Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Oct 1.
Published in final edited form as: Br J Haematol. 2016 Jun 22;175(1):87–101. doi: 10.1111/bjh.14199

Multiple myeloma and family history of lymphohaematopoietic cancers: Results from the International Multiple Myeloma Consortium

LH Schinasi 1, EE Brown 2, NJ Camp 3, SS Wang 4, JN Hofmann 5, BC Chiu 6, L Miligi 7, L Beane Freeman 5, S de Sanjose 8, L Bernstein 4, A Monnereau 9,10, J Clavel 9,10, G Tricot 11, D Atanackovic 3, P Cocco 12, L Orsi 9,10, JA Dosman 13, JR McLaughlin 14, MP Purdue 5, W Cozen 15, JJ Spinelli 16, AJ De Roos 1
PMCID: PMC5035512  NIHMSID: NIHMS792103  PMID: 27330041

Summary

Family clusters of multiple myeloma (MM) suggest disease heritability. Nevertheless, patterns of inheritance and the importance of genetic versus environmental risk factors in MM aetiology remain unclear. We pooled data from eleven case-control studies from the International Multiple Myeloma Consortium to characterize the association of MM risk with having a first-degree relative with a history of a lympho-haematapoietic cancer. Unconditional logistic regression models, adjusted for study, sex, age and education level, were used to estimate associations between MM risk and having a first-degree relative with a history of non-Hodgkin lymphoma, Hodgkin lymphoma, leukaemia or MM. Sex, African American race/ethnicity and age were explored as effect modifiers.

A total of 2,843 cases and 11,470 controls were included. MM risk was elevated in association with having a first-degree relative with any lympho-haematapoietic cancer (Odds Ratio (OR)= 1.29, 95% Confidence Interval (CI): 1.08-1.55). The association was particularly strong for having a first-degree relative with MM (OR= 1.90, 95% CI: 1.26-2.87), especially among men (OR= 4.13, 95% CI: 2.17-7.85) and African Americans (OR= 5.52, 95% CI: 1.87-16.27).

These results support the hypothesis that genetic inheritance plays a role in MM aetiology. Future studies are warranted to characterize interactions of genetic markers with environmental exposures.

Keywords: Multiple myeloma, epidemiology, lympho-haematopoietic malignancies, family history, lymphoma

Introduction

Multiple myeloma (MM) is a rare plasma cell neoplasm with a high mortality rate. Its incidence is highest in Western countries, and lowest in parts of Asia (Ferlay, et al 2013). The aetiology of this cancer remains poorly understood. Several case-control studies have reported associations between family history of any cancer (Bourguet, et al 1985) or family history of MM (Eriksson and Hallberg 1992) and increased MM risk. In a large, Swedish population-based case-control study, first-degree relatives of MM patients had over twice the risk of developing MM, compared to first-degree relatives of controls (Kristinsson, et al 2009a). Familial aggregation of MM and of MM with the premalignant state, monoclonal gammopathy of undetermined significance (MGUS), has also been documented (Bizzaro and Pasini 1990, Coleman, et al 2009, Greenberg, et al 2012, Koura and Langston 2013, Ogmundsdottir, et al 2005, Shoenfeld, et al 1982, Steingrimsdottir, et al 2011). In a large, population-based study in Sweden, first-degree relatives of MGUS patients had a nearly 3 times the risk of developing MM compared to first-degree relatives of controls (Risk ratio: 2.9, 95% CI: 1.9-4.3) (Landgren, et al 2009).

This familial clustering suggests that the aetiology of MM involves a genetic component. Genome-wide association studies have also demonstrated genetic inheritance (Broderick, et al 2012, Chubb, et al 2013). Also lending support to hypotheses of genetic inheritance are risk factors for MM, including male sex and African American race (Vangsted, et al 2012). Environmental exposures and lifestyle factors, including farming occupation and obesity, are also suspected to be associated with MM (Alexander, et al 2007). Familial clustering in close relatives could point to shared non-genetic risk factors, in addition to a genetic predisposition.

It is generally accepted that MM is a complex disease, and that multiple genes and environmental factors contribute to its aetiology. Understanding the configuration of risk in families may help to identify specific modes of inheritance. Further, examination of family history of cancer within subgroups, such as race/ethnicity or sex, may help to identify populations for which genetic inheritance is of particular importance.

With a few exceptions (Kristinsson, et al 2009a, Landgren, et al 2006), most epidemiological investigations of the association between family history of cancer and MM risk have been based on small samples, leading to low statistical power to detect patterns, especially within population subgroups. To overcome this limitation, we pooled data from 11 case-control study members of the International Multiple Myeloma Consortium to characterize the association of MM risk with family history of non-Hodgkin lymphoma (NHL), Hodgkin lymphoma, leukaemia and MM. The large pooled data set allowed investigation of risk of MM in association with family history of MM and other lymphohaematopoietic (LH) cancer types, in association with LH cancer history in specific familial relations, and these associations within subgroups. We investigated effect modification of the overall associations by sex and age. In addition, this is one of the first analyses to investigate effect modification by race/ethnicity.

Materials and Methods

Data source and case definitions

Data from 11 case-control studies included in the International Multiple Myeloma Consortium (http://epi.grants.cancer.gov/Consortia/single/immc.html)(Andreotti, et al 2013) that had information on family history of LH cancer (leukaemia, NHL, Hodgkin lymphoma, and MM) were included.

Details of the studies are given in Table I. The studies were conducted in the United States (US) [Connecticut Study (Purdue, et al 2011), Molecular and Genetic Epidemiology (iMAGE) Study of Myeloma (VanValkenburg et al 2016), Los Angeles County Case Control Multiple Myeloma Study (LACCMM; Nuyujukian, et al 2014), Nebraska Study (Zahm, et al 1992), Population Health Study (PopHealth; Baris, et al 2004), University of Southern California (USC; Cozen, et al 2006), and Utah Study (Camp et al 2008)] (, ); Canada (Canadian Study; Ghosh, et al 2011, McDuffie, et al 2009) and Europe [L'Etude des Facteurs environnemen-taux et genétique des Lymphomes de l'Adulte (ENGELA; Orsi, et al 2009), Epilymph (Casey, et al 2006, Perrotta, et al 2012), Italian Multicentre (Miligi, et al 1999)]. Nearly all of the studies included both men and women, with the exception of the Canadian (Ghosh, et al 2011, McDuffie, et al 2009) and Connecticut (Purdue, et al 2011) studies, which included only men and only women, respectively. Most of the studies used population-based controls. The multi-site Epilymph study from Europe included population-based controls at the German and Italian sites, and hospital-based controls at the Irish, French, Spanish and Czech Republic sites (Casey, et al 2006, Perrotta, et al 2012). The French study, ENGELA (Orsi, et al 2009), enrolled hospital-based controls. The Utah study included spouses of cases as controls, as well as population-based controls. Most studies matched controls to cases based on age, sex, geographic region and/or vital status.

Table I.

Individual studies contributing to the pooled analysis

Cases
 Controls
Study (reference) Location Year Family history ascertainment N Age at diagnosis, years: mean (SD) Female N (%) N Age at enrolment, years: mean (SD) Female N (%) Source Matching criteria
Canadian Ghosh et al (2011); McDuffie et al (2009) Canada 1991-1994 Self-administered postal questionnaire;
Open-ended responses for cancer.		 342 64.7 (11.1) 0 1506 54.1 (16.4) 0 P Frequency matched to the age distribution of the entire case group of HL, MM, NHL and STS cases within each province
Connecticut Study Purdue et al (2011) New Haven, CT, USA 1996-2000 In person interviews
Open ended responses for cancer.		 183 66.4 (11.5) 183 (100) 716 61.4 (14.3) 716 (100) P Frequency matched to NHL cases for a parallel study
L'Etude des Facteurs environnementaux et genétique des Lymphomes de l'Adulte (ENGELA) Orsi et al (2009) France 2000-2004 Self-administered standardized questionnaires
Prompted for HL, lymphoma, MM, lymphoid leukaemia, myeloid leukaemia, or leukaemia.		 108 59.6 (9.1) 52 (48.2) 478 58.8 (9.0) 165 (34.5) H Individually matched on centre, sex, age
Epilymph Casey et al (2006); Perrotta, et al (2012) Italy, Spain, Germany, France, Ireland, Czech Republic 1998-2004 Questionnaire administered by a locally trained interviewer
Open ended responses for cancer.		 278 62.4 (12.1) 120 (43.2) 2465 56.1 (16.0) 1143 (46.4) P in Germany and Italy; H in Ireland, France, Spain, and the Czech Republic Frequency matched by sex, age intervals, residence area
Molecular and Genetic Epidemiology (iMAGE) Study of Myeloma Van Valkenburg et al (2016) University of Alabama at Birmingham, AL 2009-2013 Prompted for leukaemia, HL, NHL, MGUS, MDS and select solid tumours; open-ended responses for other cancers 259 59.9 (9.3) 117 (45.2) 461 62.5 (11.3) 252 (54.7) P Frequency matched on sex, race, age (± 5 years), geography
Italian Multicentre Miligi et al (1999) Italy 1991-1993 In-person interviews using structured questionnaires
Open-ended responses cancer.		 270 62.5 (8.5) 137 (50.7) 1161 54.4 (14.1) 566 (48.8) P Random sample of general population, stratified by sex and 5-year age groups
Los Angeles County Case Control Multiple Myeloma Study (LACCMM) Nuyujukian et al (2014) Los Angeles County, California, USA 1985-1992 Prompted for MM, leukaemia, Hodgkin disease or lymphoma 278 61.2 (9.0) 126 (45.3) 278 61.2 (9.0) 126 (45.3) P (Neighbourhood) Individually matched by sex, race, date of birth within 5 years, neighbourhood of residence at case's diagnosis
Nebraska Zahm et al (1992) Nebraska, USA 1983-1986 Telephone interviews
Open-ended question for cancer.		 72 68.9 (9.6) 32 (44.4) 1427 67.1 (16.1) 724 (50.7) P 3:1 frequency matched by race, sex, vital status, and age to the combined age distribution of NHL, HL, CLL and MM
Population Health Study (PopHealth) Baris et al (2004) GA, MI, NJ, USA 1986-1989 In-person interviews
Open-ended responses for cancer.		 573 64.2 (10.1) 289 (50.4) 2146 61.6 (11.0) 782 (36.4) P Frequency matched to expected race, sex, age distribution of MM, oesophagus, pancreatic, and prostate cancer cases
University of Southern California (USC) Cozen et al (2006) Los Angeles, CA, USA 2000-2002 Questionnaires administered during in-person interviews (except 65 telephone interviews conducted with relatives of cases living outside LA County)
Cases prompted for family history of HL, leukaemia and myeloma. Also open-ended for other types of cancers. Controls prompted for family history of NHL, HL, leukaemia, and MM.		 150 58.9 (10.1) 59 (39.3) 252 58.8 (10.5) 112 (44.4) P Frequency matched to expected race, age, sex distribution
Utah Camp et al (2008) Utah, USA 2008-2012 Prompted for family history of NHL, MM, HL and leukaemia 330 59.7 (10.8) 128 (39.8) 580 66.2 (10.3) 296 (51.0) Population and spouses Frequency matched by birth cohort and sex
Open in a new tab

Abbreviations: P, Population; H, Hospital; NHL, non Hodgkin lymphoma; HL, Hodgkin lymphoma; MM, multiple myeloma; MGUS, monoclonal gammopathy of undetermined significance; MDS, myeloysplastic syndrome; CLL, chronic lymphocytic leukaemia; STS, soft tissue sarcoma; USA, United States of America; SD, standard deviation.

Cases were patients with a primary diagnosis of MM and were identified through cancer registries or hospitals in the individual studies. With the exception of the Utah study, where cases were predominantly prevalent rather than incidental, all studies enrolled incidental MM cases.

Explanatory variables

Data on family history of LH cancers were derived from either in-person or telephone interviews, conducted using structured questionnaires. Participants were either prompted for the type of cancer that affected a first-degree relative using a defined list of cancer types or asked about family history of cancer as an open-ended question (Table I).

Harmonized variables across all studies were created for family history of LH cancer in a first-degree relative. Family history of each of the following were considered as individual explanatory variables: MM, leukaemia, NHL and Hodgkin lymphoma family history of MM excluding cancers of the bone. Family history of leukaemia included acute lymphoblastic leukaemia, acute myeloid leukaemia, acute T-cell, and chronic lymphocytic leukaemia (CLL). In addition to considering associations of MM risk with family history of these individual LH cancer types, a composite variable describing having a first-degree relative with any LH cancer (MM, leukaemia, NHL, Hodgkin lymphoma) was created. Variables available to create the composite family history variable differed across studies, due to differences in questionnaires and reporting. First degree relatives with the following cancers were also included in the composite LH cancer definition: cancers of the blood, bone, bone marrow, corpuscles, lymph glands, neck-side, neck glands, plasmacytoma, Sezary syndrome or spleen. The LH cancer definition excluded sarcoma. In the Epilymph study, a handful of participants reported a family history of MM or Hodgkin lymphoma, but most self-reported family cancers were coded within the Epilymph study as “hematologic cancer not otherwise specified.” Therefore, individual types of LH cancer were not coded for the Epilymph study.

In analysis, associations of MM risk with a history of LH cancer in a parent, sibling, child, male first-degree relative and female first-degree relative were explored. Having a son and having a daughter with a LH cancer history were also considered as explanatory variables; results from these analyses are not presented because of the small number of such occurrences. Age of the first-degree relative at the time of diagnosis of LH cancer (<55 versus 55+ years) and the number of relatives (1 versus > 1) with any LH cancer were also investigated as explanatory variables. Cut-off points for the latter two variables were based on sample size considerations and hypotheses about associations. In all analyses, the referent category consisted of participants who did not have any first-degree relative with a history of a LH cancer.

Statistical analysis

Unconditional logistic regression models were used to calculate odds ratios (OR) and 95% confidence intervals (CI) as estimates of association of family history of LH cancers with risk of MM. Sex and age were included as covariates in all analytic models. Age was entered into the models using linear and quadratic terms (continuous age + continuous age2). This coding was selected by using likelihood ratio tests to compare the goodness-of-fit of models with different functions of age. Additionally, models were adjusted for education (less than high school graduate versus high school graduate or more), which has been shown to contribute to specificity of self-reported family history of cancer (Chang, et al 2006). A term designating study was also included in all models.

Sex, African American race (African American/Black versus European American/White), and age of the MM case at diagnosis (<55 versus 55+ years) were investigated as effect modifiers by including them in interaction terms with the main explanatory variables (e.g. age x family history of MM). Age was investigated as a modifier based on the hypothesis that stronger associations of MM risk with family history of LH cancer among patients diagnosed at younger ages might indicate a detection bias or a genetic component to MM aetiology. Likelihood ratio tests were used to assess effect modification by comparing nested models with and without interaction terms.

Assessment of heterogeneity and sensitivity analyses

Likelihood ratio tests of interaction terms between each explanatory variable and a term for study were also used to assess heterogeneity across studies. We examined the effect of removing each individual study on the overall pooled estimates. The ENGELA and LACCMM studies employed individual matching of controls to cases, which can introduce confounding (Rothman and Greenland 1998). Therefore, we re-ran the individual models for these studies to examine the effect of adjusting for the matching factors used. Race/ethnicity was not included as covariate in the main models, because some studies did not collect this information. We re-ran the pooled analyses, limiting data to studies that had a term for African American race/ethnicity and adjusted for this covariate. In addition, we examined the impact of including only studies that ascertained data on family history of LH cancers through a prompted list (USC, iMAGE, ENGELA, LACCMM, Utah) versus open-ended questions (Canada, Epilymph, Nebraska, Conneticut, PopHealth, Italy). We also ran the analyses after excluding the two studies that included hospital-based controls (ENGELA and Epilymph).

All analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC, USA)

Results

The pooled data consisted of 2,843 cases and 11,470 controls (Table II). A higher proportion of cases and controls were men (56.3% and 57.4%, respectively). The mean ages of cases and controls were 62.5 and 59.5 years, respectively. Cases and controls had similar education levels.

Table II.

Demographic characteristics of cases and controls included in the pooled analysis of family history of lymphohaematapoietic cancer and risk of multiple myeloma

Cases, n (%) Controls, n (%)
Total 2,843 (19.9) 11,470 (80.1)
Sex
    Men 1,600 (56.3) 6,588 (57.4)
    Women 1,243 (43.7) 4,882 (42.6)
Age, mean ± standard deviation 62.5 ± 10.5 59.5 ± 14.7
Education
    Less than high school graduate 983 (35.0) 4,108 (36.6)
    High school graduate or more 1,822 (65.0) 7,126 (63.4)
Open in a new tab

Figure 1 shows study-specific estimates of association between having a first-degree relative with a history of any LH cancer and MM risk, along with the pooled OR. Overall, having a family member with any LH cancer was associated with a higher risk of MM. In the pooled analysis, risk of MM was positively associated with having a first-degree relative with any LH cancer (OR=1.29, 95% CI: 1.08-1.55).

Figure 1.

Figure 1

Open in a new tab

Forest plot showing study-specific and pooled odds ratio (OR, diamond) estimates (95% confidence interval, CI, indicated by the horizontal line) of association between multiple myeloma risk and having a first-degree relative with a history of a lymphohaematopoietic cancer. All models were adjusted for sex, age (coded using a linear and quadratic terms) and education (< high school versus high school or more). The pooled OR estimates were also adjusted for study. Abbreviations: iMAGE, Molecular and Genetic Epidemiology Study of Myeloma; LACCMM, Los Angeles County Case Control Multiple Myeloma Study; USC, University of Southern California.

Figure 2 shows study-specific associations of MM risk with family history of each of the LH cancer types (NHL, leukaemia, Hodgkin lymphoma and MM), along with the pooled ORs. All study-specific OR estimates for associations with having a first-degree relative with a history of MM were 1.4 or greater, with the exception of the OR estimate from the Utah Study. Study-specific ORs could not be estimated for two studies; ENGELA had no exposed cases and the Italian study had no exposed controls. In the pooled analysis, MM risk was positively associated with having a first-degree relative with MM (OR=1.90, 95% CI: 1.27-2.88). MM risk was also elevated in those cases that had a first-degree relative with a history of NHL (OR=1.35, 95% CI: 0.91-1.99) or leukaemia (OR=1.21, 95% CI: 0.92-1.59).

Figure 2.

Figure 2

Open in a new tab

Forest plots showing study-specific and pooled odds ratio (OR, diamond) estimates (95% confidence interval, CI, indicated by the horizontal line) of association between multiple myeloma and having a first-degree relative with a history of: A) multiple myeloma, B) NHL, C) leukaemia and D) Hodgkin lymphoma. All models were adjusted for sex, age (coded using a linear and quadratic terms) and education (< high school versus high school or more). The pooled OR estimates were also adjusted for study.

Abbreviations: NHL, non Hodgkin lymphoma; NE, Not estimable; iMAGE, Molecular and Genetic Epidemiology Study of Myeloma; LACCMM, Los Angeles County Case Control Multiple Myeloma Study; USC, University of Southern California.

Having more than one relative with a history of LH cancer was associated with a higher relative risk of MM (OR=1.67, 95% CI: 0.95-2.94, Table III) than having only one relative with a history of LH cancer (OR=1.26, 95% CI: 1.01-1.58). Having a sibling with a history of LH cancer was associated with an increased risk of MM (OR=1.56, 95% CI: 1.18-2.07). Having a female first-degree relative with a LH cancer history was more strongly associated with MM risk compared to having a male relative with a LH cancer history. This was particularly evident for relationships with sibling type; MM risk was more strongly associated with having a sister (OR=1.81, 95% CI: 1.18-2.78) than a brother (OR=1.35, 95% CI: 0.92-1.98) with a LH cancer. Risk of MM was not associated with having a parent or a child with a LH cancer history. However, having a mother with a history of LH cancer was associated with an elevated relative risk of MM (OR=1.43, 95% CI: 0.97-2.11). Age of the first-degree relative with a LH cancer history was not associated with MM.

Table III.

Associations between family history of lymphohaematopoietic cancer and multiple myeloma, overall and stratified by sex 1

Males (N=8,188) Females (N=6,125) Total Likelihood ratio test of the interaction term
Cases Controls OR, 95% CI Cases Controls OR, 95% CI OR, 95% CI Chi square (p-value)
Multiple myeloma2 25 (1.9) 20 (0.4) 4.13, 2.17-7.85 20 (1.9) 40 (1.1) 1.09, 0.62-1.90 1.90, 1.26-2.87 9.76 (<0.01)
NHL3 27 (2.2) 51 (1.0) 1.65, 1.00-2.73 15 (1.6) 48 (1.40) 1.02, 0.55-1.88 1.35, 0.91-1.99 1.46 (0.23)
Leukaemia2 49 (3.2) 117 (1.8) 1.45, 1.02-2.07 36 (3.0) 103 (2.2) 0.97, 0.64-1.46 1.21, 0.92-1.59 2.15 (0.14)
Hodgkin lymphoma3 9 (0.8) 36 (0.7) 0.95, 0.44-2.05 6 (0.6) 21 (0.6) 1.00, 0.39-2.60 0.97, 0.53-1.76 0.01 (0.94)
Any LH cancer 119 (7.5) 268 (4.1) 1.58, 1.24-2.00 81 (6.5) 261 (5.4) 1.01, 0.77-1.33 1.29, 1.08-1.55 5.90 (0.02)
1 relative4 74 (5.7) 212 (3.4) 1.60, 1.20-2.12 43 (4.1) 200 (4.5) 0.92, 0.65-1.31 1.26, 1.01-1.58 5.75 (0.06)
>1 relative 12 (0.9) 17 (0.3) 1.66, 0.73-3.80 15 (1.4) 16 (0.4) 1.66, 0.77-3.60 1.67, 0.95-2.94
Male relative5 55 (4.7) 139 (2.5) 1.47, 1.05-2.06 34 (3.8) 138 (3.5) 0.89, 0.60-1.34 1.18, 0.91-1.53 3.52 (0.06)
Female relative5 46 (3.9) 118 (2.1) 1.63, 1.13-2.35 30 (3.4) 99 (2.5) 1.09, 0.71-1.68 1.37, 1.04-1.81 1.94 (0.16)
Sibling6 45 (3.5) 98 (1.6) 1.78, 1.22-2.60 36 (3.5) 103 (2.3) 1.35, 0.90-2.04 1.56, 1.18-2.07 0.93 (0.34)
    Sister7 19 (1.7) 41 (0.8) 1.70, 0.95-3.04 16 (1.9) 36 (1.0) 1.96, 1.04-3.68 1.81, 1.18-2.78 0.10 (0.75)
    Brother7 23 (2.1) 56 (1.1) 1.59, 0.94-2.67 18 (2.2) 59 (1.6) 1.12, 0.63-1.98 1.35, 0.92-1.98 0.80 (0.37)
Parent6 49 (3.8) 128 (2.1) 1.52, 1.07-2.18 27 (2.7) 117 (2.6) 0.80, 0.51-1.24 1.16, 0.88-1.53 5.14 (0.02)
    Mother8 23 (1.9) 58 (1.0) 1.93, 1.16-3.22 16 (1.7) 53 (1.2) 0.99, 0.55-1.79 1.43, 0.97-2.11 2.74 (0.10)
    Father8 26 (2.1) 66 (1.1) 1.29, 0.79-2.10 9 (1.0) 60 (1.4) 0.51, 0.24-1.05 0.91, 0.61-1.37 4.68 (0.03
Child 6 11 (0.9) 40 (0.7) 0.98, 0.49-1.96 7 (0.7) 33 (0.7) 0.73, 0.31-1.69 0.86, 0.50-1.47 0.28 (0.60)
First-degree relative aged <55 years9 16 (3.3) 25 (1.0) 2.22, 1.12-4.38 16 (2.6) 70 (2.5) 0.81, 0.46-1.42 1.17, 0.77-1.79 5.08 (0.08)
First-degree relative aged ≥55 years 8 (1.6) 31 (1.3) 0.91, 0.40-2.06 13 (2.2) 51 (1.8) 1.00, 0.53-1.89 0.97, 0.59-1.60
Open in a new tab

