Chloe AL Thompson-Peach, Daniel Thomas, Mara Dottore, Kelly Lim, Frank C. Stomski, Romana Panagopoulos, Johannes Foßelteder, Andreas Reinisch, Vashe Chandrakanthan, Sandra Jenkner, Luke Bland, Sharon Paton, Stan Gronthos, Timothy R. Hercus, Hannah R. Wardill, Winnie L. Kan, Maha M Kamel, Paul A.B. Moretti, Stuart M. Pitson, Kate Burbury, David M. Ross, Pramod C. Nair, Angel F. López, Denis Tvorogov
ABSTRACT: Somatic frameshift mutations in the gene encoding calreticulin (CALR) give rise to myelofibrosis and are classified as type 1 (del52) or type 2 (ins5) according to the degree of wild-type sequence retained adjacent to the neopeptide, with each type conferring different clinical outcomes. Targeting strategies specific for type 1 vs type 2 mutations would have enormous clinical utility in the treatment and prevention of myelofibrosis as responses to tyrosine kinase inhibitors are not durable nor mutation specific. Here, we show that dual targeting of type 1 (del52) mutant CALR with 2 monoclonal antibodies directed against distinct epitopes in CALR have significant advantages compared with single-agent treatment in the eradication of primary megakaryocyte progenitors in vitro and in a humanized ossicle microenvironment leading to improved survival in xenograft models. Dual targeting was superior in blocking constitutive STAT5 and extracellular signal-regulated kinase phosphorylation induced by del52 and prevented accumulation of Janus kinase 2 (JAK2) phosphorylation, overcoming ruxolitinib resistance. In contrast, type 2 mutations showed increased CALR dimerization and were partially resistant to antibody targeting but could be affected by a ruxolitinib triple combination. Together, our data demonstrate an ultraprecision medicine approach tailored to either type 1 or type 2 mutation classes will be required for maximal efficacy and complete blockade of JAK/STAT signaling, with far-reaching implications for patient management.