Myelodysplastic syndromes (MDS) constitute a heterogeneous group of blood cancers that most commonly affect individuals of 60 year or older. With an increasing ageing population the incidence of MDS is rising. The clinical presentation of MDS is very diverse, some patients live with low risk disease for many years whereas others rapidly progress to Acute Myeloid Leukemia (AML) with limited treatment options. Therefore early diagnosis and accurate risk assessment is critical for optimal clinical management.
Diagnosis of MDS is frequently challenged by common symptoms for the elderly such as fatigue, recurrent infections and cytopenias. Clinicians evaluate morphology of the bone marrow and cytogenetic profiles to diagnose MDS. In recent years genetic studies have identified more than 80 genes that are frequently mutated in MDS. Increasingly, MDS patients receive genetic profiling at diagnosis, which often identify several key mutations. However there is no guidance on how to use molecular findings to inform diagnosis and treatment decisions. Current guidelines for risk stratification and subsequent treatment decisions rely solely on clinical presentation and cytogenetic findings which are present in only 50% of patients.
To deliver evidenced based and reproducible standards for the incorporation of molecular biomarkers into formal clinical guidelines for MDS we called for an international collaboration. In 2015, under the aegis of the International Working Group for prognosis in MDS >25 participating centers from >12 countries joined forces in order to create the largest database of MDS patients with matched molecular and clinical information. The project is led by the Papaemmanuil laboratory at Memorial Sloan Kettering Cancer Center and is overseen by the IWG-PM oversight committee, which includes leading clinical and research investigators in MDS. Together this will deliver the new international guidelines for the incorporation of gene mutations into disease classification for MDS and for the International Prognostic Scoring System (IPSS-R).
This study represents an early finding of the IWG-PM project. While working towards the development of the new IPSS-R Dr. Elsa Bernard, post-doctoral fellow and lead author of the study performed detailed characterization of somatic alterations in MDS to include point mutations, chromosomal deletion or copy-neutral loss of heterozygosity (LOH). Today in the clinic, genes are simply reported as mutated or not mutated. Our analyses allowed us for the first time to encode the type of the mutations present and consequently assess if a gene had one or both copies mutated. One of these genes, TP53, really stood out. With our analyses 1 in 3 patients had only one copy of TP53 mutated (mono-allelic) whereas 2 in three patients had both copies mutated by diverse mechanisms (mutation and mutation, mutation and chromosomal loss or mutation and copy-neutral LOH). This is consistent with the well established mechanisms that lead to bi-allelic inactivation of TP53, which is a tumor suppressor gene.
We then set out to evaluate the genomic and clinical associations of mono- and bi-allelic TP53 mutations. Mutated TP53 have been historically associated with genome instability, dismal outcome and resistance to conventional therapies. We were very surprised by the results: the allelic state of TP53 (mono- versus bi-allelic) are associated with dramatically distinct genomic profiles and clinical presentation. Mono-allelic patients did not differ from wild-type patients with regards to genome instability, response to therapy, overall survival and progression to AML. The established associations with high-risk disease were restricted to the patients with bi-allelic mutations. The implications for accurate patient risk stratification are profound.
The clinical differences between MDS patients with mono- or bi-allelic TP53 mutations were so striking in our cohort, that we really wanted to confront the findings in other datasets. We therefore ascertained two independent cohorts. In a small cohort from St James’s University Hospital (United Kingdom) with sequential sampling of 12 TP53-mutated patients at both MDS and AML phase, we observed that patients with bi-allelic TP53 at MDS progressed to AML in just a few months. In patients that transitioned from mono- to bi-allelic during disease progression, we observed inter-clonal competition and achievement of clonal dominance for the bi-allelic clone. Those findings confirmed the importance of mapping TP53 allelic state at diagnosis and for disease surveillance. Reassuringly, we also validated the genomic and clinical associations of TP53 allelic states in a cohort of 1,120 patients from the Japanese MDS consortium led by Dr. Seishi Ogawa.
In an era where molecular biomarkers are increasingly used to guide clinical decisions, our results highlight the importance of detailed biomarker characterization that considers the influence of mutation type and allelic state for the delivery of precision medicine. Our study opens the field of the non-equivalence of TP53 mutations in clinical biomarker development, which may be pertinent across cancer indications.
This study was supported by the MDS Foundation. We would like to thank all participating patients as well as all clinical and research investigators for their contribution to the study.