Abbreviations: MM, multiple myeloma; NHL, non Hodgkin lymphoma; LH, lymphohaematapoietic; OR, Odds ratio; CI, confidence interval

1

All models were adjusted for study, sex, age (entered as a linear and quadratic term) and education (< high school versus high school or more). The referent categories consisted of those without any relative with any history of lymphohaematopoietic cancers among the entire cohort for the unstratified model or among women in the model with an interaction term between sex and family history variables.

2

Epilymph not included

3

Epilymph and LACCMM not included

4

USC and Utah not included

5

USC, Italy and LACCMM not included

6

USC and LACCMM not included

7

USC, ENGELA, Italian multicentre and LACCMM not included

8

USC, ENGELA and LACCMM not included

9

USC and Utah not included

When sex was investigated as an effect modifier, associations were generally stronger among men (Table III). Men had a strong, positive risk of MM in association with family history of MM, whereas the association for women was close to the null (men: OR=4.13, 95% CI: 2.17-7.85 versus women: OR=1.09, 95% CI: 0.62-1.90, p-value for likelihood ratio test of interaction term: <0.01). Likewise, family history of any LH cancer was associated with an elevated risk of MM for men but not women (men: OR=1.58, 95% CI: 1.24-2.00 versus women: OR=1.01, 95% CI: 0.77-1.33, p-value for likelihood ratio test of interaction term: 0.02). Similarly, the effects of having a female relative, particularly a mother, with a history of LH cancer was restricted to men (men: OR=1.63, 95% CI: 1.13-2.35 versus women: OR=1.09, 95% CI: 0.91-1.68; p-value for likelihood ratio test of interaction term: 0.16 for female relative; men: OR=1.93, 95% CI: 1.16-3.22 versus women: OR=0.99, 95% CI: 0.55-1.79; p-value for likelihood ratio test of interaction term: 0.10 for mother).

When the pooled analyses were restricted to data from studies with African American participants, and adjusted for race/ethnicity in addition to sex, age, education and study, results were similar to those from analyses of the full data set that were unadjusted for race/ethnicity (Table IV). The association between MM risk and having a first-degree relative with a history of MM was stronger among African Americans/Blacks than European Americans/Whites (African Americans: OR=5.52, 95% CI: 1.87-16.28 versus European Americans/whites: OR=1.26, 95% CI: 0.75-2.10; p-value for likelihood ratio test of interaction term: 0.01). Also, the association between MM risk and having a female relative, particularly a mother, with a LH cancer history was stronger for African Americans than European Americans (African Americans: OR=2.71, 95% CI: 1.25-5.87 versus European Americans/whites: OR=1.23, 95% CI: 0.86-1.77, p-value for likelihood ratio test of interaction term: 0.07 for female relative; African Amercians: OR=4.24, 95% CI: 1.49-12.02 versus European Americans/whites: OR=1.32, 95% CI: 0.80-2.17, p-value for likelihood ratio test of interaction term: 0.05 for mother).

Table IV.

Associations between family history of lympho-haematopoietic (LH) cancer and multiple myeloma, overall and stratified by African American race/ethnicity, in the subset of studies with race data1

African American (N=1,754) White/European (N=5,764) Total Likelihood ratio test for interaction term
Cases N (%) Controls N (%) OR, 95% CI Cases N (%) Controls N (%) OR, 95% CI OR, 95% CI Chi square (p-value)
Multiple myeloma 12 (2.9) 5 (0.4) 5.52, 1.87-16.27 24 (2.0) 47 (1.1) 1.26, 0.75-2.10 1.69, 1.08-2.64 6.39 (0.01)
NHL2 1 (0.3) 6 (0.5) 0.48, 0.06-4.10 33 (3.1) 69 (1.7) 1.43, 0.91-2.24 1.34, 0.87-2.08 1.16 (0.28)
Leukaemia 11 (2.6) 26 (2.0) 1.30, 0.63-2.70 49 (3.9) 124 (2.9) 1.20, 0.83-1.73 1.22, 0.88-1.70 0.04 (0.84)
Hodgkin lymphoma2 3 (0.9) 2 (0.2) 3.13, 0.52-18.90 10 (1.0) 39 (1.0) 1.11, 0.53-2.32 1.29, 0.66-2.51 1.12 (0.29)
Any LH cancer 26 (6.0) 47 (3.6) 1.57, 0.95-2.60 121 (9.2) 310 (7.0) 1.23, 0.97-1.56 1.28, 1.03-1.59 0.75 (0.39)
1 relative3 17 (4.3) 37 (3.0) 1.44, 0.79-2.63 56 (6.0) 212 (5.7) 1.19, 0.86-1.64 1.24, 0.93-1.65 1.33 (0.51)
>1 relative 6 (1.5) 5 (0.4) 2.68, 0.77-9.31 16 (1.7) 22 (0.6) 1.27, 0.61-2.63 1.53, 0.81-2.89
Male relative4 9 (2.8) 26 (2.2) 1.05, 0.48-2.29 52 (5.2) 167 (4.2) 1.07, 0.76-1.51 1.07, 0.78-1.46 0.00 (0.96)
Female relative4 12 (3.6) 16 (1.4) 2.71, 1.25-5.87 46 (4.6) 135 (3.4) 1.23, 0.86-1.77 1.41, 1.01-1.95 3.14 (0.07)
Sibling4 9 (2.8) 16 (1.4) 1.83, 0.79-4.26 42 (4.2) 119 (3.0) 1.38, 0.93-2.04 1.45, 1.02-2.06 0.35 (0.55)
    Sister4 5 (1.6) 7 (0.6) 2.53, 0.77-8.26 19 (2.0) 53 (1.4) 1.45, 0.82-2.55 1.60, 0.96-2.66 0.68 (0.41)
    Brother4 5 (1.6) 10 (0.9) 1.49, 0.50-4.49 23 (2.4) 72 (1.9) 1.20, 0.71-2.02 1.25, 0.78-2.00 0.12 (0.73)
Parent4 11 (3.3) 21 (1.8) 1.66, 0.78-3.54 46 (4.6) 140 (3.6) 1.07, 0.74-1.53 1.15, 0.83-1.60 1.06 (0.30)
    Mother4 8 (2.5) 7 (0.6) 4.24, 1.49-12.02 24 (2.5) 64 (1.7) 1.32, 0.80-2.17 1.62, 1.04-2.53 3.89 (0.05)
    Father4 3 (0.9) 14 (1.2) 0.59, 0.17-2.10 24 (2.5) 78 (2.0) 0.91, 0.56-1.49 0.85, 0.54-1.35 0.42 (0.52)
Child 4 2 (0.6) 6 (0.5) 1.20, 0.24-6.06 10 (1.0) 44 (1.1) 0.76, 0.37-1.55 0.81, 0.42-1.56 0.24 (0.62)
First-degree relative aged <55 years9 5 (4.0) 6 (2.7) 1.39, 0.41-4.69 15 (4.9) 45 (4.9) 0.87, 0.47-1.61 0.96, 0.56-1.65 0.57 (0.75)
First-degree relative aged ≥55 years 3 (2.4) 4 (1.8) 1.19, 0.26-5.49 9 (2.9) 28 (3.0) 0.85, 0.39-1.85 0.91, 0.45-1.82
Open in a new tab

