A Patient Derived Xenograft As a Preclinical Model for Monomorphic Epitheliotropic Intestinal T-Cell Lymphoma

Monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL) is a type of rare aggressive lymphoma accounting for 5.4% of primary intestinal peripheral T-cell lymphoma and 10-25% of all primary intestinal lymphoma. MEITL also has an extremely poor prognosis with the median overall survival of only seven months. No effective treatment or targeted therapies are currently available to manage this disease. Therefore, there is an urgent need to devise and establish an appropriate preclinical model for a better understanding of this disease and for new therapeutic strategies to be developed on it.

Here, we present the first MEITL patient derived xenograft (PDX) as both orthotropic intestinal and as a subcutaneous model. The histological analysis demonstrated high similarity in the overall immunomorphologic features between the primary tumour and PDX tumours. Importantly, all the PDX tumors carried the distinctive MEITL immunophenotype; CD3⁺CD4^-CD8⁺CD56⁺ and extensive nuclear expression of MATK. We also had Sanger sequenced 83 somatic mutations in the original tumour and rediscovered them in the 6^th passage of our xenografts, which suggest that the generated MEITL PDX tumours were genetically stable.

Moreover, whole exome sequencing results showed that the clonal architecture in the primary tumour was well preserved in the PDX tumours. The clonal architecture was further analysed with single cell whole genome amplification following by Sanger sequencing. We found that the most of heterozygous mutations in the primary tumour were a mixture of wild type alleles and heterozygous/homozygous mutations from separated clones, including DUPS14 heterozygous mutation (p.R300W) which was known to activate MEK pathway, and SETD2 mutations, (g.chr3:47061285_ 47061286insA and g.chr3:47127805C>A) which were the synthetic lethal partner of Wee1. Interestingly, we found that the two SETD2 mutations alone were able to differentiate three subclones. This result indicates that the tumour was highly subclonal, which could have a strong impact on the selection of drug resistant clones in the drug treatment.

In order to verify this hypothesis, we proceeded to test the efficacy of MEK inhibitor, Trametinib and Wee1 inhibitor, AZD1775 both separately and combined treatment, in our PDX subcutaneous model. The results showed that both 1mg/kg/day of Trametinib and 60mg/kg/day of AZD1775 treatments significantly delayed 60-70% of tumour growth as compared with the vehicle group. The tumour growth was completely arrested in the combined treatment group. In addition, 20% and 40% of the mice exhibited Trametinib and AZD1775 resistance, respectively. From our genomic analysis, all DUPS14 and SETD2 mutations from the pretreated tumours were preserved in the treatment-resistant tumours but absent in the sensitive post-treated residual tumours. The clonality analysis from the sequencing data of the pretreated tumour supports and suggests that single-agent regimes will often not be able to induce a complete response in MEITL patients.

Altogether, the current results demonstrated that our MEITL PDX model preserved the principal histological and genetic characteristics of its original tumour and is genomic stable across passages. The model also showed that MEITL could be highly subclonal and combined targeted therapy might achieve a higher efficacy for MEITL treatment.