Expression of p-STAT5, c-MYC and IKZF1 By Immunohistochemistry May Differentially Correlate with Clinical Outcome in Adult B- Lymphoblastic Leukemia (B-ALL) According to BCR-ABL Status

Background: Molecular and cytogenetic abnormalities characterize the biologic heterogeneity of B-ALL and provide a strong and independent prognostic factor for treatment outcome. Emerging data highlight the role of tyrosine kinase inhibition (TKI) in those with Philadelphia chromosome positive (Ph+) B-cell acute lymphoblastic leukemia. A new high risk subset, so-called Ph-like ALL, has been recognized, comprising 10-15% of pediatric B-ALL. These cases show similar alterations in IKZF1, and are likely driven by alternative kinase rearrangements, including CRLF2, JAK2, ABL1, and others. Increased STAT5 phosphorylation may broadly classify such patients, and, also suggest a benefit with JAK2 inhibitors or TKI’s. Finally, c-MYC has been recognized as a high risk feature for both B- and T-ALL, though it is unclear whether the risk is similar among Ph+ and Ph- patients in the era of TKI-based therapy.

Method and Materials: A retrospective analysis was performed of a series of 75 patients at Moffitt Cancer Center with B-ALL diagnosed between June 2002 and July 2013. Of them, paraffin embedded blocks from patients who had the first diagnostic bone marrow biopsy performed at MCC with a good biopsy quality (> 1.0 cm in length and tumor load over 60% of marrow cellularity) were selected for performing immunohistochemical (IHC) study. Ten normal bone marrows were also included as antibody base-line control. IHC stains were performed according to institutional standard protocol and modified based on the individually purchased antibody. Expression of pSTAT5, c-MYC, and IKZF1 on lymphoblasts was assessed semiquantitatively based on the expression score calculated with staining strength and multiplied by percentage positivity on tumor cells. Clinical outcomes were correlated to the IHC scores. Hazard ratios were generated using standard cox regression analysis. Patient survival was analyzed with Kaplan-Meier method from the date of diagnosis until death from any cause.

Results: From 75 total patients, 43 were selected with de novo B-ALL with marrow evaluation at the time of diagnosis and a minimum of 60% blasts on core/clot biopsy. 18 (42%) were Ph+ with median age 52y (18-76), median wbc 14.7 k/uL (0.7-580), and male to female ratio of 1.5. 95% received an intensive induction (HyperCVAD, CALGB8811, ECOG2993, or pediatric regimen), and 94% of Ph+ patients received TKI therapy, either with imatinib (67%) or dasatinib (33%). 35% received an allogeneic transplant in first remission.

No significant difference in age, gender, application of intensive therapy, PFS/relapse rate, or OS/death rate. There was trend towards higher wbc (p=0.11) and application of AlloSCT (p=0.11) in the Ph+ group.

Using the mean as a cutoff, the IHC staining results showed nuclear expression of pSTAT5 to be more frequently present in Ph+ B-ALL than in Ph- B-ALL (78% vs 32%, p = 0.005), however, no differences in staining were observed with IKZF1 (44% vs 52%, p=0.76), or c-MYC (50% vs 64%, p=0.53). Similar data were generated using absolute staining intensity, with Ph+ B-ALL showing a higher median pSTAT5 score (p<0.001).

Neither c-MYC, IKZF1, or pSTAT5 were associated with OS when analyzing the entire cohort. For PFS, only pSTAT5 was significant (p=0.03), with no impact of IKZF1, or c-MYC.

Analysis of the Ph- group demonstrated a decrease in PFS (p=.009) and a trend for decreased OS (p=0.125) in those with elevated pSTAT5, possibly identifying this group as ‘ph-like’ ALL. Decreased IKZF1 was not associated with PFS, though, was associated with decreased OS (p=0.031). There was a trend for worse PFS in those with MYC overexpression, with 100% of MYC low patients in sustained remission (p=0.196). This did not translate to an impact on survival, possibly related to treatment related toxicity.

Analysis of the Ph+ group demonstrated no impact of pSTAT5, IKZF1, or c-MYC on OS or PFS, suggesting that TKI therapy was able to partially abrogate these risks.

Conclusion: Ph+ ALL patients show higher baseline pSTAT5 expression by IHC than Ph- patients. However, those Ph- patients with elevated pSTAT5 show impaired PFS and a trend for worse OS, suggesting that Ph-/pSTAT5 high patients may warrant additional testing to rule out alternative, and potentially actionable, kinase translocations (Ph-like ALL). The impact of low IKZF1 and c-MYC, when assessed by IHC was limited to Ph- patients.