Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2

PMID: 24325359 · DOI: 10.1056/NEJMoa1312542 · Journal: New England Journal of Medicine (2013)

TL;DR

Nangalia et al. performed whole-exome sequencing on 151 patients with myeloproliferative neoplasms (MPN) — 48 polycythemia vera (PV), 62 essential thrombocythemia (ET), 39 myelofibrosis (MF), and 2 unclassifiable — and discovered recurrent somatic indels in exon 9 of CALR (calreticulin), a gene never previously implicated in cancer. Across exome and follow-up Sanger screening, CALR mutations were present in 70–84% of MPN patients lacking JAK2 or MPL mutations (essentially all the JAK2/MPL-negative ET and MF patients), in 8% of myelodysplastic syndromes, and in none of 1517 other (mostly solid) cancers or 550 controls. All 19 distinct variants generate a +1 base-pair frameshift that replaces the wild-type acidic C-terminal and KDEL retention motif with a novel basic peptide. Clonal phylogeny in hematopoietic colonies showed CALR mutations sat in the earliest node, consistent with an initiating role.

Cohort & data

• Discovery cohort: 151 patients with MPN passing bioinformatic QC (out of 168 sequenced), comprising 48 PV, 62 ET, 39 MF, and 2 MPNU. Average exome coverage 141×. Tumor DNA from blood granulocytes; constitutional DNA from cultured T cells (n=34), isolated T cells (n=42), or buccal swabs (n=75). Dataset: mpn_cimr_2013.
• Follow-up screening: Sanger sequencing of CALR exon 9 in an additional 1345 hematologic cancers and 52 controls, plus existing whole-exome/whole-genome data for 502 solid tumors, 498 controls, and 1015 cancer cell lines.
• Assays/methods: whole-exome-seq of matched tumor/germline pairs with orthogonal target-capture resequencing; sanger-sequencing of CALR exon 9 for follow-up; Sanger genotyping of ~300 hematopoietic colonies across 5 patients for clonal analysis; immunofluorescence and flow cytometry for protein localization; Bayesian Dirichlet-process clonal inference.

Key findings

• 1498 validated somatic mutations across 150 samples (range 1–32 per sample). Median mutations per patient: 6.5 (PV), 6.5 (ET), 13.0 (MF). The higher MF burden was significant (one-way ANOVA P=0.008; pairwise t-test P<0.001 MF vs ET, P=0.008 MF vs PV with Bonferroni correction).
• JAK2 V617F was the most prevalent driver: present in 48/48 (100%) PV, 35/62 (56%) ET, and 27/39 (69%) MF patients.
• Discovery of recurrent somatic indels in CALR exon 9 in 26/151 (17%) of the exome cohort — exclusively in patients lacking JAK2 or MPL mutations: 26/31 (84%; 95% CI 66–94) of JAK2/MPL-negative ET/MF patients, and 0/120 of those with JAK2 or MPL mutations.
• Mutual exclusivity: 146/151 patients (97%) carried JAK2, MPL, or CALR mutations in a mutually exclusive fashion (q<0.01).
• 148 CALR mutations identified across the full cohort, 19 distinct variants — all indels (14 deletions, 2 insertions, 3 complex indels) in exon 9, all producing a +1 bp frameshift. Two variants dominate: L367fs46 (52-bp deletion flanked by 7 bp of microhomology, 67/148 = 45%) and K385fs47 (5-bp insertion forming an inverse duplication, 61/148 = 41%). L367fs*46 was enriched in MF vs ET (chi-square P=0.009).
• Follow-up disease distribution of CALR mutations: 110/158 (70%; 95% CI 62–77) JAK2/MPL-negative MPN; 80/112 (71%) JAK2/MPL-negative ET; 18/32 (56%) primary MF without JAK2/MPL; 12/14 (86%) post-ET MF without JAK2/MPL; 10/120 (8%; 95% CI 4–15) MDS; 1/33 CMML; 1/29 ACML. Zero mutations in 511 JAK2/MPL-mutated MPN, 287 lymphoid cancers, 502 solid tumors (breast, lung, prostate, colorectal, skin, bone, renal, meningioma), 1015 cell lines, or 550 controls.
• Clinical phenotype: Among ET patients, CALR-mutated cases had significantly higher platelet counts (Wilcoxon P<0.001) and lower hemoglobin (Student’s t P=0.02) vs JAK2-mutated cases, and higher rates of transformation from ET to MF (Fisher P=0.03). No significant survival difference (small event count).
• Mutant protein has a novel C-terminal: All 19 indels are predicted to lose 27 amino acids of the acidic, calcium-binding C-domain (including the KDEL ER-retention signal) and gain a 36-aa novel basic peptide. The pattern (uniform +1 frameshift converging on one C-terminal sequence) is opposite that of tumor suppressors like TET2/ASXL1 (mixed nonsense and frame-shift), implying gain-of-function selection.
• Subcellular localization: Mutant CALR expressed in HEK293T and COS-7 cells localized to the ER and colocalized with wild-type CALR; no increased Golgi or cell-surface accumulation in transfected cells, 32D hematopoietic cells, or peripheral-blood leukocytes from 5 CALR-mutated patients (despite 40–100% granulocyte tumor burdens).
• Cell-of-origin and clonal order: CALR mutations were detected in flow-sorted lin−CD34+CD38−CD45RA−CD90+ hematopoietic stem cells and downstream myeloid/erythroid progenitors in 3 patients. Genotyping of ~300 individual hematopoietic colonies in 5 patients placed CALR mutation in the earliest phylogenetic node in all five — consistent with an initiating event.

