Lymphoma, Hodgkin, Classic
A number sign (#) is used with this entry because of evidence that susceptibility to classic Hodgkin lymphoma (CHL) can be conferred by variation in the KLHDC8B gene (613169) on chromosome 3p21.
DescriptionClassic Hodgkin lymphoma is a lymph node cancer of germinal center B-cell origin. Hodgkin lymphoma tumors consist of a minority of malignant cells, known as 'Reed-Sternberg' (RS) cells, mixed with reactive lymphocytes and other benign inflammatory cells. A defining feature of RS cells is the presence of 2 nuclei (summary by Salipante et al., 2009).
Clinical FeaturesManigand et al. (1964) described a brother and sister with Hodgkin disease and reviewed the literature on familial occurrence. In familial cases Vianna et al. (1974) found that the time-intervals between diagnoses were shorter than the age differences. Also the time-intervals were shorter for relatives living together than for those living apart. These findings suggested an environmental basis for familial occurrence. On the other hand, the similarity of Rye histologic type in relatives, regardless of proximity, suggested a genetic factor in host reactivity.
Fraumeni et al. (1975) described a kindred in which, in 1 sibship of 9 adults, 4 died of lymphocytic or histiocytic lymphomas and 1, a male, died of Waldenstrom macroglobulinemia complicated by adenocarcinoma of the lung. In the next generation, 1 person died of Hodgkin disease; 4 of 9 healthy persons had impaired lymphocyte transformation with phytohemagglutinin, and 3 of these had polyclonal elevation of IgM. Subsequent to the studies, adenocarcinoma of the lung developed in 1 of those with an immune defect, a woman, and her 3-year-old grandson developed lymphocytic leukemia. This was the first suggestion of a genetic or immunologic basis of lung adenocarcinoma.
Grufferman et al. (1977) found a 7-fold increased risk of Hodgkin disease in sibs under 45 years of age. Twelve of 13 sib pairs were sex-concordant. The series showed an excess of nodular sclerosis type and pairs showed an excess of concordance for this type. This type may be the form most likely to have an infectious basis (Cole et al., 1968). Apparently no excess sib risk exists for Hodgkin disease diagnosed after age 45. MacMahon (1966) suggested that the cause may differ in the young and old. In general, a primarily nongenetic basis is suggested by these findings. Conte et al. (1983) studied 4 families, each with 2 cases of Hodgkin disease of the nodular sclerosis type. All 8 patients showed B18 antigen. The affected persons included father-son, brother-sister, mother-son, and father-daughter pairs.
Salipante et al. (2009) studied a family in which the proband developed the nodular sclerosis type of classic Hodgkin lymphoma at age 39 years and 2 of her sibs also developed nodular sclerosis-type CHL as adults. Their mother died shortly after presenting with a mediastinal tumor, compatible with CHL in terms of its location, for which she declined further evaluation.
InheritanceBodmer (1986) emphasized that the sib pair method is an effective way to demonstrate linkage of a disease with HLA. It is necessary to look at only affected persons, if reduced penetrance is likely. The expected distribution of both marker phenotype 1, one marker-1 with the other marker-2, and both marker phenotype 2 is 1 in 4, 1 in 2, and 1 in 4. The distribution for Hodgkin disease in 32 sib pairs (both affected) was 16, 11 and 5--a difference from 8, 16, and 8 significant at the 0.005 level. Bodmer (1986) concluded from this that there must be a gene for Hodgkin disease in or near the HLA complex. He pointed out, furthermore, that rarity of familial clustering does not exclude an important genetic component. His argument was as follows: Assume that there is a gene d with a frequency of 1 in 1000, and that 100% of dd persons develop Hodgkin disease whereas only 5% of dD persons and 0% of DD persons develop Hodgkin disease. Under these circumstances, only about 1 of every 100 cases of Hodgkin disease will be a homozygote; the overwhelming majority will be heterozygotes. Under the terms of this model, the frequency of 2-child families with both affected can be calculated to be about 1 in 400,000. This value is calculated from the product of the proportion of 2-sib families that have dD x DD parents (about 1 in 250) and the proportion of 2-sib families in which both sibs inherit the gene (1 in 4) and express the disease phenotype (1 in 400). This model permits the possibility that in some populations the recessive gene may achieve considerably higher frequency through drift/founder effect, resulting in more impressive familial clustering.
Relatives of young adults with Hodgkin disease are at increased risk of Hodgkin disease, and, as indicated earlier, lines of evidence implicate both inheritance and environment. Mack et al. (1995) identified and followed 432 sets of twins affected by Hodgkin disease. The twins were identified through weekly advertisements between 1980 and 1992 requesting that 'twins with cancer' contact the group. The number of cases of Hodgkin disease observed before the age of 50 years in the healthy monozygotic and dizygotic cotwins of the patients with Hodgkin disease was compared with the number expected from national age-specific incidence rates. None of the 187 pairs of dizygotic twins became concordant for Hodgkin disease, whereas 10 of the 179 pairs of monozygotic twins did; in 5 of the 10 pairs, the second case appeared after the original ascertainment. During the observation period, 0.1 (monozygotic) and 0.1 (dizygotic) cases in the unaffected twins were expected. Mack et al. (1995) concluded that genetic susceptibility underlies Hodgkin disease in young adulthood.
