Abstract
Abstract 2691
Pre-treatment cytogenetics are utilized for risk-stratification in many hematologic malignancies but not in MCL, where improved risk-stratification is needed due to the heterogeneity of outcomes. A review at our institution found a significant association of CK with median time to relapse (1005 days for CK vs. not reached in those without CK (NCK), p=0.008) in pts undergoing autologous transplantation in first remission (Cohen, BMT Tandem Mtgs 2012). To evaluate the impact of CK at diagnosis on PFS and overall survival (OS) regardless of initial therapy or eligibility for transplant, we reviewed all pts with newly diagnosed MCL at the Ohio State University (OSU) from 2000–2011.
We included all pts with confirmed MCL and available bone marrow (BM, n = 68) or peripheral blood (n = 12) pretreatment cytogenetics. Unstimulated cytogenetic analysis, standardly performed at OSU from 2000–2008, was performed in 48 pts, and stimulated analysis utilizing CpG oligonucleotides was performed in 32 pts beginning in Jan 2008. CK was defined as ≥3 unrelated chromosomal abnormalities, including t(11;14). Associations between CK and clinical variables were performed using Fisher's Exact and Wilcoxon rank sum tests. PFS and OS curves were constructed from date of pathologic diagnosis until date of relapse or death by the Kaplan-Meier method and evaluated using the log-rank test. A multivariable proportional hazards model for PFS was constructed using a limited backwards selection procedure.
In 80 untreated pts, median age was 63 (range 37–81), 70% were male, 28% had bulky disease ≥5cm, and 95% had stage IV disease. Sixty-five of 77 evaluable pts (84%) had BM involvement. Median time from diagnosis to pretreatment cytogenetic analysis was 21 days (range:−20 to 1094). CK was detected in 32 pts (40%, 95% CI: 0.29 to 0.52), and CK rate was higher in stimulated versus unstimulated analyses (56% vs 29%, p=0.02). CK was associated with several adverse markers, including a higher MCL international prognostic index (MIPI; median 6.3 vs 5.9, p=0.003) and simplified MIPI scores (median 6 vs 4, p=0.0006), increased percent involvement of BM cellularity (68% vs 10%, p=0.005), splenomegaly (73% vs 38%, p=0.005), and B-symptoms (57% vs 26%, p=0.01). CK was not associated with age (p=0.98) or bulky disease ≥5cm (p=0.80). Thirty-six pts (45%) underwent transplant in first remission, 35 autologous and 1 allogeneic. Intensive induction included R-HyperCVAD (n=10), methotrexate with augmented R-CHOP (n=31), and R-CHOP (n=17); less intensive induction included R-Bendamustine (n=4) and other non-anthracycline containing therapies. There were no large differences between CK and NCK patients with respect to receipt of intensive induction (76% vs 75%) or transplant in first remission (44% vs. 46%). With a median follow-up of 24 months for PFS and 32 months for OS, the estimated PFS at 2 years for CK and NCK pts was 47% and 71%, respectively (p=0.04; Figure 1A), and the estimated OS was 57% and 84% (p=0.02). For pts with unstimulated evaluation, the association of CK with shortened PFS and OS remained significant (PFS, p=0.02 and OS, p=0.01), with no association of method of analysis with PFS (p=0.35). In a multivariable analysis, high-risk (≥6.2) vs. low-risk (<5.7) MIPI (p=0.008) and bulky disease ≥5cm (p=0.03) remained independently prognostic for shortened PFS and transplant in first remission was protective (p=0.06). In this series, CK (p=0.33),splenomegaly, and B-symptoms at diagnosis did not provide additional prognostic information independent of the other variables in the model. Six pts with CK had del(17p), while del(17p) was not present in the NCK group. Del(17p) predicted a particularly poor PFS. PFS was significantly worse among pts with CK and del(17p) than pts with CK without del(17p) (p=0.02; Figure 1B).
Pre-treatment detection of a CK, even without stimulation, is prognostic for shortened PFS and OS in MCL and should be evaluated in larger, prospective series with uniformly treated patient populations. The presence of del(17p), detected in only 8% of pts, appears to predict a particularly poor PFS and should also be prospectively evaluated.
PFS in pts with previously untreated MCL and CK vs. NCK
PFS in pts with previously untreated MCL with NCK, CK without del(17p), and CK with del(17p).
PFS in pts with previously untreated MCL with NCK, CK without del(17p), and CK with del(17p).
No relevant conflicts of interest to declare.
