PRAME mRNA Expression in AML/MDS and HLA Genotype Analysis: Impact on Population Coverage and Design of TCR-Based Immunotherapies

The cancer-testis antigen PRAME (PReferentially expressed Antigen in MElanoma) is expressed at high levels in several oncological indications, including leukemias and solid tumors, as confirmed by mRNA or protein expression studies. Its low to absent expression on normal tissues allows suitable discrimination of tumors for targeted immunotherapy. As an intracellular protein, PRAME-derived peptides can be presented by several HLA class I allotypes, including HLA-A*02:01, and specific T cell recognition has been reported for different PRAME peptide-HLA (pHLA) ligands. Thus, developing a small set of TCRs covering different HLAs could increase the number of PRAME-positive patients eligible for treatment with T cell receptor (TCR)-based immunotherapy approaches.

We determined PRAME expression in a group of Caucasian patients with acute myeloid leukemia (AML) and high-risk myelodysplastic syndrome (MDS). In addition, we analyzed the HLA-A genotype distribution in a set of healthy blood donors to determine how many and which different TCRs would be needed to optimally cover a Caucasian population in Germany.

For the analysis of PRAME expression, bone marrow (BM) and peripheral blood (PB) samples from 165 patients (AML=133, MDS=32), obtained from multiple clinical sites across Germany, were studied. Ninety-six samples were from newly diagnosed patients (58%) while 68 samples were from relapsed/refractory patients (41%), irrespective of percentage of leukemic blasts. Disease status was not known for one patient. PRAME mRNA expression was measured by quantitative reverse transcriptase polymerase chain reaction (RT-qPCR). A threshold for PRAME-negativity was derived by measuring average PRAME mRNA expression in 22 healthy individuals. Detection of 99 or more copies of PRAME, as measured in 12.5 ng RNA, was considered positive.

Of the 165 AML/MDS patients, 79 (48%; AML= 69, MDS=10) were positive for PRAME mRNA. Among relapsed/refractory patients, 52% (35/68) were classified as PRAME-positive, versus 45% (43/96) of first-diagnosis AML/MDS patients. NGS high-resolution HLA-A genotyping showed that 55% of all patients had an HLA-A*02:01 allotype, which is higher than the approximately 45% expected for a random Caucasian population and accounted for by the fact that some patients were included with known HLA-A2 status due to prior assessment for transplant eligibility. In combination, 43 (26%) AML/MDS patients were positive for PRAME above the threshold level as well as for HLA-A*02:01, thereby being potential candidates for treatment with PRAME-specific, HLA-A*02:01-restricted TCR immunotherapy.

In a group of 141 healthy local blood donors, NGS high-resolution HLA-A genotyping showed an expected distribution of HLA-A*02:01, HLA-A*01:01, HLA-A*03:01, HLA-A*24:02 and HLA-A*11:01 of 43%, 26%, 25%, 19% and 8% respectively, with some donors heterozygous for two of these particular allotypes. To determine the smallest number of TCRs needed to address the largest number of patients, we assessed how the non-overlapping distribution for each potential HLA-A allotype would contribute to patient coverage. Starting with the 43% of patients who would be covered by an HLA-A*02:01-restricted TCR immunotherapy product, additional coverage of 21%, 14%, 9%, and 1% of patients could be achieved with TCR products using HLA-A*01:01-, HLA-A*03:01-, HLA-A*24:02- and HLA-A*11:01-restricted TCRs, respectively.

In conclusion, approximately half of AML/MDS patients are PRAME-positive and could be addressed with PRAME-specific TCR immunotherapy approaches. An HLA-A*02:01-restricted TCR addresses the most patients in a population in Germany at 43%, while coverage could be increased to 87% with addition of three TCRs restricted by the next most common HLA-A allotypes, contingent upon suitable TCR discovery.