Clinical effect of driver mutations of JAK2, CALR, or MPL in primary myelofibrosis

Abstract

We studied the impact of driver mutations of JAK2, CALR, (calreticulin gene) or MPL on clinical course, leukemic transformation, and survival of patients with primary myelofibrosis (PMF). Of the 617 subjects studied, 399 (64.7%) carried JAK2 (V617F), 140 (22.7%) had a CALR exon 9 indel, 25 (4.0%) carried an MPL (W515) mutation, and 53 (8.6%) had nonmutated JAK2, CALR, and MPL (so-called triple-negative PMF). Patients with CALR mutation had a lower risk of developing anemia, thrombocytopenia, and marked leukocytosis compared with other subtypes. They also had a lower risk of thrombosis compared with patients carrying JAK2 (V617F). At the opposite, triple-negative patients had higher incidence of leukemic transformation compared with either CALR-mutant or JAK2-mutant patients. Median overall survival was 17.7 years in CALR-mutant, 9.2 years in JAK2-mutant, 9.1 years in MPL-mutant, and 3.2 years in triple-negative patients. In multivariate analysis corrected for age, CALR-mutant patients had better overall survival than either JAK2-mutant or triple-negative patients. The impact of genetic lesions on survival was independent of current prognostic scoring systems. These observations indicate that driver mutations define distinct disease entities within PMF. Accounting for them is not only relevant to clinical decision-making, but should also be considered in designing clinical trials.

Figure 1

Cumulative incidence of anemia, thrombocytopenia, and marked leukocytosis in PMF patients stratified according to their driver mutation. The thresholds for hemoglobin level and WBC count are those of the IPSS, whereas that for PLT count is the lower limit of normal range. Cumulative incidences were estimated with a competing risk approach, considering death for any cause as a competing event. Vertical tick marks indicate right-censored patients. (A) Cumulative incidence of anemia (hemoglobin <10 g/dL). CALR-mutant patients had a lower incidence of anemia compared with the remaining patients (maximum P value equal to .004). (B) Cumulative incidence of thrombocytopenia (PLT count <100 × 10⁹/L). The cumulative incidence of thrombocytopenia was significantly lower in CALR-mutant patients compared with the remaining ones (P < .001 in all comparisons). (C) Cumulative incidence of marked leukocytosis (WBC count >25 × 10⁹/L). The cumulative incidence of marked leukocytosis was significantly lower in CALR-mutant patients compared with JAK2-mutant (P = .004) or triple-negative patients (P < .001).

Figure 2

Cumulative incidence of thrombosis in PMF patients stratified according to their driver mutation. Vertical tick marks indicate right-censored patients. JAK2-mutant patients had a higher incidence of thrombosis than those with CALR mutation (P = .021). This difference remained statistically significant after adjusting for age (SHR, 2.19; 95% CI, 1.15-4.18; P = .017), the estimated risk of thrombosis being about 2-fold in JAK2-mutant compared with CALR-mutant patients.

Figure 3

Cumulative incidence of leukemic transformation in PMF patients stratified according to their driver mutation. Vertical tick marks indicate right-censored patients. Triple-negative patients had higher incidence of leukemic transformation compared with both CALR-mutant and JAK2-mutant patients (maximum P value equal to .043).

Figure 4

Kaplan-Meier analysis of survival of PMF patients stratified according to their driver mutation. Vertical tick marks indicate right-censored patients. In univariate analysis, CALR-mutant patients had a better OS than JAK2-mutant (HR 2.3, P < .001), MPL-mutant (HR 2.6, P = .009), and triple-negative patients (HR 6.2, P < .001). Three JAK2-mutant patients had short follow-up and were not included in the analysis.

Figure 5

Kaplan-Meier analysis of survival of PMF patients stratified according to their driver mutation and subdivided according to their IPSS risk. Vertical tick marks indicate right-censored patients. (A) “Lower” IPSS risk subgroup, including patients with low or intermediate-1 IPSS risk: CALR-mutant patients had longer survival compared with either JAK2 (V617F)-mutant (P = .011) or triple-negative patients (P < .001). (B) “Higher risk” subgroup, including patients with intermediate-2 or high IPSS risk: CALR-mutant patients had longer survival compared with the remaining genetic subgroups (maximum P value equal to .023).

Figure 6

Kaplan-Meier analysis of survival of PMF patients stratified according to the risk categories defined by a clinical-molecular prognostic model. This model includes the variables reported in Table 2, that is, IPSS variables plus CALR, JAK2, and MPL mutation status. We assigned each factor an integer weight according to the corresponding HR in the multivariable-Cox regression of Table 2. Scores were then recoded into the 5 risk categories shown in this figure: details are reported in the last section of “Results.” Based on the Akaike information criterion, which compares quality of models, the clinical-molecular model provided a better stratification than the IPSS. This analysis serves as a proof of concept that accounting for driver mutations improves the risk stratification provided by IPSS.

Figure 7

Kaplan-Meier analysis of survival of PMF patients stratified according to their genotype, as it was known in different time periods. (A) OS of the whole population of PMF patients: the genetic basis of MPNs was unknown before 2005, and therefore no genotypic subgroup could be defined. (B) PMF patients stratified according to JAK2 or MPL mutation status: these mutant genes where identified in 2005 and 2006, respectively. (C) PMF patients stratified according to JAK2, CALR, or MPL mutation status: somatic mutations of calreticulin were identified in 2013.