Abstract 4326

Background:

Extramedullary (EM) disease is a well-known manifestation of acute myeloid leukemia (AML). Despite its recognized incidence, little is known about organ-specific EM-AML, including genitourinary (GU) AML. The purpose of this study is to identify the patients (pts) who develop GU-AML and to characterize the clinicopathologic, cytogenetic, and molecular features of this population.

Methods:

A database of 2,181 consecutive patients who were diagnosed with AML and underwent induction therapy from 2000 to 2011 at M.D. Anderson Cancer Center was reviewed retrospectively. All pts with histologically proven EM-AML were included in this series. Clinicopathologic, cytogenetic, and molecular data were examined and statistically analyzed.

Results:

A total of 1,120 pts underwent at least one EM biopsy and 244 were diagnosed with EM-AML. Of these, 9 pts (6 females) demonstrated GU-AML (0.4% of total population, 3.7% of EM-AML pts). Furthermore, 3 GU-AML pts demonstrated additional EM-AML in non-GU sites. At AML dx, GU-AML pts demonstrated median bone marrow blasts of 35% (range 1–69%) and median peripheral blood blasts of 1% (range 0–46%). CBC included median WBC of 3.5 K/uL (range 1.6–21.0 K/uL), median Hgb level of 9.4 g/dL (range 8.0–14.3 g/dL), and median platelet count of 118 K/uL (range 28–206 K/uL). Median age of AML dx in GU-AML pts was 45 years (range 28–69 years) and was significantly younger than the median age of AML dx in all other non-GU pts (60 years, range 12–89 years, p=0.025, Student's t-test). A total of 78% of GU-AML dx were made before or at AML presentation and 89% of GU-AML dx were made within 3 months of AML presentation. A total of 67% of GU-AML pts demonstrated cytogenetic abnormalities. Cytogenetic features included inversion 16 (inv (16), 33%), trisomy 8 (33%), diploid (33%), trisomy 22 (22%) and complex (22%). For all pts with GU-AML, no molecular mutations were present in RAS (0/9), FLT3 (0/7), NPM1 (0/2) or JAK2 (0/2). CR was achieved by 78% of pts with GU-AML. The pts who did not achieve CR expired early in induction therapy (within 29 days) due to sepsis. Of the GU-AML pts with CR, CR duration was 50.7 months (95% CI 15.2–86.2 months). CR duration of GU-AML pts was significantly longer than that of EM-AML pts with no GU sites (18.0 months, 95% CI 14.1–22.0 months, p=0.03, Kaplan-Meier method). Overall survival (OS) for all GU-AML pts was 41.6 months (95% CI 12.7–70.5 months) and was statistically equal to OS of pts without GU-AML and to OS of EM-AML pts with no GU sites.

Table.

Clinicopathologic and cytogenetic features of GU-AML pts.

PtGenderFABAge at AML dxSite(s) of EM-AMLCytogeneticsResponse
M4eo 44 Ovary Inv(16), trisomy 22 CR 
N/A 59 Uterus, ovary, fallopian tube, ureter, parametrial soft tissue, and lymph nodes Diploid Induction death 
M4 46 Bladder, epididymis Inv(16), trisomy 22 CR 
M1 68 Labia Diploid CR 
M2 28 Labia, breast, nasopharynx, skin, chest soft tissue Complex, including trisomy 8 CR 
M4eo 45 Uterus, fallopian tube Inv(16), trisomy 8 CR 
M0 61 Pleural fluid, kidney Complex Induction death 
M0 35 Kidney Diploid CR 
M5b 34 Testicle, spermatic cord Trisomy 8 CR 
PtGenderFABAge at AML dxSite(s) of EM-AMLCytogeneticsResponse
M4eo 44 Ovary Inv(16), trisomy 22 CR 
N/A 59 Uterus, ovary, fallopian tube, ureter, parametrial soft tissue, and lymph nodes Diploid Induction death 
M4 46 Bladder, epididymis Inv(16), trisomy 22 CR 
M1 68 Labia Diploid CR 
M2 28 Labia, breast, nasopharynx, skin, chest soft tissue Complex, including trisomy 8 CR 
M4eo 45 Uterus, fallopian tube Inv(16), trisomy 8 CR 
M0 61 Pleural fluid, kidney Complex Induction death 
M0 35 Kidney Diploid CR 
M5b 34 Testicle, spermatic cord Trisomy 8 CR 
Conclusion:

GU-AML is a rare but noteworthy manifestation of AML that tends to be diagnosed before or at AML presentation. Pts with GU-AML developed AML at a significantly younger age by 15 years than pts without GU-AML (p=0.025). Most GU-AML pts demonstrated cytogenetic abnormalities but none demonstrated molecular mutations. The presence of GU-AML, rather than EM-AML in other sites, may contribute to extended duration of CR (p=0.03). However, despite this finding and other advantages such as majority achievement of CR and young age of AML dx, there was no statistical advantage in OS in pts with GU-AML compared to those pts without GU-AML or to pts with EM-AML in non-GU sites.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution