Gene Copy Number Alterations Identify Subsets of Mycosis Fungoides/Sézary Syndrome Patients with Worse Survival Outcomes

Background:
Mycosis Fungoides/ Sézary syndrome (MF/SS) are primary cutaneous T-cell lymphomas (CTCL)s characterized by cutaneous involvement by clonal CD4+ mature T cells. With advanced disease, many patients develop circulating tumor cells. Even low levels of blood involvement (<250 clonal tumor cells/ul) have been associated with poor prognosis. While many molecular aberrations have been identified in MF/SS, prognostication based on molecular alterations including gene copy number alterations (GCNAs) by fluorescence in situ hybridization (FISH) has not been well established. We investigated the impact of GCNAs in circulating tumor cells and clinical features on survival outcomes.

Methods:
We reviewed MF/SS patients diagnosed from 2012 to 2024 and identified 125 patients with blood involvement who had molecular studies. A validated FISH panel using 11 FISH probes designed to capture 97.5% of GCNAs in patients with CTCL and SS was used. The panel included probes for TP53, MYC, RB1, CDKN2A, ATM, STAT3, STAT5B, ARID1A, ZEB1, FAS, CARD11, and DNMT3A genes. Overall survival (OS) was estimated using the Kaplan-Meier method and log-rank test was used to compare survival distributions between different groups. Univariate Cox proportional hazards (CPH) regression identified variables associated with OS, and variables with a p-value < 0.5 were included in the multivariate CPH model to identify predictors of survival. Only patients with complete data were included in the analysis.

Results:
The median age of patients was 74 (IQR:67-81) years (68% male, 32% female). Overall, 49% were alive at the time of analysis. Clonality by Vβ and T-cell receptor (TCR) rearrangement by PCR were identified in 70% and 84% of the patients, respectively. The median absolute lymphocyte count (ALC) was 1.7(IQR: 0.9-3.6). A total of 85 patients had FISH performed at time of diagnosis. Among these, 54% showed GCNAs. Deletions were identified in genes including ATM (11q22.3) in 14%, DNMT3A (2p23) in 17%, TP53 (17p13.1) in 40%, ZEB1 (10p11.2) in 18%, RB1 (11q14.2) in 13%, ARID1A (1p35.3) in 13%, and CDKN2A (9p21.3) in 15% of all patients. Genes showing amplifications were STAT3 (17q21.31) in 26%, MYC (8q24.21) in 37%, and CARD11 (7p22) in 11% of all patients.

The median OS of all patients was 54 months (95% CI: 33-84). Univariate Cox regression analysis identified several significant prognostic factors for OS. Older age (HR=1.02, p=0.05) was significantly associated with worse OS. Elevated ALC (hazard ratios [HR] = 1.1, p < 0.001), elevated WBC (HR=1.02, p < 0.001), elevated LDH (HR = 1.001, p < 0.000) and elevated ANC (HR = 1.2, p = 0.016) were significantly associated with worse OS. Conversely, higher Hgb (HR = 0.7, p < 0.001) levels were associated with better OS. SS stage (HR=1.4, p = 0.02), presence of TCR rearrangement (HR=1.55, p = 0.005) and tumor cell count (HR = 1, p<0.001) were also significant predictors of OS. The presence of ARID1A deletion (HR = 3.3, p = 0.001) and ZEB1 deletion (HR=2.05, p=0.04) were strongly associated with poorer OS. Amplifications of STAT3 (HR= 1.4, p=0.3), MYC (HR= 1.2, p=0.5), or CARD11 (HR= 1.3, p=0.6) were not significantly associated with poorer OS.

Furthermore, a total of 78 patients had full data and were included in the multivariate CPH model. Among GCNAs, ZEB1 deletion (HR = 5.8, p = 0.008), ARID1A deletion (HR = 5.7, p = 0.009), and ATM deletion (HR = 5.5, p = 0.01) were significantly associated with poorer OS. Clinical variables including male gender (HR=7.3, p<0.001), higher ANC (HR=1.3, p = 0.05) and lower Hgb (HR= 0.6, p <0.001) were all significant predictors. The median OS for patients with ZEB1 deletion was (34 months, 95%CI 14-NA) compared with wild-type ZEB1 (86 months, 46-NA). Patients with ATM deletion (39 months, 20-NA) and AIRD1A deletion (21 months, 19-NA) also had significantly lower OS compared to wild type cases (93 months and 94 months, respectively).

Summary:

Herein, we identified novel molecular prognostic markers based on a validated FISH panel in circulating tumor cells. In conjunction with clinical data, our results help identify high risk patients with specific GCNAs who may benefit from a more intensive and novel treatment approaches, potentially improving survival outcomes. Future research should aim to validate these findings in larger cohorts.