• myelofibrosis;
  • leukaemia;
  • single nucleotide polymorphism

Primary myelofibrosis (PMF) is a chronic myeloproliferative neoplasm (MPN) characterized by heterogeneous clinical presentation and a dismal outcome, with median survival of <6 years (Cervantes et al, 2009). In 15–20% of patients the disease evolves into an aggressive form of acute leukaemia. Variables associated with leukaemia transformation are peripheral blood blasts >2%, thrombocytopenia, and high-risk karyotypic abnormalities (Tefferi et al, 2012).

The molecular mechanisms underlying the leukaemic transformation are still poorly characterized. There is growing information in the literature concerning germline variants that, by themselves or by predisposing to acquiring somatic mutations, could affect disease occurrence, phenotype and outcome. A JAK2 haplotype, named 46/1, is strongly associated with MPN and accounts for a large part of sporadic MPN (Jones et al, 2009). The A3669G polymorphism, which stabilizes the mRNA of glucocorticoid receptor beta, is significantly more represented in patients with polycythaemia vera (PV) and PMF (Varricchio et al, 2011). Furthermore, the G/G haplotype was associated with shorter overall and leukaemia transformation-free survival, particularly in JAK2 V617F mutated patients (Poletto et al, 2012).

Recently, possible associations among five single nucleotide polymorphisms (SNPs) in four genes involved in DNA repair mechanisms and leukaemic transformation were analysed in a case-control study of 422 subjects with PV and essential thrombocythaemia (ET) including 64 cases of acute leukaemia (Hernández-Boluda et al, 2012). The Gln/Gln genotype (‘CC’ allele) in codon 751 of exon 23 of the ERCC2 (XPD) gene was associated with a 4·9-fold greater risk (95% confidence interval [CI], 2·0–12) of leukaemia compared with the other genotypes, as well as with a significantly higher rate of non-myeloid malignancies (odds ratio 4·2; 95% CI, 1·5–12). The 5′[RIGHTWARDS ARROW]3′ helicase protein ERCC2 (XPD) has an important role in the nucleotide excision repair (NER) pathway, is involved in transcription initiation and in the control of cell cycle and apoptosis. The Gln variant was found to predict for a significantly increased risk of developing acute myeloid leukaemia after chemotherapy (Allan et al, 2004).

This study aimed to determine whether the ERCC2 751 Gln/Gln polymorphism is associated with an increased risk of leukaemic progression in PMF patients and/or influenced disease phenotype and outcome. We examined 434 PMF patients diagnosed according to the 2008 World Health Organization criteria. Patients were recruited from six European haematology units (Florence, Pavia, Southampton, Barcelona, Athens and Mannheim) following local Ethical Committee approval. Informed consent was provided according to the Declaration of Helsinki. We used a population-based case-control approach to determine the predisposition to leukaemic transformation in 67 cases who progressed to leukaemia and 367 patients who did not.

The candidate SNP (rs13181) consists of an A to C shift leading to Lys751Gln in exon 23 of ERCC2. Genotyping was performed by real-time reverse transcription polymerase chain reaction (RTqPCR) in peripheral blood DNA, obtained at diagnosis, using a commercially available TaqMan SNP assay (ID: C_3145033_10; Applied Biosystem, Life Technologies). Allelic discrimination analysis was performed using the ABI Prism 7300 Detection System. JAK2 V617F mutation and burden were determined by RTqPCR (Guglielmelli et al, 2009) or pyrosequencing; MPL W515L/K mutations were determined using RTqPCR (Pancrazzi et al, 2008). The rs13181 AA, AC, CC variants were considered as ordered categorical variables. The chi-square or Fisher's exact test (two-tailed) or chi-square test for trend (larger contingency table) with Bonferroni correction were used to compare variables among groups of patients that had been categorized according to ERCC2 genotype. Continuous variables were analysed using the Mann–Whitney U test or Kruskal–Wallis test for multiple comparisons. Kaplan–Meier analysis and the log-rank test were used to estimate overall and leukaemia-free survival. A P value of <0·05 was considered to indicate statistical significance; all tests were two-tailed. Data were processed using spss Version 17·0 software (StatSoft).

The main clinical and haematological characteristics at diagnosis of the 67 leukaemic cases and the 367 control patients are reported in Table 1. Statistically significant differences in leukaemic cases compared with control patients were found in relation with mortality rate (83·6% vs. 24·5%, P < 0·001), mean haemoglobin (P = 0·002), percentage of blasts in peripheral blood (P < 0·001) and V617F allele burden (P = 0·005); case patients were more prone to have leucocytosis (P = 0·004) and >1% peripheral blood blasts (P < 0·001). Finally, the case group included a greater proportion of patients (41·8%) with higher International Prognostic Scoring System (IPSS) risk scores (Intermediate 2 plus high-risk) than the control group (31·5%; P = 0·019).

Table 1. Clinical characteristics and ERCC2 genotype in study population
CharacteristicCases (n = 67)Controls (n = 367) P
  1. Cases, patients who progressed to acute leukaemia; controls, patients who did not progress to acute leukaemia; *, number of patients with available information; IPSS, International Prognostic Scoring System; SD, standard deviation.

Age (years), median (range)62 (22–88)61 (14–90)0·315
Male gender, n (%)24 (35·8)145 (39·5)0·569
Follow up (months), median (range)59·6 (4–337)57·0 (1–340)0·398
Deaths, n (%)56 (83·6)90 (24·5) <0·001
Leucocytes (× 109/l), mean ± SD15·7 ± 19·211·5 ± 9·90·103
Haemoglobin (g/l), mean ± SD106 ± 23115 ± 24 0·002
Platelets (× 109/l) (N = 482)*, mean ± SD408·2 ± 335·8429·2 ± 364·10·446
% peripheral blast cells, mean ± SD1·4 ± 3·20·5 ± 1·6 <0·001
Splenomegaly (N = 426), n (%)55/67 (82·1)267/359 (74·4)0·085
Abnormal karyotype (N = 204)*, n (%)10/32 (31·3)35/172 (20·3)0·226
Unfavourable karyotype (N = 202)*, n (%)3/31 (9·7)9/171 (5·3)0·339
JAK2 V617F, n (%)35 (52·2)229 (62·4)0·101
% JAK2 V617F allele burden (N = 231), mean ± SD60·0 ± 28·745·0 ± 21·8 0·005
MPL W515L/K (N = 429)*, n (%)5/66 (7·6)18/363 (5·0)0·229
Constitutional symptoms, n (%)25 (37·3)110 (27·2)0·110
Age > 65 years, n (%)25 (37·3)146 (39·8)0·759
Haemoglobin < 100 g/l, n (%)23 (34·3)100(27·2)0·091
Leucocytes > 25 × 109/l, n (%)10 (14·9)25 (6·8) 0·004
Circulating blasts ≥ 1%, n (%)19 (28·4)53 (14·4) <0·001
IPSS, n (%)
Low26 (38·8)131/366 (35·8) 0·019
Int-113 (19·4)120/366 (32·8)
Int-212 (17·9)74/366 (20·2)
High16 (23·9)41/366 (11·3)
ERCC2 Lys751Gln, N (%)
AA29 (43·3)152 (41·4)0·493
AC26 (38·8)166 (45·2)
CC12 (17·9)49 (13·4)
ERCC2 Lys751Gln: CC/AA+AC, N (%)
AA + AC55 (82·1)318 (86·6)0·323
CC12 (17·9)49 (13·4)

Genotype distribution of the ERCC2 rs13181 SNP in the entire patient population was: AA 41·7%, AC 44·2%, CC 14·1%, AA/AC 85·9%; the frequency of minor C allele was 0·361, similar to healthy controls (Hernández-Boluda et al, 2012). Among those patients who evolved to leukaemia, genotype distribution (AA 43·3%, AC 38·8%, CC 17·9%) was similar to the control group (AA 41·4%, AC 45·2%, CC 13·4%); the frequency of the minor C allele was 0·373 compared with 0·360 in controls.