In a study in Sweden, Shugart et al. (2000) found a difference between the age of onset in parents and offspring who were affected with Hodgkin lymphoma, indicating anticipation. Anticipation has been found also in non-Hodgkin lymphoma (605027). They estimated the heritability of Hodgkin lymphoma in the Swedish population to be approximately 28.4%.
PathogenesisA link between Hodgkin disease and Epstein Barr virus (EBV) was first hinted at through epidemiologic studies (Munoz et al., 1978; Mueller et al., 1989). A possible role for EBV in the pathogenesis of Hodgkin disease was suggested by direct detection of EBV DNA in tumor biopsies in up to 50% of Hodgkin disease cases (Weiss et al. (1987, 1989); Wu et al., 1990) and the finding of a consistent pattern of EBV latent gene transcription in EBV-positive Reed-Sternberg cells (Pallesen et al., 1991; Deacon et al., 1993), as demonstrated by in situ hybridization.
As reviewed by Ambinder (2003), refinements in molecular techniques remove all doubts about the origin and character of the Reed-Sternberg cells in classic Hodgkin disease: they are neoplastic cells of B cell lineage. They differ from most other B cell-lineage lymphomas in that, although they carry immunoglobulin genes that have undergone the rearrangements that should allow them to produce immunoglobulins, they do not express or produce immunoglobulins. Infectious mononucleosis has long been recognized as a risk factor for Hodgkin lymphoma. Hjalgrim et al. (2003) presented evidence for a causal association between infectious mononucleosis-related EBV infection and the EBV-positive subgroup of Hodgkin lymphoma in young adults. Ambinder (2003) noted as a puzzle the fact that both EBV-positive and EBV-negative Hodgkin lymphoma can occur within a family. He suggested that it is not appropriate to presume that EBV-positive and EBV-negative Hodgkin lymphomas are distinct entities. There appear to be several paths to Hodgkin lymphoma; they all must involve signals that allow lymphocytes of B cell lineage that do not express immunoglobulin to escape apoptosis.
Tecchio et al. (2007) found increased serum B-lymphocyte stimulator (BLYS; 603969) in 87 patients with Hodgkin lymphoma compared to controls. Among those with Hodgkin lymphoma, higher BLYS levels were associated with more advanced disease stage, systemic symptoms, extranodal involvement, and age. Five-year failure-free survival was higher in patients with serum BLYS below the median. Statistical analyses confirmed the prognostic significance of serum BLYS in Hodgkin lymphoma. The findings confirmed the role of B cell signaling in Hodgkin disease.
MappingPerforming molecular typing of HLA class II loci, Klitz et al. (1994) found an overall association of the nodular sclerosing (NSHD) group with the HLA class II region. The results of the study suggested that susceptibility to NSHD is influenced by more than 1 locus within the class II region. Haplotypes and alleles defined for the 4 loci typed (DRB1, DQA1, DQB1, and DPB1; see 142857 and 142858) were present in both excess and deficit in the sample of NSHD (N = 155; age range 9-62 years). Klitz et al. (1994) suggested that the data explain why earlier studies showed HLA linkage but not association and that they substantiate the specific involvement of the immune system in certain neoplastic diseases.
Through a literature search, Shugart and Collins (2000) identified 59 nuclear families with Hodgkin disease. Affected individuals in these families had been HLA haplotyped. To evaluate the role of HLA-linked determinants of Hodgkin disease and investigate the mode of inheritance of familial Hodgkin disease, a combined segregation and linkage analysis was performed, together with a semiparametric analysis of sib pairs. Among single-locus models, a general single-locus model showed no improvement over an additive model. However, a 2-locus additive model gave the best overall fit and a lod score of 3.55. Shugart and Collins (2000) concluded that Hodgkin disease is most likely determined by both an HLA-associated major gene and other non-HLA genetic factors together with environmental effects.
Goldin et al. (2005) performed a genomewide linkage screen in 254 individuals from 44 high-risk Hodgkin lymphoma families, among which there were 95 cases of Hodgkin lymphoma and 4 cases of non-Hodgkin lymphoma. The strongest linkage was on chromosome 4p near the marker D4S394, with a lod score of 2.6. The mean identity-by-descent sharing among 35 affected sib pairs was 72% in this region.
Population GeneticsFerraris et al. (1997) quoted an estimate of 4.5% as the proportion of all cases of Hodgkin disease represented by familial Hodgkin disease. They reviewed 28 articles on familial Hodgkin disease, published between 1972 and 1995, with detailed analysis of data from 18 papers, reporting on 328 patients. The male-to-female ratio was 1.5, similar to that reported for sporadic Hodgkin disease. A significant difference was found between sporadic and familial Hodgkin disease in the age at diagnosis: only 1 major peak between 15 and 34 years was present in the group of patients with familial Hodgkin disease. For the sporadic disease, the age of onset is characteristically bimodal, with a peak in the late twenties and another after the age of 50. A notable exception is Japan, where the young-adult incidence peak of Hodgkin disease is barely detectable.
