Introduction

Calreticulin (CALR) exon 9 mutations are the second most common mutations in patients with essential thrombocythaemia (ET) and primary myelofibrosis (PMF). The overall risk of myelofibrotic change, either PMF or secondary myelofibrosis (SMF), appears to be more frequent in patients with CALR type 1 than type 2 mutations [1] and male patients are more likely than females to have CALR type 1 mutations [2].

It had been noted locally that a significant number of young male patients with CALR mutations appeared to have a more aggressive/myelofibrotic disease phenotype. The aim of this study was to identify if there were further differences in the incidence of myelofibrotic change within subgroups of patients with CALR mutations, specifically looking to see if males with CALR type 1 mutations were more likely to have myelofibrosis than females and if they developed it at a younger age.

Methods

The study reviewed all patients with a positive CALR exon 9 mutation identified within our unit between February 2016 and September 2020. CALR mutations were identified as previously described [3]. Mutation sub-types were identified and compared with patient sex, age at diagnosis, initial MPN diagnosis, and subsequent disease transformation. Statistical analysis was performed with p-values for the total male and female patients with myelofibrosis within each age group calculated using Fisher’s exact test.

Results

CALR type 1 mutations (including type 1-like mutations throughout) or type 2 mutations were identified in 123 patients and were overall more frequent in males (74 male, 49 female, p = 0.024) as shown in Table 1. Seventy-five (61.0%) of these patients had CALR type 1 mutations and 48 (39.0%) type 2. The incidence of CALR type 2 mutations in male compared with female patients were significantly higher in those under the age of 60 but not different in those over 60; however, further detailed analysis of myelofibrotic variances in CALR type 2 mutations was prevented by low patient numbers within each subgroup.

Table 1 Diagnoses (initial and subsequent) by age of male and female patients with CALR mutations
Full size table

Of the 44 male patients with CALR type 1 mutations, 9 were diagnosed with PMF and 35 with ET. Twelve of those with ET subsequently transformed to SMF. Thirty-one female patients had CALR type 1 mutations with 5 of these being diagnosed with PMF and 26 with ET. Two transformed to SMF. The overall risk of myelofibrosis in patients with CALR type 1 mutations was therefore significantly higher in male than in female patients with 21 of 44 (47.7%) versus 7 of 31 (22.6%), respectively, p = 0.0315.

Whilst the overall risk of myelofibrosis with CALR type 1 mutations is highest in males, this risk appears to vary between age groups. The mean age of developing myelofibrosis, either PMF or SMF, in the cohort was 61 overall with it being 59 for males and 66 for females. The median was 62 overall with a median age of 60 for males and 65 for females. These findings indicate that myelofibrosis occurs at a younger age in male patients who have CALR type 1 mutations.

To consider this further, the data was split by age group and is summarised in Table 1. Of the 32 patients aged 60 or over with CALR type 1 mutations, 19 were male and 13 were female. Eight of the male and 3 of the female patients had myelofibrosis. Whilst in keeping with the known increased frequency of males having CALR type 1 mutations, the risk of myelofibrosis between sexes was not statistically different in this older age group, p = 0.4501.

The risk of myelofibrosis was also not quite statistically different when looking at those with CALR type 1 mutations under 60 years of age overall at 0.0637; however, myelofibrosis became increasingly common in the male sex among younger age groups. Of the 18 patients under 50 years old, 13 were male and 5 were female. All females had ET without progression but 9 of the males had developed myelofibrosis, p = 0.0294. Of the 7 patients under 40 years old, the 2 females had ET but all 5 males had myelofibrosis, p = 0.0476.

Discussion

This study highlights that in our cohort, CALR type 1 mutations are associated with an overall increased incidence of myelofibrosis in male compared with female patients. This variation is particularly evident in those diagnosed at a young age where despite low patient numbers, there is a statistically increased risk of myelofibrosis among male patients under the age of 40 which persists in those under 50 but becomes less relevant in those diagnosed later in life.

In our cohort, all 9 patients with myelofibrosis under the age of 50 were male and this finding does not appear to be an outlier. Although not specifically investigated in their study, previously published data by Cabagnols et al. [4] showed a similar trend in their cohort, with males accounting for 83.3% of patients under 50 with both CALR type 1 mutations and myelofibrosis. This was despite them making up only 53.4% of CALR type 1 mutation positive patients overall.

The reason for these variations in incidence and MPN phenotype in younger age groups is not currently understood but may relate to hormonal variations. The incidence of myelofibrosis between male and female patients being significantly different in those under 50 years old but not significant beyond this age appears to correlate with the age of menopause.

High oestrogen levels present in premenopausal females may be protective against both CALR type 1 mutation or allele burden, and transformation to myelofibrosis. Tamoxifen, a selective oestrogen receptor modulator, has long been known to exhibit a complex spectrum of both oestrogen agonistic and antagonistic effects depending on the specific tissue affected [5]. The recent Tamarin study showed that tamoxifen could induce apoptosis of CALR mutated haematopoietic progenitors, and based on preclinical studies, this appeared to work by modulation of oestrogen receptor alpha [6]. It remains unclear if Tamoxifen’s modulatory role in this area is oestrogen agonistic or antagonistic, but these results give further weight to oestrogen’s relevance in MPN associated with CALR mutations.

Conclusion

Whilst it is already known that CALR type 1 mutations are more common in males, this study is the first to demonstrate that even within the demographic of those with CALR type 1 mutations, male patients have a worse MPN phenotype than female patients with it occurring at a younger age and being more myelofibrotic in nature. The potential reasons for this variability as discussed could lead to new opportunities for treatment and warrant further investigation.