We compared the clinical characteristics of 61 patients having the homozygous CC genotype with those having the AA/AC genotype (Table 2). The number of leukaemic transformations was similar in the CC and AA/AC genotypes (19·7% vs. 14·7%), and no other meaningful difference emerged from the comparison of patients' characteristics. The absence of an unique clinical phenotypic profile associated with the minor ERCC2 allele in homozygosis was also noted in the study involving PV and ET patients (Hernández-Boluda et al, 2012). Finally, there were no statistically significant correlations between any of the possible genotypes (AA vs. AC vs. CC; not shown) or the homozygous CC genotype versus AA/AC genotype and the rate of leukaemia transformation, leukaemia-free survival and overall survival.

Table 2. Clinical characteristics of the study patients stratified according to ERCC2 genotype
CharacteristicCC (n = 61)AA + AC (n = 373) P
  1. *, number of patients with available information; IPSS, International Prognostic Scoring System; SD, standard deviation.

Age (years), median (range)65 (31–90)61 (14–90)0·120
Male gender, n (%)40 (65·6)225 (60·3)0·435
Follow-up (months), median (range)44·7 (2–337)43·8 (0–295)0·970
AML transformation, n (%)12 (19·7)55 (14·7)0·323
Leucocytes (x 109/l), mean ± SD12·0 ± 11·412·2 ± 12·00·647
Haemoglobin (g/l), mean ± SD111 ± 22114 ± 240·795
Platelets (x 109/l) (N = 433)*, mean ± SD449·4 ± 319·3422·2 ± 366·10·259
% peripheral blast cells, mean ± SD0·3 ± 0·80·7 ± 2·10·924
JAK2 V617F, n (%)38 (62·3)226 (60·6)0·800
% JAK2 V617F allele burden (N = 241), mean ± SD43·3 ± 19·547·3 ± 23·60·464
MPL W515L/K (N = 429)*, n (%)6/60 (10·0)17/369 (4·6)0·085
Splenomegaly (N = 426)*, n (%)42/60 (70·0)280/366 (76·5)0·277
Abnormal karyotype (N = 204)*, n (%)9/31 (29·0)36/173 (20·8)0·309
Unfavourable karyotype (N = 202)*, n (%)2/31 (6·4)10/171 (5·8)0·896
IPSS, n (%)
Low16 (26·2)129 (34·6)0·181
Int-126 (42·6)107 (28·7)
Int-212 (19·7)85 (22·8)
High7 (11·5)52 (13·9)
Leukaemia transformation, n (%)12 (19·7)55 (14·7)0·323
Leukaemia-free survival (months), median (range)176·7 (95·8–257·6)254·3 (123·2–385·5)0·548
Overall survival (months), median (range)84·2 (81·3–133·4)107·4 (74·0–94·3)0·209

The complexity of pathogenesis in PMF is indicated by the overwhelming greater number of mutation abnormalities other than JAK2 V617F and MPL W515 (Vainchenker et al, 2011); it is therefore intuitive that genetic processes leading to leukaemia in PMF may also differ from those in PV and ET. These reasoning may help to explain why in this study we failed to confirm a leukaemia-predisposing role of the CC ERCC2 polymorphism as reported in PV and ET.


  1. Top of page
  2. Acknowledgements
  3. Authorship and disclosure
  4. References

This study was supported by a special grant from Associazione Italiana per la Ricerca sul Cancro-”AIRC 5 per Mille”- to AGIMM, ‘AIRC-Gruppo Italiano Malattie Mieloproliferative’ (#1005); for a description of the AGIMM project, see at and by FIRB grant #RBAP11CZLK (to AMV, MC).

Authorship and disclosure

  1. Top of page
  2. Acknowledgements
  3. Authorship and disclosure
  4. References

MCS, AS, FB, CM performed analysis; PG, GB, KZ, AR, AD, FC, MC, NC contributed patients; MCS, PG and AMV designed research and contributed to manuscript writing. The authors declare no potential conflict of interest.