CytogeneticsComparative genomic hybridization studies showed gains in chromosome region 2p as the most common imbalance in classic Hodgkin lymphoma. The minimal region of gain contained 2 candidate oncogenes, REL (164910) and BCL11A (606557). Martin-Subero et al. (2002) studied 44 primary cases of classic Hodgkin disease by combined immunophenotyping and interphase cytogenetics. A median 2p13 copy number above the tetraploid range was detected in 24 (55%) cases. One case displayed selective amplification of the REL locus not affecting BCL11A; 2 other cases showed evidence of breakpoints in the region spanned by the REL probe. Martin-Subero et al. (2002) interpreted these data as indicating that REL rather than BCL11A may be the target of the 2p13 alterations in classic Hodgkin lymphoma.
In a family with the nodular sclerosis type of classic Hodgkin lymphoma (CHL), Salipante et al. (2009) identified a t(2,3)(q11.2;p21.31) translocation. Analysis of breakpoints by FISH and sequencing revealed that the chromosome 3 breakpoint resided in the first intron of the KLHDC8B gene (613169), whereas the chromosome 2 breakpoint was not located near known or hypothetical genes.
To identify novel fusion transcripts resulting from translocations, Steidl et al. (2011) investigated 2 Hodgkin lymphoma cell lines by whole-transcriptome paired-end sequencing and showed a highly expressed gene fusion involving the major histocompatibility complex (MHC) class II transactivator CIITA (MHC2TA; 600005) in KM-H2 cells. In a subsequent evaluation of 263 B-cell lymphomas, Steidl et al. (2011) also demonstrated that genomic CIITA breaks are highly recurrent in primary mediastinal B-cell lymphoma (38%) and classical Hodgkin lymphoma (15%). Furthermore, they found that CTIIA is a promiscuous partner of various in-frame gene fusions, and reported that these gene alterations impact survival in primary mediastinal B-cell lymphoma. As functional consequences of CIITA gene fusions, Steidl et al. (2011) identified downregulation of surface HLA class II expression and overexpression of ligands of the receptor molecule programmed cell death-1 (CD274/PDL1, 605402 and CD273/PDL2, 605723). These receptor-ligand interactions have been shown to impact antitumor immune responses in several cancers, whereas decreased MHC class II expression has been linked to reduced tumor cell immunogenicity. Thus, Steidl et al. (2011) concluded that recurrent rearrangements of CIITA may represent a novel genetic mechanism underlying tumor-microenvironment interactions across a spectrum of lymphoid cancers.
Molecular GeneticsIn a family with the nodular sclerosis type of classic Hodgkin lymphoma in which affected individuals carried a t(2,3)(q11.2;p21.31) translocation that disrupted the KLHDC8B gene, Salipante et al. (2009) found that family members with the translocation expressed only about half as much KLHDC8B as those without the translocation and had correspondingly reduced expression of protein. Salipante et al. (2009) then analyzed the KLHDC8B gene in 52 probands from families with 2 or more individuals with CHL and identified a SNP in the 5-prime UTR (28C-T; 613169.0001) that was present in probands from 3 (5.8%) of 52 families compared with 4 (1.3%) of 307 controls (odds ratio, 4.64; 95% confidence interval, 1.01-21.4). The variant segregated with disease and was associated with a reduction in translation of mRNA. In addition, Reed-Sternberg cells from 1 of 3 informative sporadic CHL patients showed loss of heterozygosity for markers within and flanking the KLHDC8B gene on chromosome 3p21.31.
Lamprecht et al. (2010) found that Reed-Sternberg cells in Hodgkin lymphoma demonstrated upregulation of CSF1R (164770) and CSF1 (120420) mRNA and constitutive activation of CSF1R, which correlated with increased proliferation of the Reed-Sternberg cells. Non-Hodkgin cell lines did not express either gene, suggesting that the expression in Reed-Sternberg cells was aberrant. Analysis of CSF1R transcripts in Reed-Sternberg cells showed use of an alternative transcription start site located about 6.2-kb upstream of the normal myeloid transcription start site: this sequence corresponded to a long terminal repeat (LTR) of the MaLR THE1B family. LTRs derived from ancient retroviral infections have accumulated in the mammalian genome, and mammalian organisms have devised a number of surveillance mechanisms to silence these elements early in development, usually by DNA methylation. The LTR region was found to contain a number of putative binding sites for transcription factors (i.e., NFKB; 164011) that were expressed in the Reed-Sternberg cells. Further studies indicated that the LTR is normally repressed by epigenetic methylation, and that Reed-Sternberg cells had lost this methylation. In addition, nearly all Reed-Sternberg cells studies had lost expression of the transcriptional repressor CBFA2T3 (603870). LTR-driven CSF1R transcripts were also found in anaplastic large cell lymphoma. Lamprecht et al. (2010) suggested that inhibition of CSF1R signaling may be of therapeutic value in Hodgkin lymphoma.