Correspondence
Elevated incidence of chronic myeloid leukaemia in immunosuppressed solid organ transplant recipients
The BCR-ABL1 fusion gene is considered to be the molecular cause of leukaemic transformation in chronic myeloid leukaemia (CML). However, low levels of BCR-ABL1 transcripts may also be detectable sporadically in healthy individuals (Biernaux et al, 1995; Bose et al, 1998). The acquisition of additional genetic modifications or alterations in immune surveillance may be necessary to produce transformation to CML. Therefore, a normal immune system may effectively control the randomly acquired BCR-ABL1 transcripts. Interestingly, chronic immunosuppression has not yet been clearly identified as a risk factor for CML. Here, we report a higher incidence of CML in chronically immunosuppressed solid organ transplant (SOT) recipients at one institution. Cases of CML following SOT were identified, based on outpatient International Classification of Diseases (ICD)-9 diagnosis codes for adult patients undergoing SOT between 1 January 1 2000 and 31 December 2011, with a subsequent ICD-9 diagnosis code for CML. This information was crossreferenced with aggregate data for the institution obtained from the Organ Procurement and Transplant Network database as of May 10, 2013. Details on immunosuppression, diagnosis of CML, treatment and response were obtained from medical records. The age-adjusted incidence of CML in this cohort of patients was calculated and compared to the incidence in the age- and gender-matched US population based on the Poisson distribution. Also, the incidence of all secondary malignancies excluding non-melanoma skin cancer was calculated for this cohort of patients and compared to the incidence of CML in this cohort. A total of 3089 adult SOT were performed between 1 January 2000 and 31 December, 2011. Five of these recipients developed CML, 1–10 years after transplant (Table I). No additional cytogenetic abnormalities besides the Philadelphia chromosome t(9;22) were identified. The total follow-up time was 13 264 person-years. The unadjusted incidence of CML was 377 per 100 000 person-years, and the ageadjusted incidence rate was 347 per 100 000 person-years. This incidence significantly exceeds the US incidence of 16 cases per 100 000 person years (P < 00001) in an age- and gender-matched population. Also, in this cohort of 3089 patients, 310 patients had a subsequent diagnosis of any malignancy, excluding non-melanoma skin cancer. This finding represents an overall secondary cancer incidence of 2337 cases/100 000 person years as compared to the cancer incidence in the general US population of approximately 460 cases/100 000 person years (Howlader et al, 2009) ª 2014 John Wiley & Sons Ltd British Journal of Haematology, 2014, 166, 612–628
The frequency of CML in SOT recipients at our institution was approximately 20-times higher than the reported incidence in the United States (Howlader et al, 2009). The overall frequency of any malignancy after SOT was approximately 5 times higher than in the general US population, suggesting that the risk of CML is particularly increased out of proportion to the overall increase in secondary malignancy for SOT patients. Importantly, in this series of patients we did not observe any difference in response to treatment with imatinib as compared to immunocompetent patients suggesting these patients can be managed according to standard treatment algorithms. Single cases of CML in SOT recipients have been previously reported (le Coutre et al, 2010), but this is the first estimate of incidence. It remains to be determined to what extent factors, such as type of immunosuppression, HLA-type and latency between transplantation and diagnosis of CML, may contribute to this phenomenon. Evidence exists that immune surveillance and response play important roles in developing CML. Certain human leucocyte antigen (HLA) types seem to have a protective effect against CML. HLA-A3 and HLA-B8 have been shown to have binding motifs for BCR-ABL1 encoded peptides that can elicit an immune response (Bocchia et al, 1995; Clark et al, 2001). A registry-based study revealed that patients with HLA-B8 with or without coexpression of HLA-A3 had a lower incidence of CML, suggesting this immune response has a protective effect (Posthuma et al, 1999). While this does not prove the role of native immune surveillance in preventing the development of CML, it provides evidence that an immune response can be protective. Immune suppression may thus hinder this early immune response and allow propagation of the leukaemic clone. In the US population HLA-B8 is expressed in 128% of African Americans and 225% of Caucasians (Gonzalez-Galarza et al, 2011). Information on HLA typing was available for four of five patients in our study, none of whom had HLA-B8 expression. However, an association cannot be established within our cohort because of such small numbers. Alternatively, immunosuppressive agents may promote tumourigenesis. Calcineurin inhibitors, including ciclosporin and tacrolimus are associated with increased risk of secondary malignancies not only through immune suppression, but possibly through direct promotion of tumourigenesis (Hojo et al, 1999; Dantal & Soulillou, 2005). In contrast, the mammalian target of rapamycin (mTOR) inhibitors, including sirolimus, have been shown to have an antitumour effect (Hojo et al, 1999). However, we did not observe a specific association with the type of immunosuppression and the development of CML. 619
Correspondence Table I. Characteristics of patients with CML following SOT.
Patient
Age (years)/ gender
Transplant type/ primary disease
1
52/male
2
48/female
3
50/male
Liver/hepatocellular carcinoma Kidney and pancreatic islet cell/type I diabetes mellitus and diabetic nephropathy Lung transplant/ pulmonary sarcoidosis
4
59/male
5
77/female
Kidney/polycystic kidney disease Kidney/polycystic kidney disease and hypertensive nephropathy
Immunosuppression at the time of CML diagnosis
Time from transplant to CML diagnosis (years)
CML phase at diagnosis
TKI used
Best response to TKI
Time to Best Response (months)
Prednisone, tacrolimus Prednisone, tacrolimus
4
Chronic
Imatinib
CMR
18
6
Chronic
Imatinib
CCyR
6
Prednisone, tacroliums, azathioprine Sirolimus
15
Chronic
Imatinib
CMR
22
10
Chronic
Imatinib
CCyR
6*
1
Chronic
Imatinib
HR
1†
Prednisone, tacrolimus, mycophenolate
CML, chronic myeloid leukaemia; SOT, solid organ transplantation; TKI, tyrosine kinase inhibitor; CMR, complete molecular response; CCyR, complete cytogenetic response; HR, haematological response. *The patient died from septic shock. †The patient completed only 1 month of therapy before dying from allograft failure unrelated to CML.
In summary, our data indicate an elevated risk for CML in SOT patients. This finding needs to be validated in a larger cohort of patients to exclude the possibility of statistical clustering. If increased CML risk in SOT patients is confirmed in larger studies it may have significant implications for the clinical management of SOT patients with a consideration of early molecular testing for CML in the appropriate clinical scenario.
Conflicts of interest The authors have no conflicts of interest to disclose. Asha Dhanarajan1,2 Jack W. Hsu2 Philipp le Coutre3 John R. Wingard2 Myron Chang4 Maxim Norkin2 1
Acknowledgments
Division of Hematology/Oncology, Department of Medicine, Loyola
This project was supported in part by the University of Florida Clinical and Translational Science Institute.
University Medical Center, Maywood, IL, 2Division of Hematology/ Oncology, Department of Medicine, University of Florida, Gainesville, FL, USA, 3Charite Universit€atsmedizin Berlin, Berlin, Germany and 4
Department of Biostatistics, University of Florida, Gainesville, FL, USA
Author contributions
E-mail:
[email protected]
All authors contributed extensively to the work presented in this paper. AD and MN performed the data collection and drafted the manuscript. MC performed the statistical analysis. AD, JH, PLC, JW and MN all contributed equally to the conception, design, and analysis of this project as well as to the manuscript revisions at all stages.
References Biernaux, C., Loos, M., Sels, A., Huez, G. & Stryckmans, P. (1995) Detection of major BCR-ABL gene expression at a very low level in blood cells of some healthy individuals. Blood, 86, 3118–3122.
620
Keywords: chronic myeloid leukaemia, immunosuppression, incidence, solid organ transplantation First published online 4 April 2014 doi: 10.1111/bjh.12885
Bocchia, M., Wentworth, P.A., Southwood, S., Sidney, J., McGraw, K., Scheinberg, D.A. & Sette, A. (1995) Specific binding of leukemia oncogene fusion protein peptides to HLA class I molecules. Blood, 85, 2680– 2684.
Bose, S., Deininger, M., Gora-Tybor, J., Goldman, J.M. & Melo, J.V. (1998) The presence of typical and atypical BCR-ABL fusion genes in leukocytes of normal individuals: biologic significance and implications for the assessment of minimal residual disease. Blood, 92, 3362–3367.
ª 2014 John Wiley & Sons Ltd British Journal of Haematology, 2014, 166, 612–628
Correspondence Clark, R.E., Dodi, I.A., Hill, S.C., Lill, J.R., Aubert, G., Macintyre, A.R., Rojas, J., Bourdon, A., Bonner, P., Wang, L., Christmas, S., Travers, P., Creaser, C., Rees, R. & Madrigal, J.A. (2001) Direct evidence that leukemia cell present HLAassociated immunogenic peptides derived from the bcr-abl b3a2 fusion protein. Blood, 98, 2887–2893. le Coutre, P., Reinke, P., Neuhaus, R., Trappe, R., Ringel, F., Lalancette, M., Hemmati, P., D€ orken, B. & Daniel, P. (2010) BCR-ABL positive cells and chronic myeloid leukemia in immune suppressed organ transplant recipients. European Journal of Haematology, 84, 26–33. Dantal, J. & Soulillou, J. (2005) Immunosuppressive drugs and the risk of cancer after organ
transplantation. New England Journal of Medicine, 352, 1371–1373. Gonzalez-Galarza, F.F., Christmas, S., Middleton, D. & Jones, A.R. (2011) Allele frequency net: a database and online repository for immune gene frequencies in worldwide populations. Nucleic Acid Research, 39, D913–D919. Hojo, M., Morimoto, T., Maluccio, M., Asano, T., Morimoto, K., Lagman, M., Shimbo, T. & Suthanthiran, M. (1999) Cyclosporine induces cancer progression by a cell-autonomous mechanism. Nature, 397, 530–534. Howlader, N., Noone, A.M., Krapcho, M., Neyman, N., Aminou, R., Waldron, W., Altekruse, S.F., Kosary, C.L., Ruhl, J., Tatalovich, Z., Cho, H., Mariotto, A., Eisner, M.P., Lewis, D.R.,
Chen, H.S., Feuer, E.J. & Cronin, K.A. (eds). 2009 SEER Cancer Statistics Review, 1975-2009 (Vintage 2009 Populations). National Cancer Institute, Bethesda, MD. http://seer.cancer.gov/ csr/1975_2009_pops09/, based on November 2011 SEER data submission, posted to the SEER web site, April 2012. Posthuma, E.F.M., Falkenburg, J.H.F., Apperley, J.F., Gratwohl, A., Roosnek, E., Hertenstein, B., Schipper, R.F., Schreuder, G.M.T., D’Amaro, J., Oudshoorn, M., vBiezen, J.H., Hermans, J., Willemze, R. & Niederwieser, D. (1999) HLA-B8 and HLA-A3 coexpressed with HLA-B8 are associated with a reduced risk of the development of chronic myeloid leukemia. Blood, 93, 3863–3865.
The frequency of TP53 gene defects differs between chronic lymphocytic leukaemia subgroups harbouring distinct antigen receptors
In chronic lymphocytic leukaemia (CLL) tumour initiation and progression are linked to the interplay between cellextrinsic and cell-intrinsic mechanisms. Cell-extrinsic mechanisms relate to the crosstalk of CLL cells with their microenvironment through various receptors, including the B-cell receptor (BcR). The pivotal role of the BcR is supported by the biased immunoglobulin gene usage in CLL; the discrimination of CLL into two subsets with different prognoses based on the somatic hypermutation (SHM) status of the immunoglobulin heavy variable (IGHV) genes; the existence of so called stereotyped subsets with highly similar BcR immunoglobulins showing remarkable clinico-biological similarities; and, the clinical efficacy of BcR signalling inhibitors (Stamatopoulos et al, 2007; Stevenson et al, 2011). Cellintrinsic mechanisms implicated in CLL pathobiology include more general cancer-associated mechanisms e.g. activation of oncogenes and/or elimination of tumour suppressor genes. Amongst these, dysregulation of the TP53 gene – i.e. deletions of the TP53 locus at 17p13.1 [del(17p)] and mutations in the TP53 gene – represent the most adverse prognostic factor in CLL (Pospisilova et al, 2012). Whether microenvironmental stimulation and cancer-associated gene defects drive CLL independently or mutually cooperate remains unresolved. To obtain insight into this issue, we investigated the frequency of TP53 defects in relation to particular immunogenetic features, especially IGHV gene usage and BcR immunoglobulin stereotypy. To this end, we analysed patients from a large multi-centre series with available immunogenetic and TP53 mutation data. As TP53 defects are significantly enriched among patients with ª 2014 John Wiley & Sons Ltd British Journal of Haematology, 2014, 166, 612–628
unmutated IGHV (U-CLL), we restricted the analysis to U-CLL, moreover, we included cases utilizing the IGHV3-21 gene, which is associated with adverse prognosis regardless of IGHV mutational status (Bomben et al, 2009). This intentional selection resulted in a cohort comprising of 1173 patients (Table I). The detailed description of the methods employed in this study are provided in the Supplementary methods (Data S1). Overall, 1188 productive IGHV-IGHD-IGHJ gene rearrangements were detected. Due to our selection criteria 1146/1188 (965%) rearrangements were unmutated [98% germline identity (GI) cut-off value]. The remaining 42/ 1188 (35%) rearrangements were IGHV-mutated (all IGHV3-21). Three hundred anf forty-four cases were assigned to major stereotyped subsets (Table SI) (Agathangelidis et al, 2012). TP53 mutations were detected in 211/1173 patients (179%) and information regarding del(17p) was available in 977/1173 patients (83%). Thus, the overall frequency of TP53 defects, including TP53 mutations and del(17p), was 209% (204/977), reflecting the adverse composition of our cohort. Focusing on the distribution of TP53 defects according to SHM load, we observed a higher frequency in cases expressing truly unmutated IGHV genes (GI = 100%; 188/866 cases, 217%) compared to cases with minimally/borderline mutated IGHV genes (GI = 98–999%; 43/280 cases, 154%; P = 0021). This was partly attributed to a high number of IGHV3-21 cases with minimally/borderline mutated status (41/280), in which TP53 defects were rare – only 4% (4/100) 621