Genes & alterations

• CALR — discovery: recurrent exon 9 +1 bp frameshift indels (19 distinct variants; L367fs46 deletion and K385fs47 insertion dominate). Found in 70–84% of JAK2/MPL-negative MPN. Mutually exclusive with JAK2 and MPL; arises in HSC compartment; sits in earliest phylogenetic node. Produces a mutant protein with novel basic C-terminus lacking KDEL ER-retention signal — proposed gain-of-function oncogene. First report of somatic mutations in an ER chaperone in any cancer.
• JAK2 — V617F in all 48/48 PV, 35/62 (56%) ET, 27/39 (69%) MF. Strictly mutually exclusive with CALR mutations.
• MPL — mutated in 7 patients, all ET or MF with unmutated JAK2. Mutually exclusive with CALR.
• TET2 — 25 somatic variants in 22 patients (epigenetic regulator).
• DNMT3A — 13 somatic variants in 12 patients (epigenetic regulator).
• ASXL1 — 13 somatic variants in 12 patients; co-mutated with splicing factors U2AF1 and SRSF2.
• EZH2 — somatic variants in 4 patients.
• IDH1 / IDH2 — somatic variants in 3 patients combined; IDH1 co-mutated with SRSF2.
• U2AF1 — 4 patients (splicing factor).
• SF3B1 — 3 patients (splicing factor).
• SRSF2 — 2 patients; co-mutated with TET2, IDH1, and ASXL1 — pattern echoing MDS.
• CBL — 1 patient.
• NFE2 — 2 patients.
• SH2B3 (LNK) — 1 patient.
• CHEK2 — missense somatic mutations in 1 patient each with PV, ET, and MF (q=0.008); not previously reported in MPN.

Clinical implications

• Diagnostic: Adding CALR exon 9 sequencing to JAK2 and MPL testing captures a driver in 146/151 (97%) of MPN patients in this cohort. The authors propose CALR testing in peripheral blood as a diagnostic tool for JAK2/MPL-negative thrombocytosis, paralleling how JAK2 testing transformed MPN diagnosis. Confirmatory work needed.
• Prognostic phenotype within ET: CALR-mutated ET presents with higher platelet counts, lower hemoglobin, and higher risk of transformation to MF than JAK2-mutated ET (P=0.03). Survival difference not demonstrated in this study (small deaths).
• Disease specificity: CALR indels are essentially MPN/MDS-restricted (none in 287 lymphoid cancers, 502 solid tumors, or 1015 cell lines screened here) — strong diagnostic discrimination from reactive thrombocytosis and from non-myeloid malignancies.
• Therapeutic implication (mechanistic): Loss of the KDEL ER-retention signal and acquisition of a uniform novel basic C-terminal suggest a gain-of-function mechanism distinct from the JAK2/MPL signaling axis, raising the prospect of CALR-mutant–specific therapeutic strategies (not pursued in this paper).

Limitations & open questions

• Authors note low exome coverage at CALR (median depth 10×) caused initial under-detection; orthogonal Sanger sequencing of exon 9 was required and recovered an additional case among 6 “triple-negative” patients.
• The lower CALR mutation prevalence in the follow-up Sanger cohort (70%) vs the exome cohort (84%) is attributed to Sanger sensitivity on whole-blood samples, not biology.
• Mechanism of oncogenesis remains undefined. No altered Golgi or cell-surface CALR accumulation observed; authors explicitly state their data do not exclude partial ER loss via the secretory pathway. Whether mutant CALR drives proliferation via altered ER chaperone function, calcium handling, or extracellular signaling is unresolved.
• Survival impact unresolved — too few deaths to power survival comparison between CALR-mutant and JAK2-mutant MPN.
• Clonal analysis placing CALR at the earliest phylogenetic node was performed in only 5 patients; Dirichlet-process inference was limited by sparse CALR coverage and could not always resolve subclones with similar tumor burdens.
• No drug treatment data; therapeutic relevance untested.

Citations from this paper used in the wiki

• “Somatic CALR mutations were found in 70 to 84% of samples of myeloproliferative neoplasms with nonmutated JAK2, in 8% of myelodysplasia samples, in occasional samples of other myeloid cancers, and in none of the other cancers.” (Abstract)
• “A total of 148 CALR mutations were identified with 19 distinct variants. Mutations were located in exon 9 and generated a +1 base-pair frameshift, which would result in a mutant protein with a novel C-terminal.” (Abstract)
• “Exome sequencing showed that 146 of 151 patients with myeloproliferative neoplasms (97%) had mutations in JAK2, MPL, or CALR in a mutually exclusive manner (q<0.01).” (Results)
• “Among patients with essential thrombocythemia, those with CALR mutations, as compared with those with JAK2 mutations, presented with significantly higher platelet counts (P<0.001 by the Wilcoxon rank-sum test) and lower hemoglobin levels (P = 0.02 by Student’s t-test). Patients with CALR mutations had a significantly higher incidence of transformation from essential thrombocythemia to myelofibrosis than did those with JAK2 mutations (P = 0.03 by Fisher’s exact test).” (Results)
• “In all five patients, CALR mutations arose in the earliest phylogenetic node, consistent with mutation of CALR being an initiating event in these patients.” (Results)
• “Missense somatic mutations in CHEK2, which have not been reported previously in myeloproliferative neoplasms, were found in 1 patient each with polycythemia vera, essential thrombocythemia, and myelofibrosis (q = 0.008).” (Results)

This page was processed by crosslinker on 2026-05-09.