Common Genetic Variants of Fetal Hemoglobin Modify Hematological Phenotypes in MDS/MPN/Myeloma Patients Receiving Cytotoxic Drugs

    

    

    Figure 2: Analysis of minor allele f requencies of HbF genetic variants in hematological malignancies in comparison with wider British-European population genome
data using linear regression and t-test.
    

Discussion

Genetic studies of HbF can present opportunities for useful insight into hematological patterns of response to treatment and prognostic biomarkers. We report common fetal hemoglobin variants do modify hematological phenotypes in MPN/Myeloma/ MDS patients receiving cytotoxic drugs. HbF was variably expressed in our study group on HbF-inducing agents highlighting degree of genetic influence in treatment response. MDS on AZA group showed highest levels of baseline and induced HbF followed by MPN on HU and Myeloma on Lenalidomide groups respectively. Elevated baseline HbF was seen in 1 in 4 of our hematological malignancy pre-treatment cohort. MDS on AZA group had more than double mean induced HbF levels at 2.36% compared to MPN on HU at 1.06% and threetimes more than the Myeloma patients on Lenalidomide at 0.77%. Our HbF induction profile differs from SCD HbF induction profile studies which illustrate Pomalidomide (Lenalidomide analogue) HbF levels equivalent to or greater than levels seen with HU. [22,23] We also found that our MPN population on HU had less prevalence of HbF responders (28.6%) in comparison to SCD population on HU owing probably to the lower baseline HbF levels comparatively to begin with. [24,25] Nevertheless, it remains impossible to make full comparisons due to the differences in disease populations. The various hematological phenotypes such as anemia and macrocytosis observed in our study population reflect treatment and/or disease specific myelosuppression and stress erythropoiesis. Likewise, we did not include dosage of cytotoxic agent, as this is difficult to compare between the disease sub-group populations. Nevertheless, it is likely, that a dosage effect will be observed between cytotoxic agent and HbF level if the diseases are further studied individually. Similarly, more larger and focused disease phenotype studies should be undertaken to determine HbF quantitative trait loci effect and respective treatment dosage regimens on disease-specific clinical outcomes like achievement of disease control, need for packed RBC transfusions, normalization of hemoglobin concentration or platelet count for MPNs, myeloma response rate as well as event-free survival for MDS.

Whilst the influence of HbF genetic loci which modify hematological phenotype leading to favorable outcomes in SCD & β-thalassemia is well known, there is a distinct lack of comparable evidence evaluating HbF variant influence in patients with hematological malignancies. Lübbert et al. [4] however demonstrated the prognostic value of elevated pre-treatment HbF levels in predicting better outcome for MDS/AML patients receiving decitabine therapy. Developing on this, our study reveals the genotypic significance of HbF and effects on hematological traits amongst patients with hematological malignancy on HbF-inducing cytotoxic therapy thus indicating the potential for these HbF variants in prediction of treatment response. These findings within this small study are further contextually validated by Tapper et al. [26] Genome Wide Association Study (GWAS) of 3437 MPN cases which showed that, in addition to established MPN driver mutations like JAK2 and CALR which predispose to MPN, they also found surprisingly HBS1L-MYB variant rs9376092 as significantly associated with MPN, in particular JAK2V617F-positive ET cases. For JAK2V617F-negative cases, rs9376092 (HBS1L-MYB) achieved significance conditional on CALR and/or MPL mutations. [26] Unfortunately, we were not able to replicate and corroborate this exact Single Nucleotide Polymorphism (SNP) amongst our MPN group as we did not select this SNP to genotype from the outset of the study. However, we can of novel verify another HBS1L-MYB SNP rs9494142 as significantly associated with HbF in our MPN group on HU. Furthermore, in line with Tapper et al. [26] showing influence of HBS1L-MYB variant towards the development of ET amongst MPN patients, our study also highlights PLT involvement with association of increased PLT count and HbF levels (Beta = 0.35, p = 0.04, Table 2) amongst our MPN group on HU. Studies describing SCD HU-associated HbF induction involve numerous complex signal transduction pathways through key genomic variants BCL11A, HBS1L-MYB and SAR1. [27] In our study, a patient with MPN had an intriguing HbF induction to 25% while he was on HU. This patient had Type 2 Diabetes and was also on Metformin (now known to be a potent HbF inducer). [28] SNP assay for this patient revealed he was heterozygous for four HBS1L-MYB variants and homozygous for the mutant allele at the BCL11A (rs6545816) locus.

Thein et al. [11] identified genetic variants at three principal linkage disequilibrium blocks at HMIP blocks 1, 2, and 3 spanning approximately 79kb responsible for 19% of HbF trait variance. HMIP also has the most abundant pleiotropic effects on human hematological traits. [29] The most powerful of these effects are mostly attributable to HMIP-2 variants such as rs9494142, 3bp-del rs66650371 and rs6920211 as these variants reside within a core-enhancer for MYB, a major hematopoietic gene regulator which disrupt binding for key transcription factors like GATA. [30] Our study showed HMIP-2 variants rs9494142 and rs6920211 influencing increased baseline PLT count amongst MPN on HU group. These variants were also seen to affect different hematological parameters at different stages of treatment such as when these HMIP-2 variants switched influence of baseline PLT count to regulation of induced Hb levels 3-months post treatment initiation. Amongst, our total hematological malignancy group, HMIP-2 (rs6920211) was also associated with BILI whilst on HbF-inducing cytotoxic therapy.

HBS1L-MYB variants have been shown to be associated with baseline HbF in African American patients with sickle cell anemia. [14] We confirm another HBS1L-MYB variant rs6920211 as associated with baseline HbF levels amongst Myeloma patients. BCL11A variant rs1427407 was strongly associated with induced HbF levels in this group on Lenalidomide. BCL11A association in Myeloma is shown by previous work from Dulmovits et al. [22] and Kubi-Appiah et al. [31] in Pomalidomide-induced reactivation of HbF through BCL11A. Dulmovits et al. [22] also found HbF reactivation involving Ikaros, which is the degradation target in Myeloma treatment with Pomalidomide. Our study also showed rs1427407 (BCL11A) significantly associated with increased platelet count seemingly withstanding the thrombocytopenic tendency of Lenalidomide. Sub-groups with this variant should be stratified and evaluated for increased risk of thromboembolism amongst patients with Myeloma on Lenalidomide highlighting potential functional significance of rs1427407 (BCL11A) variant on hematological traits. BCL11A loci account for 15% of HbF variance in human populations. [12] Differences in BCL11A expression level of transcript have been recently implicated in disease biology of Acute Myeloid Leukemia and acute phase Chronic Myeloid Leukemia [32,33].

The XMN1-HGB2 (rs782144) locus within the β-globin gene cluster contributes 10% of HbF variance. [12] Although no significant HbF association was found, XMN1-HBG2 (rs782144) did influence various hematological parameters including increased markers of hemolysis e.g. LDH and reticulocytes amongst MPN patients on HU in our study. XMN1-HBG2 (rs782144) was also associated with reduction in PLT count amongst MDS patients on AZA making this a useful line of investigation for thrombocytopenia within this disease group.

Our study of novel shows that these major HbF-associated variants may also be implicated in the disease biology of patients with MPN/Myeloma/MDS evidenced by MAF discordance when compared to wider British/European population MAF data derived from 1000 genome project. Although our sample data is small, this fits and corroborates findings from Tapper et al. [26] MPN GWAS study showing HBS1L-MYB affecting MPN disease biology predisposing to the development of ET. The marker found associated in this paper (rs9376092) tracks the same causative genetic variant (3-bp deletion) as the markers used in the present study. [34] JAK2 is also known for its pleiotropic effect on various hematological parameters [6], as such, we analyzed the association between JAK2 status of MPN disease subtypes and influence on HbF levels but found no significant association. We did note however that JAK2V617F-positive ET patients had most elevated HbF levels among MPN disease sub-groups but this did not achieve statistical significance possibly due to our small sub-group sample size. Amongst MDS patients, a previous study had identified higher HbF levels amongst those with karyotypic abnormality. [35] We, however found no association between karyotype and HbF in MDS.

Whilst 89 patients is reasonable statistical study strength, disease sub-group analysis have smaller sample sizes and smaller numbers genotyped particularly for our MDS group as such it is possible that not all statistically significant associations could be identified within this study. Consequently, larger powered, prospective genotype-phenotype association studies are warranted. Despite our small sample size, the data validates previous studies and highlights that the common HbF variants influence HbF and hematological traits in MPN, Myeloma and MDS and suggests that these variants are involved in disease biology and treatment response for these conditions. Consequently, illustrating that HbF serves as a conduit bio-marker for the presence of genetic variants that influence the hematological response in these diseases.

Acknowledgements

We would like to thank The British Society of Haematology (BSH) for providing financial support for this study. Author NI was the recipient of a BSH Early Stage Research start-up grant and was a UK NIHR Academic Clinical Fellow in Haematology at the time of the study. This study also received an American Society of Hematology (ASH) Abstract Achievement Award.

We thank Dr Muzaffar Malik for statistical advice with the study. We thank Tim Futter for ethics and administrative support to this study at his time at King’s College London.

We would like to also thank the patients at the haematology clinic at Brighton & Sussex University Hospitals NHS Trust for their kind participation in this study. We also thank the staff at the BSUH haematology department for their help in facilitating this study.

Author Contribution

NI, TC, SLT, and SM conceived and designed the study. NI performed the research. NI, AP, HH, DJ, and TC recruited the patients. NI, HR and SM performed the genotyping. AB and HR performed the FACS. NI, AP, HH, DJ and DM curated the data. NI and SM performed the data and genetic analyses. NI wrote the manuscript. All authors reviewed the manuscript. TC, SLT and SM provided critical review for the manuscript and provided senior supervision the overall study.

Conflict of Interest

The authors declare no conflict of interest or competing financial interest.

References

1. Thein SL, Menzel S, Lathrop M, Garner C. Control of fetal hemoglobin: new insights emerging f Rom genomics and clinical implications. Hum Mol Genet. 2009; 18: R216-23.
2. Eridani S, Mosca A. Fetal hemoglobin reactivation and cell engineering in the treatment of sickle cell anemia. J Blood Med. 2011; 2: 23-30.
3. Bauer DE, Kamran SC, Orkin SH. Reawakening fetal hemoglobin: prospects for new therapies for the β – globin disorders. Blood. 2012; 120: 2945-53.
4. Lübbert M, Ihorst G, Sander PN, Bogatyreva L, Becker H, Wijermans PW et al. Elevated fetal haemoglobin is a predictor of better outcome in MDS/AML patients receiving 5- aza-2'-deoxycytidine (decitabine). Br J Haematol. 2017; 176: 609-17.
5. Ware RE, Eggleston B, Redding-Lallinger R, Wang WC, Smith-Whitley K, Daeschner C et al. Predictors of fetal hemoglobin response in children with sickle cellanemia receiving hydroxyurea therapy. Blood. 2002; 99: 10-4.
6. Chami N, Chen MH, Slater AJ, Eicher JD, Evangelou E, Tajuddin SM et al. Exome genotyping identifies pleiotropic variants associated with red blood cell traits. Am J Hum Genet. 2016; 99: 8-21.
7. Garner C, Tatu T, Reittie JE, Littlewood T, Darley J, Cervino S et al. Genetic inf luences on F cells and other hematologic variables: a twin heritability study. Blood. 2000; 95: 342-6.
8. Sankaran VG, Orkin SH. The switch f Rom fetal to adult hemoglobin. Cold Spring Harb Perspect Med. 2013; 3: a011643.
9. Tumburu L, Thein SL. Genetic control of erythropoiesis. Curr Opin Hematol. 2017; 24: 173-82.
10. Alter BP, Rosenberg PS, Day T, Menzel S, Giri N, Savage SA et al. Genetic regulation of fetal haemoglobin in inherited bone marrow failure syndromes. Br J Haematol. 2013; 162: 542-6.
11. Thein SL, Menzel S, Peng X, Best S, Jiang J, Close J et al. Intergenic variants of HBS1L-MYB are responsible for a major quantitative trait locus on chromosome 6q23 inf luencing fetal hemoglobin levels in adults. Proc Natl Acad Sci U S A. 2007; 104: 11346-51.
12. Menzel S, Garner C, Gut I, Matsuda F, Yamaguchi M, Heath S et al. A QTL inf luencing F cell production maps to a gene encoding a zinc-finger protein on chromosome 2p15. Nat Genet. 2007; 39: 1197-9.
13. Khanh T et al. The XmnI G γ polymorphism inf luences hemoglobin F synthesis contrary to BCL11A and HBS1L- MYB SNPs in a cohort of 57 β-thalassemia intermedia patients. Blood Cells Mol Dis. 2010; 45: 124-7.
14. Akinsheye I, Alsultan A, Solovieff N, Ngo D, Baldwin CT, Sebastiani P et al. Fetal hemoglobin in sickle cell anemia. Blood. 2011; 118: 19-27.
15. Bauer DE, Orkin SH. Update on fetal hemoglobin gene regulation in hemoglobinopathies. Curr Opin Pediatr. 2011; 23: 1-8.
16. Lettre G, Sankaran VG, Bezerra MA, Araújo AS, Uda M, Sanna S et al. DNA polymorphisms at the BCL11A, HBS1L-MYB, and beta-globin loci associate with fetal hemoglobin levels and pain crises in sickle cell disease. Proc Natl Acad Sci U S A. 2008; 105: 11869-74.
17. Banan M, Bayat H, Azarkeivan A, Mohammadparast S, Kamali K, Farashi S et al. The XmnI and BCL11A single nucleotide polymorphisms may help predict hydroxyurea response in Iranian β-thalassemia patients. Hemoglobin. 2012; 36: 371-80.
18. Campbell PJ, Green AR. The myeloproliferative disorders. N Engl J Med. 2006; 355: 2452-66.
19. Gryn J, Zeigler ZR, Shadduck RK, Lister J, Raymond JM, Sbeitan I et al. Treatment of myelodysplastic syndromes with 5-azacytidine. Leuk Res. 2002; 26: 893-7.
20. Holstein SA, McCarthy PL. Immunomodulatory drugs in multiple myeloma: mechanisms of action and clinical experience. Drugs. 2017; 77: 505-20.
21. Singh A, Xu YJ. The cell killing mechanisms of hydroxyurea. Genes (Basel). 2016; 7: 99.
22. Dulmovits BM, Appiah-Kubi AO, Papoin J, Hale J, He M, Al-Abed Y et al. Pomalidomide reverses -globin silencing through the transcriptional reprogramming of adult hematopoietic progenitors. Blood. 2016; 127: 1481- 92.
23. Meiler SE, Wade M, Kutlar F, Yerigenahally SD, Xue Y, Moutouh-de Parseval LA et al. Pomalidomide augments fetal hemoglobin production without the myelosuppressive effects of hydroxyurea in transgenic sickle cell mice. Blood. 2011; 118: 1109-12.
24. Chand AR, Xu H, Wells LG, Clair B, Neunert C, Spellman AE et al. Are there true non-responders to hydroxyurea in sickle cell disease? a multiparameter analysis. Blood. 2014; 124: 4073.
25. Green NS, Barral S. Genetic modifiers of HbF and response to hydroxyurea in sickle cell disease. Pediatr Blood Cancer. 2011; 56: 177-81.
26. Tapper W, Jones AV, Kralovics R, Harutyunyan AS, Zoi K, Leung W et al. Genetic variation at MECOM, tert, JAK2 and HBS1L-MYB predisposes to myeloproliferative neoplasms. Nat Commun. 2015; 6: 6691.
27. Pule GD, Mowla S, Novitzky N, Wiysonge CS, Wonkam A. A systematic review of known mechanisms of hydroxyurea-induced fetal hemoglobin for treatment of sickle cell disease. Expert Rev Hematol. 2015; 8: 669-79.
28. Zhang Y, Weiss M, Sumazin P, Sheehan VA. Metformin induces FOXO3- dependent fetal hemoglobin production in primary erythroid cells. Blood. 2016; 128.
29. Menzel S, Jiang J, Silver N, Gallagher J, Cunningham J, Surdulescu G et al. The HBS1L-MYB intergenic region on chromosome 6q23.3 inf luences erythrocyte, platelet, and monocyte counts in humans. Blood. 2007; 110: 3624-6.
30. Stadhouders R, Aktuna S, Thongjuea S, Aghajanirefah A, Pourfarzad F, van Ijcken W et al. HBS1L-MYB intergenic variants modulate fetal hemoglobin via long -range MYB enhancers. J Clin Invest. 2014; 124: 1699-710.
31. Appiah-Kubi AO, Blanc L, Singh SA, Didier S, Dsilva S, Chan KWH et al. Pomalidomide augments fetal hemoglobin production in primary erythroid cells by a novel mechanism modulating BCL11A but not KLF-1. Blood. 2013; 122: 314.
32. Tao H, Ma X, Su G, Yin J, Xie X, Hu C et al. BCL11A expression in acute myeloid leukemia. Leuk Res. 2016; 41: 71-5.
33. Yin J, Zhang F, Tao H, Ma X, Su G, Xie X et al. BCL11A expression in acute phase chronic myeloid leukemia. Leuk Res. 2016; 47: 88-92.
34. Menzel S, Rooks H, Zelenika D, Mtatiro SN, Gnanakulasekaran A, Drasar E et al. Global genetic architecture of an erythroid quantitative trait locus, HMIP- 2. Ann Hum Genet. 2014; 78: 434-51.
35. Craig JE, Sampietro M, Oscier DG, Contreras M, Thein S. Myelodysplastic syndrome with karyotype abnormality is associated with elevated F-cell production. Br J Haematol. 1996; 93: 601-5.