  1. Top of page
  2. Acknowledgements
  3. Authorship and disclosure
  4. References
  • Allan, J.M., Smith, A.G., Wheatley, K., Hills, R.K., Travis, L.B., Hill, D.A., Swirsky, D.M., Morgan, G.J. & Wild, C.P. (2004) Genetic variation in XPD predicts treatment outcome and risk of acute myeloid leukemia following chemotherapy. Blood, 104, 38723877.
  • Cervantes, F., Dupriez, B., Pereira, A., Passamonti, F., Reilly, J.T., Morra, E., Vannucchi, A.M., Mesa, R.A., Demory, J.L., Barosi, G., Rumi, E. & Tefferi, A. (2009) New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood, 113, 28952901.
  • Guglielmelli, P., Barosi, G., Specchia, G., Rambaldi, A., Lo Coco, F., Antonioli, E., Pieri, L., Pancrazzi, A., Ponziani, V., Delaini, F., Longo, G., Ammatuna, E., Liso, V., Bosi, A., Barbui, T. & Vannucchi, A.M. (2009) Identification of patients with poorer survival in primary myelofibrosis based on the burden of JAK2V617F mutated allele. Blood, 114, 14771483.
  • Hernández-Boluda, J.-C., Pereira, A., Cervantes, F., Alvarez-Larrán, A., Collado, M., Such, E., Arilla, M.J., Boqué, C., Xicoy, B., Maffioli, M., Bellosillo, B., Marugán, I., Amat, P., Besses, C. & Guillem, V. (2012) A polymorphism in the XPD gene predisposes to leukemic transformation and new nonmyeloid malignancies in essential thrombocythemia and polycythemia vera. Blood, 119, 52215228.
  • Jones, A.V., Chase, A., Silver, R.T., Oscier, D., Zoi, K., Wang, Y.L., Cario, H., Pahl, H.L., Collins, A., Reiter, A., Grand, F. & Cross, N.C. (2009) JAK2 haplotype is a major risk factor for the development of myeloproliferative neoplasms. Nature Genetics, 41, 446449.
  • Pancrazzi, A., Guglielmelli, P., Ponziani, V., Bergamaschi, G., Bosi, A., Barosi, G. & Vannucchi, A.M. (2008) A sensitive detection method for MPLW515L or MPLW515K mutation in chronic myeloproliferative disorders with locked nucleic acid-modified probes and real-time polymerase chain reaction. The Journal of molecular diagnostics, 10, 435441.
  • Poletto, V., Rosti, V., Villani, L., Catarsi, P., Carolei, A., Campanelli, R., Massa, M., Martinetti, M., Viarengo, G., Malovini, A., Migliaccio, A.R. & Barosi, G. (2012) A3669G polymorphism of glucocorticoid receptor is a susceptibility allele for primary myelofibrosis and contributes to phenotypic diversity and blast transformation. Blood, 120, 31123117.
  • Tefferi, A., Pardanani, A., Gangat, N., Begna, K.H., Hanson, C.A., Van Dyke, D.L., Caramazza, D., Vannucchi, A.M., Morra, E., Cazzola, M., Pereira, A., Cervantes, F. & Passamonti, F. (2012) Leukemia risk models in primary myelofibrosis: an International Working Group study. Leukemia, 26, 14391441.
  • Vainchenker, W., Delhommeau, F., Constantinescu, S.N. & Bernard, O.A. (2011) New mutations and pathogenesis of myeloproliferative neoplasms. Blood, 118, 17231735.
  • Varricchio, L., Masselli, E., Alfani, E., Battistini, A., Migliaccio, G., Vannucchi, A.M., Zhang, W., Rondelli, D., Godbold, J., Ghinassi, B., Whitsett, C., Hoffman, R. & Migliaccio, A.R. (2011) The dominant negative beta isoform of the gluco-corticoid receptor is uniquely expressed in erythroid cells expanded from polycythemia vera patients. Blood, 118, 425436.