Juvenile Myelomonocytic Leukemia (JMML) is a rare childhood myelodysplastic / myeloproliferative overlap disorder. JMML exhibits myeloid populations with mutations in Ras-Erk signaling genes, most commonly PTPN11, which confer growth hypersensitivity to GM-CSF. While allogeneic hematopoietic stem cell transplant (HSCT) is the treatment of choice for children with JMML, 50% of children succumb to leukemia relapse; however, the mechanism leading to this high relapse rate is unknown. We hypothesized that the hyperinflammatory nature of JMML may damage the bone marrow microenvironment, leading to poor engraftment of normal donor cells following transplant, permitting residual leukemia cells to outcompete the normal graft, and thus promoting leukemia relapse.

Using Vav1 promoter-directed Cre, we generated a mouse model of JMML that conditionally expresses gain-of-function PTPN11D61Yin utero during development. While PTPN11D61Y/+; VavCre+embryos did not demonstrate in utero lethality, we observed a modest reduction of PTPN11D61Y/+; VavCre+ mice at the time of weaning compared to predicted Mendelian frequencies. Further, surviving PTPN11D61Y/+; VavCre+ mice developed elevated peripheral blood leukocytosis and monocytosis as early as 4 weeks of age compared to PTPN11+/+; VavCre+ controls.

To address the hypothesis that an aberrant bone marrow microenvironment in the PTPN11D61Y/+ mice leads to poor engraftment of wild-type donor cells following transplant, we examined engraftment of wild-type hematopoietic stem and progenitor cells (HSPCs) in the PTPN11D61Y/+; VavCre+ mice and monitored animals for disease relapse.

16-24 week-old diseased PTPN11D61Y/+; VavCre+ and control PTPN11+/+; VavCre+ mice were lethally irradiated (11 Gy split dose) and transplanted with 5 x 105 CD45.1+ wild-type bone marrow low density mononuclear cells (LDMNCs), which simulates a limiting stem cell dose commonly available in a human HSCT setting. 6 weeks post-HSCT, PTPN11D61Y/+; VavCre+recipients demonstrated an unexpected elevated CD45.1+ donor cell contribution in peripheral blood compared to the control PTPN11+/+; VavCre+ recipients. However, despite superior engraftment in the PTPN11D61Y/+; VavCre+ recipients, these mice had a significantly shorter median survival post-HSCT due to a resurgence of recipient CD45.2-derived leukemic cells.

We repeated the experiment using a high dose of CD45.1+ LDMNCs (10 x 106 cells) to determine if providing a saturating dose wild-type cells could prevent the relapse of recipient-derived leukemogenesis and normalize the survival of the PTPN11D61Y/+; VavCre+recipients. While this saturating dose of wild-type cells resulted in high peripheral blood chimerism in both the PTPN11D61Y/+; VavCre+ and PTPN11+/+; VavCre+ recipients, the PTPN11D61Y/+; VavCre+ animals nevertheless demonstrated significantly reduced overall survival.

When we examined the cause of mortality in the HSCT-treated PTPN11D61Y/+; VavCre+mice, we found enlarged spleens, hypercellular bone marrow, and enlarged thymuses. Flow cytometry revealed that the majority of cells in the peripheral blood, bone marrow, and spleen were recipient-derived CD45.2+ CD4+ CD8+ T cells. To verify that the disease was neoplastic in origin, secondary transplants into CD45.1/.2 recipients were performed from two independent primary PTPN11D61Y/+; VavCre+and two independent primary PTPN11+/+; VavCre+ controls. Secondary recipients of bone marrow from PTPN11D61Y/+; VavCre+ animals rapidly succumbed to a CD45.2-derived T-cell acute lymphoid leukemia (T-ALL).

Previous studies demonstrated that wild-type PTPN11 is needed to protect the integrity of the genome by regulating Polo-like kinase 1 (Plk1) during the mitosis of the cell cycle (Liu et al., PNAS, 2016). We now demonstrate that even when PTPN11 mutant animals are provided with saturating doses of wild-type HSCs, dysregulated residual recipient cells are able to produce relapsed disease. Collectively, these studies highlight the propensity of residual mutant PTPN11 cells to transform after being subjected to mutagenic agents that are commonly used for conditioning regimens prior to allogeneic HSCT. These findings suggest that modified pre-HSCT conditioning regimens bearing reduced mutagenicity while maintaining adequate cytoreductive efficacy may yield lower post-HSCT leukemia relapse in children with PTPN11mutations.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution