Preclinical Analyses Of The Chemical JAK2 Inhibitor, SAR302503, In Classical Hodgkin Lymphoma and Primary Mediastinal Large B-Cell Lymphoma

Classical Hodgkin lymphoma (cHL) and primary mediastinal large B-cell lymphoma (MLBCL) are diseases which share certain clinical, pathologic and genetic features. We previously characterized chromosome 9p24 amplification as a disease-specific structural alteration in cHL and MLBCL and identified the immunoregulatory genes, PD-L1 and PD-L2, and JAK2 as key targets of the 9p24 amplification. In a panel of informative cHL and MLBCL cell lines with defined 9p24 copy number, JAK2 amplification increased JAK2 protein expression and activity and enhanced sensitivity to chemical JAK2 inhibition with commercially available tool compounds. Given the importance of JAK/STAT signaling as a survival pathway in primary cHL and MLBCL and the pathway’s additional role in augmenting PD-1 ligand transcription, we postulated that JAK2 was a promising rational therapeutic target in cHLs and MLBCLs with 9p24 amplification.

SAR302503 (fedratinib), formerly TG101348, is a selective, potent and oral JAK2 inhibitor that has demonstrated clinical activity in myeloproliferative disorders with activating JAK2 mutations. For these reasons, we have evaluated the preclinical activity of SAR302503 in in vitro and in vivo model systems of cHL and MLBCL with known 9p24/JAK2 copy number. In in vitro analyses, SAR302503 inhibited the cellular proliferation of cHL and MLBCL cell lines and induced their apoptosis. Of note, there was an inverse correlation between 9p24/JAK2 copy number and the EC50 of SAR302503, indicating that cHLs and MLBCLs with JAK2 amplification exhibited increased sensitivity to chemical JAK2 inhibition. In this series of cHLs and MLBCLs of defined JAK2 copy number, JAK2 copy gain was associated with higher baseline phosphorylated JAK2 (pJAK2) and increased abundance of phosphorylated STAT family members including pSTAT1, pSTAT3 and pSTAT6. Consistent with the copy number-dependent anti-proliferative effects of SAR302503, the compound decreased pJAK2, pSTAT1, 3 and 6 in a copy-number dependent manner in cHLs and MLBCLs. These effects were apparent within 2 hours of SAR302503 treatment.

We next developed a comprehensive phosphoJAK/STAT immunohistochemical signature to assess baseline pathway activity and sensitivity to targeted chemical JAK2 inhibition. In cHL cell lines with 9p24 amplification, SAR302503 treatment abrogated pJAK2, pSTAT1 and pSTAT3 immunohistochemical staining at early and late timepoints. Furthermore, chemical JAK2 inhibition decreased PD-L1 transcript abundance in 9p24-amplified cHL and MLBCL cell lines and reduced the abundance of downstream JAK2 targets including MYC and PIM1 in cHLs with high JAK2 copy number.

After demonstrating the activity and specificity of SAR302503 in in vitro assays, we evaluated the JAK2 inhibitor in murine xenograft models of cHL and MLBCL with 9p24/JAK2 amplification (Karpas 1106 [MLBCL] and HDLM2 [cHL]). In systemic MLBCL (Karpas 1106) and subcutaneous cHL (HDLM2) xenograft models, tumor growth was monitored via bioluminescent imaging (Karpas 1106) and mass measurements (HDLM2) and SAR302503 treatment was begun following the establishment of ≅ 100 mm³ tumors. In both models, pSTAT3 was analyzed as a pharmacodynamic marker following 5 days of treatment and SAR302503-treated animals had markedly decreased tumor cell pSTAT3 expression. In the Karpas 1106 MLBCL murine model, SAR302503 treatment significantly prolonged overall survival (p = .0002). In the HDLM2 cHL xenograft model, the JAK2 inhibitor significantly decreased subcutaneous tumor growth (p< .0001) and pSTAT3 expression in tumor cells (p = .0004). Transcriptional profiling confirmed that the HDLM2 tumors from SAR302503-treated animals exhibited coordinately decreased expression of STAT3 targets and downregulation of a functionally validated JAK2 gene set. Taken together, these data indicate that SAR302503 specifically decreases cHL and MLBCL growth in a 9p24/JAK2 copy number-dependent manner in vitro and in vivo and highlight the utility of pSTAT3 immunohistochemistry as a biomarker of pathway responsiveness. Clinical evaluation of JAK2 pathway inhibition in patients with cHL and MLBCL with known 9p24/JAK2 copy number status is warranted.