Abstract 1014

Poster Board I-36

A database was created of 305 women with leukemic breast (Br) or ovarian (Ov) tumors, reported 1970-2009 (35% since 2000), to analyze their clinical behavior and reasons why survival remains poor. Follow-up was elicited from 96 authors. Distribution of cases and 2-year DFS are shown.

BreastOvaryBreast+Ovary
Presentation of Br or Ov tumor AMLn=129 ALLn= 59 AMLn=53 ALLn=56 AMLn=6 ALLn=2 
GS 11 of 35 3 of 22 0 of 2 
At marrow dx 2 of 30 2 of 17 4 of 18 0 of 3 1 of 2 
After chemo 0 of 34 2 of 23 1 of 8 11 of 50 1 of 3 
After SCT 1 of 30 4 of 19 1 of 5 1 of 3 1 of 1 
BreastOvaryBreast+Ovary
Presentation of Br or Ov tumor AMLn=129 ALLn= 59 AMLn=53 ALLn=56 AMLn=6 ALLn=2 
GS 11 of 35 3 of 22 0 of 2 
At marrow dx 2 of 30 2 of 17 4 of 18 0 of 3 1 of 2 
After chemo 0 of 34 2 of 23 1 of 8 11 of 50 1 of 3 
After SCT 1 of 30 4 of 19 1 of 5 1 of 3 1 of 1 

There is no common feature (FAB, karyotype, wbc) except age; 90% were under 51 (range:3mo-67yr). Br and Ov tumors were generally the first site of extramedullary (EM) leukemia, but 21 had prior or concurrent CNS involvement. The median time of Br and Ov relapse after diagnosis was 11-13 mo in AML and in Br cases with ALL, but for ALL Ov cases, the median was 44 mo. Late relapse at 5 to 13 yrs was reported in 24 cases (7 Br, 17 Ov). Median survival of 59 cases of granulocytic sarcoma (GS), before marrow leukemia, was 30mo for Br and 13mo for Ov cases. For the 71 Br or Ov relapses after chemo without marrow relapse, median survivals were 12mo (AML) and 17mo (ALL). When marrow was simultaneously involved in 115, median survivals were <8mo (AML) and <12mo (ALL). Of 60 cases of Br or Ov relapse after SCT, marrow had not relapsed in 36 cases with median survival of 16-22mo; when marrow was involved in 24, median survival was 4mo. Failure to control primary and subsequent EM disease was the principal obstacle to cure, as tumor response to chemotherapy was asynchronous with marrow response, and rarely complete or documented serially. Br tumors were single or multiple, axillary nodes were often involved, and recurrence in ipsi- or contralateral breasts was seen in at least 14%. Ovarian tumors grew rapidly to large size (median 10cm) before symptoms and tended to invade contiguous organs and to involve the contralateral ovary, so excision was difficult and recurrence common. After treatment of Br and Ov tumors, further relapse in EM soft tissue sites was almost twice as common as marrow relapse. The most common subsequent site was CNS in both AML (14 cases) and ALL (23 cases), despite universal prophylaxis in ALL. The 46 cases disease-free at 2 yrs provide a focus for analyzing successful initial treatment. In addition to systemic chemo, the majority received local therapy (tumor excision and/or RT); in at least 7 cases, local followed systemic therapy. Nine underwent SCT. Only 12 long survivors had been in simultaneous marrow relapse (5 Br, 6 Ov, 1Br/Ov); 2 with ALL subsequently relapsed in CNS with marrow. Of the 34 pts with isolated Br or Ov tumors disease-free at 2yr, 13 subsequently relapsed, predominantly in EM sites (8 EM, 1 EM+BM, 4BM). Twenty-six pts (11ALL, 15AML) have been documented disease-free at 5 to 31 yr. Six additional patients who relapsed and responded to salvage therapies survived 5 to 26 yr. Neither age nor karyotype predicted DFS; survivors were ages 5mo to 55yr at dx, and 13 had abnormal karyotypes, only 2 were 'favorable'. This study documents the behavior and drug resistance of leukemic tumors in breast and ovary, the greater risk for subsequent extramedullary than marrow relapse, an association with resistant CNS disease, and short survival, but also that lengthy survival and cure are possible. Current systemic agents alone are inadequate to cure most EM tumors. Local and aggressive systemic therapy, promptly initiated, with reliable, probably intraventricular, CNS treatment and long-term monitoring for further extramedullary involvement should improve prognosis until more effective agents are available.

Disclosures:

No relevant conflicts of interest to declare.

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution