Abstract
This study uses clinical sequencing data to examine the association between JAK2-V617F detected by solid tumor sequencing and mutations in the tumor, clonal hematopoiesis of indeterminate potential, or myeloproliferative neoplasms.
Clinical sequencing assays aim to identify somatic mutations in cancer cells for accurate diagnosis and treatment. However, most clinical-grade implementations lack patient-matched germline DNA, and supplemental analyses are needed to infer the mutational status of variants. In addition, genomic heterogeneity confounds the ability to distinguish subclonal tumor alterations from those possibly originating from the microenvironment’s nontumor component. Recent studies have shown that certain mutations identified in sequencing assays did not reflect alterations in the tumor but instead revealed alterations in infiltrating hematopoietic cells possibly from undiagnosed clonal hematopoiesis of indeterminate potential (CHIP),1,2 an age-related expansion of hematopoietic stem cells harboring somatic mutations predominantly in DNMT3A (GenBank 1788), TET2 (GenBank 54790), and ASXL1 (GenBank 171023). Mutations in other genes associated with hematologic diseases are detected less frequently in CHIP, and when these mutations are encountered, reports have attributed them either to the tumor or to CHIP.1,2,3
We examined whether detection of these mutations might indicate the presence of other disorders. Specifically, the JAK2-V617F mutation has been directly linked to myeloproliferative neoplasms (MPNs), and detection of this mutation, along with activating exon 12 JAK2 (GenBank 3717) mutations and alterations in MPL (GenBank 4352) or CALR (GenBank 811) can establish a diagnosis of MPN.4
Methods
We analyzed clinical sequencing data from 2030 solid tumors tested at Rutgers University (New Brunswick, NJ) between November 1, 2012, and August 31, 2018. Specimens from 8 patients (5 men and 3 women; median age, 74 years [range, 60-82 years]) had JAK2-V617F mutations, but on examination, mutations were present at variant allele frequencies (VAFs) significantly different from those expected from tumor purity (Table). Three patients had available specimens. To validate that JAK2 mutations existed in hematopoietic elements, we performed manual macrodissection on paraffin-embedded specimens to enrich for cancer or hematopoietic cells and sequenced these samples at high depth (>2500×) using a 49-gene panel (RainDance Technologies) on Illumina MiSeq. We identified all sites that were different from the reference using an inclusive variant caller5 and used a Bayesian approach to detect true mutations against background error6 in which mutations were tested in each sample against 33 previously sequenced, JAK2 wild-type samples. After correcting for multiple hypotheses using the Benjamini-Hochberg method, we generated a list of variants with a false-discovery rate less than 0.001. Clinical records and patient samples for this study were obtained under approval from Rutgers University Institutional Review Board protocol 2012002075 with written consent (4 of 8 patients) and protocol 20170001364 with waived consent (4 of 8 patients).
Table. Clinical and Sequencing Data for Patients With Detected JAK2-V617F Mutation.
| Diagnosis | Age at Diagnosis, y | Sequenced Specimen | Sequenced Specimen Tumor Purity, % | MPN Diagnosis | VAF by Solid Tumor Clinical Sequencing | VAF in Tumor-Enriched Sample | VAF in Lymphocyte-Enriched Sample | VAF of Tumor-Sepcific Mutations | VAF of CHIP-Associated Mutations | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Clinical Sequencing | Tumor Enriched | Lymphocyte Enriched | Clinical Sequencing | Tumor Enriched | Lymphocyte Enriched | ||||||||
| Kidney urothelial carcinoma | 73 | Kidney | 40 | ET | 2.0 | NA | NA | NA | NA | NA | NA | NA | NA |
| Unknown primary melanoma | 77 | Lymph node | 50 | PV | 4.0 | 3.5 | 5.6 | ASXL1 (20.00), BRAF (64.00), BCOR (12.00) | ASXL1 (25.66), BRAF (77.07), BCOR (17.20) | ASXL1 (1.55), BRAF (4.28), BCOR (ND) | NA | NA | NA |
| Kidney renal cell carcinoma | 60 | Kidney | 35 | None | 2.4 | NA | NA | NA | NA | NA | NA | NA | NA |
| Unknown primary melanoma | 77 | Chest wall | 70 | Myelofibrosis | 17.5 | 5.9 | 28.1 | NRAS (49.76) | NRAS (54.37) | NRAS (ND) | TET2 (2.04), U2AF1 (7.70) | TET2 (ND), U2AF1 (1.86) | TET2 (4.37), U2AF1 (10.86) |
| Lung adenocarcinoma | 67 | Lung | 20 | None | 4.8 | NA | NA | NA | NA | NA | NA | NA | NA |
| Lung adenocarcinoma | 75 | Lung | 20 | PV | 19.1 | NA | NA | NA | NA | NA | NA | NA | NA |
| Unknown primary melanoma | 82 | Salivary gland | 50 | None | 8.0 | 12.0 | 2.2 | TP53 (16.00), TET2 (18.00) | TP53 (33.39), TET2 (49.58) | TP53 (ND), TET2 (ND) | DNMT3A (6.00) | DNMT3A (5.62) | DNMT3A (10.30) |
| Colon adenocarcinoma | 65 | Liver | 30 | NA (platelets, 529 × 103/μL) | 1.8 | NA | NA | NA | NA | NA | NA | NA | NA |
Abbreviations: CHIP, clonal hematopoiesis of indeterminate potential; ET, essential thrombocythemia; MPN, myeloproliferative neoplasm; NA, not applicable; ND, not detected; PV, polycythemia vera; VAF, variant allele frequency.
SI conversion factor: To convert platelets to cells ×109 per liter, multiply by 1.0.
Results
Examination of patients’ clinical histories indicated that 4 of the 8 patients with JAK2-V617F mutations detected on solid tumor sequencing had a diagnosis of MPN; 2 had polycythemia vera, 1 had essential thrombocythemia, and 1 had myelofibrosis. One additional patient had a platelet count of 529 × 103/μL (to convert to cells ×109 per liter, multiply by 1.0) before initiating chemotherapy, whereas interpretation of the results of blood tests for the 3 other patients without a diagnosis of MPN was confounded by receipt of concurrent chemotherapy. In the validation analysis of the macrodissected specimens, 2 of the 3 patients had JAK2-V617F in lymphocyte-enriched samples at significantly higher VAFs (5.6% at 2807× and 28.1% at 3029×, respectively) than in tumor-enriched samples (3.5% at 3264× and 5.9% at 1772×, respectively) (P < .001, Fisher exact test). For these patients, tumor-specific mutations were all detected at significantly higher VAFs in tumor-enriched samples. In 1 patient, CHIP-associated mutations in U2AF1 (GenBank 7307) and TET2 were also detected at significantly higher VAFs in the lymphocyte-enriched sample. In the third patient without a prior diagnosis of MPN, JAK2-V617F was detected at a significantly higher VAF in the tumor-enriched sample (12% at 1788× vs 2.2% at 2757×). In this patient, a CHIP-associated mutation in DNMT3A was detected at a significantly higher VAF in the lymphocyte-enriched sample, suggesting that JAK2-V617F was likely present in tumor cells (Table).
Discussion
There have been conflicting reports in the literature as to whether JAK2-V617F detected by solid tumor sequencing is associated with a mutation in the tumor or CHIP.1,3 Our analysis suggests that although both of these results are possible, detection of JAK2-V617F may instead be associated with a coexistent MPN. Limitations may arise in distinguishing underlying JAK2-V617F mutations from CHIP versus those from an MPN in patients receiving chemotherapy for their solid tumors. Therefore, when MPN-associated mutations are observed in solid tumor sequencing data, caution is necessary for proper patient treatment, and a hematologic workup should be considered in the appropriate clinical context.
References
- 1.Ptashkin RN, Mandelker DL, Coombs CC, et al. . Prevalence of clonal hematopoiesis mutations in tumor-only clinical genomic profiling of solid tumors. JAMA Oncol. 2018;4(11):1589-1593. doi: 10.1001/jamaoncol.2018.2297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Severson EA, Riedlinger GM, Connelly CF, et al. . Detection of clonal hematopoiesis of indeterminate potential in clinical sequencing of solid tumor specimens. Blood. 2018;131(22):2501-2505. doi: 10.1182/blood-2018-03-840629 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li SD, Ma M, Li H, et al. . Cancer gene profiling in non-small cell lung cancers reveals activating mutations in JAK2 and JAK3 with therapeutic implications. Genome Med. 2017;9(1):89. doi: 10.1186/s13073-017-0478-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rumi E, Cazzola M. Diagnosis, risk stratification, and response evaluation in classical myeloproliferative neoplasms. Blood. 2017;129(6):680-692. doi: 10.1182/blood-2016-10-695957 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hadigol M, Khiabanian H. MERIT reveals the impact of genomic context on sequencing error rate in ultra-deep applications. BMC Bioinformatics. 2018;19(1):219. doi: 10.1186/s12859-018-2223-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rabadan R, Bhanot G, Marsilio S, Chiorazzi N, Pasqualucci L, Khiabanian H. On statistical modeling of sequencing noise in high depth data to assess tumor evolution. J Stat Phys. 2018;172(1):143-155. doi: 10.1007/s10955-017-1945-1 [DOI] [PMC free article] [PubMed] [Google Scholar]