-AP (n=4) with a median of 39.5 mo (range 2 to 222) prior to transformation to EMD CML-BP. Of the 30 pts in CP, disease status in BM or peripheral blood at time of isolated EMD was CHR in 10 pts (33%), major CyR in 4 (13%), CCyR in 1 (3%), MMR in 7 (23%) (including 3 with MR4.5). Twenty-two pts (44%) had TKI prior to the diagnosis of isolated EMD. Immunophenotype of blasts in isolated EMD was myeloid in 87% and lymphoid in 13%. At the time of first diagnosis, isolated EMD involved central nervous system in 19 pts (38%), soft tissues 5 (10%), bones 7 (14%), skin 5 (10%), lymph nodes 2 (4%), small bowel 2 (6%), and testis or pleura (1 each). Eight pts (16%) presented with isolated EMD at ≥2 sites. BCR-ABL mutation testing was performed in 8 pts at time of isolated EMD (performed on bone marrow and/or peripheral blood samples only), and mutations were identified in 3 pts: T315I, T315L and G250E. Forty-seven (94%) pts received treatment for isolated EMD (1 pt kept under observation and 2 pts died before starting treatment; one from pneumonia, another from unknown cause). Treatment information for initial isolated EMD was available in 47 pts, 33 pts treated with TKI (8 alone, 6 with systemic chemotherapy, 6 with local therapy (XRT and/or surgery) and 19 with both systemic and local therapy). Fourteen pts treated without TKI (5 with systemic chemotherapy alone and 9 with only local and/or systemic therapy). Response to first line treatment for isolated EMD was complete remission in 29 (62%), partial remission in 3 (6%), progressive disease in 12 (25%), and 3 pts lost to follow up (6%). Overall, the median follow up for all 128 pts with EMD was 90 mo (63 mo for those with isolated EMD). Overall survival (OS) of pts with isolated EMD was significantly longer compared to those with non-isolated EMD (n=73; p=0.002, Figure-1A). OS according to the sites involved in pts with isolated EMD showed that pts with ≥2 sites involved at initial presentation of isolated EMD had a trend for inferior outcome (5 year OS 12% vs 20% in skin alone, 25% in CNS alone, 36% in others; Figure-1B). The median FFS in pts with isolated EMD was 6 months. Thirteen pts (28%) relapsed: 2 with BM and extramedullary relapse (3 and 6 months after first isolated EMD, 3 in the same EMD site (63, 36 and 8 months after isolated EMD), and 8 pts at a site different from original EMD site (at a median of 8.5 months after initial isolated EMD. Three pts had 2 relapses. At the time of last follow up, 13 pts were alive and 37 pts had died. Causes of death in pts with isolated EMD were CML progression (n=26) and unknown (n=11).

Conclusions: Isolated EMDduring the course of CML is uncommon. Pts with isolated EMD CML-BP have poor outcome but better than pts who have concordant BM involvement with CML-BP. Further studies to understand the biologic features of EMD CML-BP are ongoing.

View largeDownload slide

View largeDownload slide

Close modal