Abbreviations: NHL, non Hodgkin lymphoma; LH, lymphohaematapoietic; OR, Odds ratio; CI, confidence interval

1

Data from Epilymph, Canada, Italian multicentre and ENGELA were not included because of lack of information on race. All models were adjusted for study, sex, age at diagnosis (entered as a linear and quadratic term), race and education (< high school versus high school or more). The referent category for the models consisted of those without any relative with any history of lymphohaematopoietic cancers among the entire cohort for the unstratified model or among whites in the stratified model.

2

LACCMM not included

3

USC and Utah not included

4

USC and LACCMM not included

Many of the associations were slightly stronger among MM cases diagnosed at younger ages (Supplementary Table 1). However, most of the differences across categories of age of diagnosis or enrolment were not substantially different.

Heterogeneity

For the most part, formal statistical tests of heterogeneity of effects across studies, which were assessed using likelihood ratio tests of interaction terms between study and the explanatory variables, indicated homogeneity; the p-values for the test of homogeneity were 0.23 or greater (a small p-value would be suggestive of heterogeneity). There was evidence of some heterogeneity in the association of MM risk with family history of MM (p-value for likelihood ratio test: 0.07). However, excluding data from the Utah study removed the suggestion of heterogeneity (p for likelihood ratio test of interaction term after removing Utah study: 0.49).

Sensitivity analyses

When models were re-run for the ENGELA and LACCMM studies with adjustment for the individual level matching variables in each, the results were nearly identical to those in which models were adjusted for age, sex and education (data not shown).

For the most part, exclusion of individual studies did not substantially impact the pooled OR estimates. However, when data from the Utah study were excluded, associations with having a family history of MM became stronger, moving from an OR=1.90, 95% CI: 1.26-2.87 to an OR of 2.71, 95% CI: 1.69-4.36.

When pooled analyses were restricted to data from studies that used an open-ended question to ascertain information about first-degree relatives’ cancer type (Canada, Epilymph, Nebraska, Connecticut, PopHealth, Italy), the OR estimate of association between MM risk and family history of MM became stronger, moving from OR=1.90 (95% CI: 1.26-2.87) to OR=3.40 (95% CI: 1.88-6.14). By comparison, when analyses were restricted to include only data from studies that ascertained family history of LH cancer using a prompted list, the estimate of association between MM risk and family history of MM decreased towards the null, to OR=1.32 (95% CI: 0.75-2.30). Removing data from the two studies that enrolled hospital based controls did not cause any substantial changes in results.

Discussion

We pooled data from studies conducted in the US, Canada and several European countries to investigate associations between history of LH cancer in first-degree relatives and MM risk. Elevated risk of MM was associated with having a first-degree relative with a history of any LH cancer, and especially with a history of MM. Higher risk of MM was associated with having a female first-degree relative, a sibling, and especially a sister, with a history of a LH cancer. Most associations were stronger among men. The association between MM and having a first degree relative with a history of MM was stronger in African Americans; this is the largest analysis to show familial aggregation patterns among African Americans.

The finding of a higher risk of MM associated with a family history of LH cancer, and especially MM, is consistent with previous reports (Eriksson and Hallberg 1992, Bourguet, et al 1985). Associations of MM with family history of NHL, Hodgkin lymphoma and leukaemia were less substantial than with family history of MM, suggesting that the associations between MM risk and family history of any LH cancer were driven by family history of MM. This is consistent with previous research (Kristinsson, et al 2009a, Kristinsson, et al 2009b). Most associations were stronger among men, including having a family member with a history of LH cancer or a history of MM, and having a mother with a history of LH cancer. These sex-related results are somewhat in conflict with previous studies, which have found higher risk of MM in daughters of mothers with MM, and in sons of fathers with a history of MM (Lynch, et al 2005).

It is not clear why most associations were stronger among men. It is possible that these results reflect a sex-related reporting bias, especially among controls. A higher proportion of female than male controls reported having a family history of a LH cancer. Past validation studies have shown that females report a family history of cancer with higher sensitivity but lower specificity than men (Chang, et al 2006). If, in the data used in this pooled analysis, female controls reported family history with greater sensitivity and lower specificity than male controls, this could partially explain the results.

This is one of the first analyses to show that familial aggregation of MM is stronger among African Americans. Higher risk of MM in association with having a relative with a history of LH cancer, and with a history of MM, has been observed previously in Blacks versus Whites (Brown, et al 1999). Rates of MM have been shown to be higher among African Americans than in European Americans/Whites. In 2011 in the US, the age-adjusted rate of MM in African Americans was more than twice the rate among European Americans/Whites (Howlader, et al 2015). The stronger association among African Americans could suggest genetic susceptibility to MM.

Our results are supported by evidence of genetic components to MM disease aetiology. Carriage of the hyperphosphorylated form of the paratarg-7 protein (pP-7), which is a single target protein whose function remains unknown, is a molecularly defined autosomal dominantly inherited risk factor for MGUS and MM (Grass, et al 2009). A higher prevalence of pP-7 carriage has been observed among African American MM patients compared to other ethnic groups, although the pP-7 carrier state is also a strong risk factor in European MM cases (Zwick, et al 2014). Hyperphosphorylation of other autoantigenic paraprotein targets, including paratargs 2,5,6,8,9,10 and 11, have been identified in MM patients (Koura and Langston 2013). Human leucocyte antigen (HLA) types, including HLA-Cw2 and HLA-B18, have been investigated as potentially associated with MM (Koura and Langston 2013). Additionally, genome wide association studies (GWAS) have identified regions on chromosomes 3p22, 7p15.3 and 2p23.3 as potentially associated with MM (Broderick, et al 2012, Koura and Langston 2013). The SNP rs2456449 polymorphism on chromosome 8q24, associated with CLL (Crowther-Swanepoel, et al 2010), has also been associated with MM (Campa, et al 2012). This could indicate common susceptibility between MM and CLL. CLL is a subtype of NHL. We observed an elevated risk of MM in association with having a first-degree relative with NHL. However, due to sample size considerations and a lack of information on NHL subtypes from all of the contributing studies, we did not evaluate the association between family history of CLL, specifically, and MM risk.

In sensitivity analyses, exclusion of data from the Utah study caused the pooled OR to increase. Also, the test of heterogeneity suggested that the Utah study contributed to heterogeneity in the estimate of association between family history of MM and MM risk. This difference in association might be due to the Utah study having included predominantly prevalent rather than incidental cases. The inclusion of predominantly prevalent cases might have caused a selection bias; by including prevalent cases, the Utah study might have included MM cases with a better prognosis and, in association with this superior prognosis, weaker inheritance patterns. There is genetic evidence that could support this theory. A GWAS study identified SNPs at chromosome 16p13 that were associated with mortality among MM patients (Ziv, et al 2015), suggesting that poorer prognosis might be associated with inheritance.

In support of our finding of an association between having a sibling with a LH cancer history and MM risk, several reports have shown a strong clustering of MM among siblings (Jain, et al 2009, Shoenfeld, et al 1982). However, in a population-based study in Sweden (Altieri, et al 2006), having parents rather than siblings with a history of MM was associated with a higher risk of MM, although the estimates of association among siblings were based on small numbers. Strong sibling risk is consistent with an aetiology that includes recessive genetic components. Other models could also be suggested, such as gene-environment interactions with early life exposures. Thus far, such models have not been well studied for MM.

This analysis had several strengths. The large data set accommodated investigation of associations with MM within strata of sex, age, and race/ethnicity. Because this analysis was restricted to history of cancer among first-degree relatives of MM cases and controls, the participant's recall of relative's cancer history should have been fairly good, leading to less misclassification. Indeed, probands with breast, ovarian, and colorectal cancers have been shown to better recall the cancer history of first-degree versus second-degree relatives (Ziogas and Anton-Culver 2003).

Limitations in this analysis include the potential for recall bias. Cases may have more accurately recalled family history than controls and, among controls, women may have had better recall than men. We were unable to validate self-reports of family history of cancer. In an investigation of self-reports of cancer among case and population-based control participants of a study of incident malignant lymphoma in Denmark and Sweden, the sensitivity of self-reports for any cancer, and more specifically for history of haematopoietic cancer, was higher for cases than controls, but the specificity was marginally higher for controls than cases (Ahmed, et al 2006). If a similar phenomenon occurred in the current report, it would have inflated our effect estimates. Although our sensitivity analysis that restricted analyses to data from studies that ascertained family history of cancer type using an open-ended question suggested that the risk estimate of association with family history of MM was inflated, this could be partially a result of removing the data from the Utah Study.

Although unlikely due to the rarity of MM, it is possible that cases may have been diagnosed earlier and more frequently because of a family history of the disease. However, when age at diagnosis/enrolment was investigated as an effect modifier, the differences in associations were not substantially different for younger versus older participants. To some extent, this indicates that diagnosis bias is unlikely to explain the results. Restricting analyses to family history of MM in first-degree relatives makes it difficult to separate shared genetics from shared environmental exposures. Family history of MM in more distant relatives might be more indicative of genetic associations; however, those data are difficult to acquire outside of genealogical studies. Because most of the contributing studies did not have information on family size, we could not adjust for this variable, which could have confounded the association of family history of LH cancer with risk of MM. Finally, MGUS has been shown to be more prevalent in relatives of MM patients than in the general population (Vachon, et al 2009). However, we were unable to evaluate associations of MM risk with a family history of MGUS due to lack of data.

Results from this pooled analysis provide compelling evidence to support hypotheses that genetic inheritance plays a role in the aetiology of MM. Future studies should investigate the interaction of genetic markers with shared environmental exposures or lifestyle factors within families, such as pesticide use in the home environment and obesity. These results also suggest the need for further study of genetic inheritance among African Americans and identification of genetic markers that may account for associations of MM risk with family history of specific types of LH cancers.

Supplementary Material

Supp Table S1

Acknowledgements

This study was supported by the National Institute of Environmental Health Sciences grant R21 ES021592 (A.J.D.). The Connecticut Study and the Population Health Study were supported in part by the Intramural Research Program of the NIH. The ENGELA study (JC) was supported by grants from the Association pour la Recherche contre le Cancer, the Fondation de France, AFSSET and a donation from Faberge employees. The work conducted by E.E. Brown was supported, in part, by grants from the NCI (U54CA118948, R21CA155951, R25CA76023, R01CA186646, and the University of Alabama at Birmingham Comprehensive Cancer Center Support Grant P30CA13148) and the American Cancer Society (IRG60-001-47).

Footnotes

Place work was carried out:

Department of Environmental and Occupational Health, Drexel University Dornsife School of Public Health

Authorship: LH Schinasi analysed and interpreted data and led the manuscript writing; AJ De Roos designed the research, analysed and interpreted data, and contributed to manuscript writing; G Tricot and D Atanackovic performed the clinical reviews for the Utah Study; EE Brown, NJ Camp, SS Wang, JN Hofmann, BC Chiu, L Miligi, L Beane Freeman, S de Sanjose, L Bernstein, A Monnereau, J Clavel, P Cocco, L Orsi, P Pahwa, JA Dosman, JR McLaughlin, MP Purdue, W Cozen and JJ Spinelli contributed data. All authors contributed to drafting or revising of the paper and approved the submitted version. All authors have no conflicts of interest to disclose.

References

  1. Ahmed B, Estey E, Manning J, David C, Keating MJ, Kantarjian H, Tsimberidou AM. Anaplastic large cell lymphoma with involvement of the pancreas presenting as panniculitis in a patient with a history of acute myeloid leukemia--case report and review of the literature. Haematologica. 2006;91:ECR55. [PubMed] [Google Scholar]
  2. Alexander DD, Mink PJ, Adami HO, Cole P, Mandel JS, Oken MM, Trichopoulos D. Multiple myeloma: a review of the epidemiologic literature. International Journal of Cancer. 2007;120(Suppl 12):40–61. doi: 10.1002/ijc.22718. [DOI] [PubMed] [Google Scholar]
  3. Altieri A, Chen B, Bermejo JL, Castro F, Hemminki K. Familial risks and temporal incidence trends of multiple myeloma. European Journal of Cancer. 2006;42:1661–1670. doi: 10.1016/j.ejca.2005.11.033. [DOI] [PubMed] [Google Scholar]
  4. Andreotti G, Birmann B, de Roos AJ, Spinelli J, Cozen W, Camp NJ, Moysich K, Chiu B, Steplowski E, Krzystan J, Boffetta P, Benhaim-Luzon V, Brennan P, de SS, Costas L, Costantini AS, Miligi L, Cocco P, Becker N, Foretova L, Maynadie M, Nieters A, Staines A, Tricot G, Milliken K, Weisenburger D, Zheng T, Baris D, Purdue MP. A pooled analysis of alcohol consumption and risk of multiple myeloma in the international multiple myeloma consortium. Cancer Epidemiology Biomarkers and Prevention. 2013;22:1620–1627. doi: 10.1158/1055-9965.EPI-13-0334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baris D, Silverman DT, Brown LM, Swanson GM, Hayes RB, Schwartz AG, Liff JM, Schoenberg JB, Pottern LM, Greenberg RS, Stewart PA. Occupation, pesticide exposure and risk of multiple myeloma. Scandinavian Journal Work Environment and Health. 2004;30:215–222. doi: 10.5271/sjweh.782. [DOI] [PubMed] [Google Scholar]
  6. Bizzaro N, Pasini P. Familial occurrence of multiple myeloma and monoclonal gammopathy of undetermined significance in 5 siblings. Haematologica. 1990;75:58–63. [PubMed] [Google Scholar]
  7. Bourguet CC, Grufferman S, Delzell E, DeLong ER, Cohen HJ. Multiple myeloma and family history of cancer. A case-control study. Cancer. 1985;56:2133–2139. doi: 10.1002/1097-0142(19851015)56:8<2133::aid-cncr2820560842>3.0.co;2-f. [DOI] [PubMed] [Google Scholar]
  8. Broderick P, Chubb D, Johnson DC, Weinhold N, Forsti A, Lloyd A, Olver B, Ma YP, Dobbins SE, Walker BA, Davies FE, Gregory WA, Child JA, Ross FM, Jackson GH, Neben K, Jauch A, Hoffmann P, Muhleisen TW, Nothen MM, Moebus S, Tomlinson IP, Goldschmidt H, Hemminki K, Morgan GJ, Houlston RS. Common variation at 3p22.1 and 7p15.3 influences multiple myeloma risk. Nature Genetics. 2012;44:58–61. doi: 10.1038/ng.993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brown LM, Linet MS, Greenberg RS, Silverman DT, Hayes RB, Swanson GM, Schwartz AG, Schoenberg JB, Pottern LM, Fraumeni JF., Jr. Multiple myeloma and family history of cancer among blacks and whites in the U.S. Cancer. 1999;85:2385–2390. [PubMed] [Google Scholar]
  10. Camp NJ, Werner TL, Cannon-Albright LA. Familial myeloma. N Engl J Med. 2008;359:1734–1735. doi: 10.1056/NEJMc081677. author reply 1735. [DOI] [PubMed] [Google Scholar]
  11. Campa D, Martino A, Sainz J, Buda G, Jamroziak K, Weinhold N, Vieira Reis RM, Garcia-Sanz R, Jurado M, Rios R, Szemraj-Rogucka Z, Marques H, Lesueur F, Bugert P, Moreno V, Szemraj J, Orciuolo E, Gemignani F, Rossi AM, Dumontet C, Petrini M, Goldschmidt H, Landi S, Canzian F. Comprehensive investigation of genetic variation in the 8q24 region and multiple myeloma risk in the IMMEnSE consortium. British Journal of Haematology. 2012;157:331–338. doi: 10.1111/j.1365-2141.2012.09047.x. [DOI] [PubMed] [Google Scholar]
  12. Casey R, Brennan P, Becker N, Boffetta P, Cocco P, Domingo-Domenech E, Foretova L, Nieters A, de SS, Staines A, Vornanen M, Maynadie M. Influence of familial cancer history on lymphoid neoplasms risk validated in the large European case-control study epilymph. European Journal of Cancer. 2006;42:2570–2576. doi: 10.1016/j.ejca.2006.03.031. [DOI] [PubMed] [Google Scholar]
  13. Chang ET, Smedby KE, Hjalgrim H, Glimelius B, Adami HO. Reliability of selfreported family history of cancer in a large case-control study of lymphoma. Journal of the National Cancer Institute. 2006;98:61–68. doi: 10.1093/jnci/djj005. [DOI] [PubMed] [Google Scholar]
  14. Chubb D, Weinhold N, Broderick P, Chen B, Johnson DC, Forsti A, Vijayakrishnan J, Migliorini G, Dobbins SE, Holroyd A, Hose D, Walker BA, Davies FE, Gregory WA, Jackson GH, Irving JA, Pratt G, Fegan C, Fenton JA, Neben K, Hoffmann P, Nothen MM, Muhleisen TW, Eisele L, Ross FM, Straka C, Einsele H, Langer C, Dorner E, Allan JM, Jauch A, Morgan GJ, Hemminki K, Houlston RS, Goldschmidt H. Common variation at 3q26.2, 6p21.33, 17p11.2 and 22q13.1 influences multiple myeloma risk. Nature Genetics. 2013;45:1221–1225. doi: 10.1038/ng.2733. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Coleman EA, Lynch H, Enderlin C, Stewart B, Thome SD, Kennedy R, Richardson-Nelson T, Barlogie B. Initial report of a family registry of multiple myeloma. Cancer Nursing. 2009;32:456–464. doi: 10.1097/NCC.0b013e3181a90168. [DOI] [PubMed] [Google Scholar]
  16. Cozen W, Gebregziabher M, Conti DV, Van Den Berg DJ, Coetzee GA, Wang SS, Rothman N, Bernstein L, Hartge P, Morhbacher A, Coetzee SG, Salam MT, Wang W, Zadnick J, Ingles SA. Interleukin-6-related genotypes, body mass index, and risk of multiple myeloma and plasmacytoma. Cancer Epidemiology Biomarkers and Prevention. 2006;15:2285–2291. doi: 10.1158/1055-9965.EPI-06-0446. [DOI] [PubMed] [Google Scholar]
  17. Crowther-Swanepoel D, Broderick P, Di Bernardo MC, Dobbins SE, Torres M, Mansouri M, Ruiz-Ponte C, Enjuanes A, Rosenquist R, Carracedo A, Jurlander J, Campo E, Juliusson G, Montserrat E, Smedby KE, Dyer MJ, Matutes E, Dearden C, Sunter NJ, Hall AG, Mainou-Fowler T, Jackson GH, Summerfield G, Harris RJ, Pettitt AR, Allsup DJ, Bailey JR, Pratt G, Pepper C, Fegan C, Parker A, Oscier D, Allan JM, Catovsky D, Houlston RS. Common variants at 2q37.3, 8q24.21, 15q21.3 and 16q24.1 influence chronic lymphocytic leukemia risk. Nature Genetics. 2010;42:132–136. doi: 10.1038/ng.510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Eriksson M, Hallberg B. Familial occurrence of hematologic malignancies and other diseases in multiple myeloma: a case-control study. Cancer Causes Control. 1992;3:63–67. doi: 10.1007/BF00051914. [DOI] [PubMed] [Google Scholar]
  19. Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray F. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet] International Agency for Research on Cancer; Lyon, France: 2013. [Google Scholar]
  20. Ghosh S, McLaughlin JR, Spinelli JJ, Dosman JA, McDuffie HH, Pahwa P. Multiple myeloma and occupational exposures: a population-based case-control study. Journal of Occupational and Environmental Medicine. 2011;53:641–646. doi: 10.1097/JOM.0b013e31821a1ebb. [DOI] [PubMed] [Google Scholar]
  21. Grass S, Preuss KD, Ahlgrimm M, Fadle N, Regitz E, Pfoehler C, Murawski N, Pfreundschuh M. Association of a dominantly inherited hyperphosphorylated paraprotein target with sporadic and familial multiple myeloma and monoclonal gammopathy of undetermined significance: a case-control study. Lancet Oncology. 2009;10:950–956. doi: 10.1016/S1470-2045(09)70234-7. [DOI] [PubMed] [Google Scholar]
  22. Greenberg AJ, Rajkumar SV, Vachon CM. Familial monoclonal gammopathy of undetermined significance and multiple myeloma: epidemiology, risk factors, and biological characteristics. Blood. 2012;119:5359–5366. doi: 10.1182/blood-2011-11-387324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Howlader N, Noone AM, Krapcho M, Garshell J, Miller D, Altekruse SF, Kosary CL, Yu M, Ruhl J, Tatalovich Z, Mariotto A, Lewis DR, Chen HS, Feuer EJ, Cronin KA. SEER Cancer Statistics Review, 1975-2011, based on November 2013 SEER data submission. 2015.
  24. Jain M, Ascensao J, Schechter GP. Familial myeloma and monoclonal gammopathy: a report of eight African American families. American Journal of Hematology. 2009;84:34–38. doi: 10.1002/ajh.21325. [DOI] [PubMed] [Google Scholar]
  25. Koura DT, Langston AA. Inherited predisposition to multiple myeloma. Ther.Adv.Hematol. 2013;4:291–297. doi: 10.1177/2040620713485375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kristinsson SY, Bjorkholm M, Goldin LR, Blimark C, Mellqvist UH, Wahlin A, Turesson I, Landgren O. Patterns of hematologic malignancies and solid tumors among 37,838 first-degree relatives of 13,896 patients with multiple myeloma in Sweden. International Journal of Cancer. 2009a;125:2147–2150. doi: 10.1002/ijc.24514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kristinsson SY, Goldin LR, Bjorkholm M, Koshiol J, Turesson I, Landgren O. Genetic and immune-related factors in the pathogenesis of lymphoproliferative and plasma cell malignancies. Haematologica. 2009b;94:1581–1589. doi: 10.3324/haematol.2009.008979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Landgren O, Linet MS, McMaster ML, Gridley G, Hemminki K, Goldin LR. Familial characteristics of autoimmune and hematologic disorders in 8,406 multiple myeloma patients: a population-based case-control study. International Journal of Cancer. 2006;118:3095–3098. doi: 10.1002/ijc.21745. [DOI] [PubMed] [Google Scholar]
  29. Landgren O, Kristinsson SY, Goldin LR, Caporaso NE, Blimark C, Mellqvist UH, Wahlin A, Bjorkholm M, Turesson I. Risk of plasma cell and lymphoproliferative disorders among 14621 first-degree relatives of 4458 patients with monoclonal gammopathy of undetermined significance in Sweden. Blood. 2009;114:791–795. doi: 10.1182/blood-2008-12-191676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lynch HT, Watson P, Tarantolo S, Wiernik PH, Quinn-Laquer B, Isgur BK, Huiart L, Olopade OI, Sobol H, Sanger W, Hogg D, Weisenburger D. Phenotypic heterogeneity in multiple myeloma families. Journal of Clinical Oncology. 2005;23:685–693. doi: 10.1200/JCO.2005.10.126. [DOI] [PubMed] [Google Scholar]
  31. McDuffie HH, Pahwa P, Karunanayake CP, Spinelli JJ, Dosman JA. Clustering of cancer among families of cases with Hodgkin Lymphoma (HL), Multiple Myeloma (MM), Non-Hodgkin's Lymphoma (NHL), Soft Tissue Sarcoma (STS) and control subjects. BMC Cancer. 2009;9:70. doi: 10.1186/1471-2407-9-70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Miligi L, Seniori CA, Crosignani P, Fontana A, Masala G, Nanni O, Ramazzotti V, Rodella S, Stagnaro E, Tumino R, Vigano C, Vindigni C, Vineis P. Occupational, environmental, and life-style factors associated with the risk of hematolymphopoietic malignancies in women. American Journal of Industrial Medicine. 1999;36:60–69. doi: 10.1002/(sici)1097-0274(199907)36:1<60::aid-ajim9>3.0.co;2-z. [DOI] [PubMed] [Google Scholar]
  33. Nuyujukian DS, Voutsinas J, Bernstein L, Wang SS. Medication use and multiple myeloma risk in Los Angeles County. Cancer Causes Control. 2014;25:1233–1237. doi: 10.1007/s10552-014-0424-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ogmundsdottir HM, Haraldsdottirm V, Johannesson GM, Olafsdottir G, Bjarnadottir K, Sigvaldason H, Tulinius H. Familiality of benign and malignant paraproteinemias. A population-based cancer-registry study of multiple myeloma families. Haematologica. 2005;90:66–71. [PubMed] [Google Scholar]
  35. Orsi L, Delabre L, Monnereau A, Delval P, Berthou C, Fenaux P, Marit G, Soubeyran P, Huguet F, Milpied N, Leporrier M, Hemon D, Troussard X, Clavel J. Occupational exposure to pesticides and lymphoid neoplasms among men: results of a French case-control study. Occupational and Environmental Medicine. 2009;66:291–298. doi: 10.1136/oem.2008.040972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Perrotta C, Staines A, Codd M, Kleefeld S, Crowley D, T MA, Becker N, Brennan P, de SS, Foretova L, Maynadie M, Nieters A, Boffetta P, Cocco P. Multiple Myeloma and lifetime occupation: results from the EPILYMPH study. Journal of Occupational Medicine and Toxicology. 2012;7:25. doi: 10.1186/1745-6673-7-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Purdue MP, Lan Q, Menashe I, Zheng T, Zhang Y, Yeager M, Hosgood HD, III, Zahm SH, Chanock SJ, Rothman N, Baris D. Variation in innate immunity genes and risk of multiple myeloma. Hematology Oncology. 2011;29:42–46. doi: 10.1002/hon.954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rothman K, Greenland S. Modern Epidemiology. Philadelphia, PA: 1998. [Google Scholar]
  39. Shoenfeld Y, Berliner S, Shaklai M, Gallant LA, Pinkhas J. Familial multiple myeloma. A review of thirty-seven families. Postgraduate Medical Journal. 1982;58:12–16. doi: 10.1136/pgmj.58.675.12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Steingrimsdottir H, Einarsdottir HK, Haraldsdottir V, Ogmundsdottir HM. Familial monoclonal gammopathy: hyper-responsive B cells in unaffected family members. European Journal of Haematology. 2011;86:396–404. doi: 10.1111/j.1600-0609.2011.01593.x. [DOI] [PubMed] [Google Scholar]
  41. Vachon CM, Kyle RA, Therneau TM, Foreman BJ, Larson DR, Colby CL, Phelps TK, Dispenzieri A, Kumar SK, Katzmann JA, Rajkumar SV. Increased risk of monoclonal gammopathy in first-degree relatives of patients with multiple myeloma or monoclonal gammopathy of undetermined significance. Blood. 2009;114:785–790. doi: 10.1182/blood-2008-12-192575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Vangsted A, Klausen TW, Vogel U. Genetic variations in multiple myeloma I: effect on risk of multiple myeloma. European Journal of Haematology. 2012;88:8–30. doi: 10.1111/j.1600-0609.2011.01700.x. [DOI] [PubMed] [Google Scholar]
  43. VanValkenburg ME, Pruitt GI, Brill IK, Costa L, Ehtsham M, Justement IT, Innis-Shelton RD, Salzman D, Reddy ES, Godby KN, Mikhail FM, Carroll AJ, Reddy VB, Sanderson RD, Justement LB, Sanders PW, Brown EE. Family history of hematologic malignancies and risk of multiple myeloma: differences by race and clinical features. Cancer Causes Control. 2016;27:81–91. doi: 10.1007/s10552-015-0685-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zahm SH, Weisenburger DD, Babbitt PA, Saal RC, Vaught JB, Blair A. Use of hair coloring products and the risk of lymphoma, multiple myeloma, and chronic lymphocytic leukemia. American Journal of Public Health. 1992;82:990–997. doi: 10.2105/ajph.82.7.990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ziogas A, Anton-Culver H. Validation of family history data in cancer family registries. American Journal of Preventative Medicine. 2003;24:190–198. doi: 10.1016/s0749-3797(02)00593-7. [DOI] [PubMed] [Google Scholar]
  46. Ziv E, Dean E, Hu D, Martino A, Serie D, Curtin K, Campa D, Aftab B, Bracci P, Buda G, Zhao Y, Caswell-Jin J, Diasio R, Dumontet C, Dudzinski M, Fejerman L, Greenberg A, Huntsman S, Jamroziak K, Jurczyszyn A, Kumar S, Atanackovic D, Glenn M, Cannon-Albright LA, Jones B, Lee A, Marques H, Martin T, Martinez-Lopez J, Rajkumar V, Sainz J, Vangsted AJ, Watek M, Wolf J, Slager S, Camp NJ, Canzian F, Vachon C. Genome-wide association study identifies variants at 16p13 associated with survival in multiple myeloma patients. Nature Communications. 2015;6:7539. doi: 10.1038/ncomms8539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zwick C, Held G, Auth M, Bernal-Mizrachi L, Roback JD, Sunay S, Iida S, Kuroda Y, Sakai A, Ziepert M, Ueda R, Pfreundschuh M, Preuss KD. Over one-third of African-American MGUS and multiple myeloma patients are carriers of hyperphosphorylated paratarg-7, an autosomal dominantly inherited risk factor for MGUS/MM. International Journal of Cancer. 2014;135:934–938. doi: 10.1002/ijc.28731. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Table S1
NIHMS792103-supplement-Supp_Table_S1.docx (20KB, docx)

RESOURCES