Identification of Calr Mutations in Individuals of the Background Population Reveals Undiagnosed Myeloproliferative Neoplasms and Surprisingly Stable Low Levels of Calr Mutant Allele Burden in Individuals Not Developing Myeloproliferative Neoplasm during 10-14 Year Follow-up

Background: The goal of the emerging ‘early diagnosis and treatment intervention’ paradigm is to prevent complications and ultimately mortality associated with the current late-stage treatment of myeloproliferative neoplasms (MPNs). Screening of the background population indicates that MPNs are massively under-diagnosed and likely never discovered - possibly due to early deaths from thrombo-embolic events. Somatic gain of function mutations in exon 9 of the calreticulin gene are drivers for development of MPNs. Individuals with CALR mutations are suspected to develop disease faster than individuals with the JAK2V617F mutation, but this assumption is based on the level of mutant allele burden at diagnosis. Furthermore, it is unclear whether the presence of a CALR driver mutation always results in development of disease and how the clone develops in the early phase of the pre-diagnostic stage. Previously, we have demonstrated that the dynamics of the mutant allele burden are closely associated with the blood cell counts and may predict development of disease years before life-threatening blood cell counts required for fulfilling contemporary diagnostic criteria. In the current study we investigate follow-up data of CALR individuals from the background population and describe the development of MPN and allele burden during a decade of follow-up.

Methods: Fifteen individuals identified with type 1 (n=12) and type 2 (n=3) mutations in the CALR gene by mutational screening of the General Suburban Population Study (GESUS) with no MPN diagnosis at baseline and at least two measurements of CALR mutant allele burden (1 baseline and 1 follow-up). Data included the CALR mutant allele burden identified by digital droplet PCR and subsequently measured by quantitative PCR. The samples were collected in 2010-2013 and analyzed for CALR 2016-2017. Biochemical measurements included hemoglobin, erythrocytes, leukocytes, and thrombocytes (Trb). Data was analyzed using Graphpad prism 10.

Results: Six of the fifteen individuals were diagnosed with MPNs after the baseline identification of the CALR mutation (4 type 1 and 2 type 2). Of the six individuals diagnosed, five had a high mutant allele burden (≥20%) with elevated Trb at baseline, suggesting undiagnosed MPNs (median allele burden 38% (range 20 – 44%), median Trb 706 x10⁹ cells/L (range, 647 – 877 x10⁹ cells/L)). The median time from baseline to time of diagnosis of the undiagnosed MPNs was 5.2 years (range, 1.9 – 9,5 years) and the mutant allele burden at diagnosis a median of 36% (range, 31 – 41%). The final individual had a very low mutant allele burden at baseline: 0,071% and Trb 180 x10⁹ cells/L and was diagnosed with MPN when the CALR mutant allele burden had exponentially increased to 19% and Trb to 570 x10⁹ cells/L 12 years after the baseline sample. Individuals not developing MPNs during follow-up, had a median follow-up period of 12 years (range, 10-14 years) with a median mutant allele burden at baseline of 0.17% (range, 0.02 – 3.4%) and at the median mutant allele burden at the last follow-up was 1,0% (range, 0.034 – 3.1%). For individuals with CALR and elevated Trb at baseline had a median relative fold-change in mutant allele burden of -1.15-fold (range, -1.22 – 1.53-fold). The individual with exponential growth of the malignant clone developing MPN had a relative increase of 2692 fold, while it for individuals with no diagnosis was a median 1.51-fold (range, -2.74 – 238-fold). Two non-MPNs with mutant allele burdens exceeding 3% experienced transient ischemic attacks.

Conclusion: A CALR mutant allele burden above 20% at baseline identified individuals that likely had an undiagnosed MPN and could have been diagnosed and treated at the time of baseline sampling. The mutational dynamics for individuals not developing MPN revealed surprisingly stable levels of CALR allele burden and suggested no or very slow exponential growth within the follow-up period, contrasting previously published expectations of rapid development of CALR mutated clones. Screening for CALR can identify individuals at high risk of being diagnosed with MPN, either by a mutant allele burden above 20% or by exponential growth dynamics of the malignant clone several years before blood cell counts reveal the neoplasm.