Phenotypic Heterogeneity in Multiple Myeloma Families
http://www.100md.com
《临床肿瘤学》
the Creighton University School of Medicine
University of Nebraska Medical Center, Omaha, NE
Weill Medical College of Cornell University
Our Lady of Mercy Cancer Center, New York Medical College, New York, NY
University of Chicago, Chicago, IL
Institut Paoli-Calmettes, Marseille, France
University of Toronto, Toronto, Ontario, Canada
ABSTRACT
PATIENTS AND METHODS: This observational study consisted of 39 families with multiple cases of MM or related disorders from four collaborating research centers. Each center followed its usual family study method. Probands were interviewed, and, when possible, cancers were verified by medical records and pathology review. A working pedigree was compiled on each family.
RESULTS: Seventeen families had affected members in two or more generations, and eight families had two or more affected members in a single generation. Four families had two or more members with plasma cell dyscrasias, with or without a single case of MM. In the remaining 10 families, a single MM case occurred with a family history of other cancers. Other cancers observed in family members included hematologic malignancies and solid tumors. In families with MM in multiple generations, there was a decrease in the age at MM diagnosis in successive generations.
CONCLUSION: The study of familial MM may provide insights into the pathogenesis and, ultimately, the control and prevention of MM and related disorders. Population-based epidemiologic studies are crucial, but because of the rarity of familial MM, a concerted case-finding approach may also be fruitful. Therefore, we propose an international consortium to study familial MM, and we invite all interested colleagues to participate.
INTRODUCTION
The estimated number of new MM patients in the United States is 15,270 for 2004, and the estimated mortality is 11,070 individuals.3 These figures demonstrate the dismal survival rate of patients with MM. MM constitutes approximately 10% to 15% of all hematologic malignancies and approximately 1% of all forms of cancer, and it accounts for approximately 2% of all cancer deaths and close to 20% of the deaths caused by hematologic malignancies.4 It is a disease of middle-aged and elderly persons, with approximately 40% of patients being less than 60 years of age at the time of diagnosis. Only 2% of cases occur in persons less than 40 years old.5
Among MM patients, a small but unknown fraction are familial. Chance occurrence, familial environmental exposures such as pesticides or solvents,5 and inherited factors may contribute to these familial clusters. Reports of community clusters of MM6,7 and connubial occurrences of MM8,9 suggest a role for environmental factors in the etiology of familial MM.
Our interest in familial MM was stimulated by our initial study of a single family10 (family 18, Fig 1) that contained a sibship of seven persons in whom three had histologically verified MM. Two other siblings had monoclonal gammopathy of undetermined significance (MGUS), as determined by immunoelectrophoresis of serum and urine, and one of these siblings was recently diagnosed with amyloidosis and has died. In addition, other relatives had acute lymphocytic leukemia, melanoma, and prostate cancer. These remarkable observations suggested the potential significance of inherited genetic factors in MM. Therefore, we organized this study of a larger series of familial MM to generate hypotheses about the mode of genetic transmission for MM. We had previously observed evidence of anticipation,11 which is the tendency in multigenerational families of earlier disease onset or greater disease severity in later generations. Therefore, we also aimed to study the phenomenon of anticipation in this larger collection of families.
PATIENTS AND METHODS
Each center followed its usual family study method. Thus, there was no uniformity in family ascertainment or data collection. Whenever possible, the probands from these centers were interviewed regarding details about their family histories of cancer as well as their personal medical histories. Particular attention was given to documentation of the diagnosis of MM when present in a proband and the proband's history of other types of cancer. When possible, cancers in other family members were also verified by medical records and pathology review, and a working pedigree was then compiled.
Analyses were performed using SAS/STAT for Windows (SAS Institute, Cary, NC). Pedigrees were classified by the number of MM cases observed and the number of generations in which they occurred. MM patients from the multipatient, multigeneration families were compared with the MM patients from the multipatient, single-generation families using the {chi}2 test of independence (for sex ratio) and analysis of variance (for age at diagnosis). Evidence for anticipation was assessed in families with members with MM in two or more successive generations using the paired t test. In families in which there were multiple patients with MM in a generation, the mean age of those patients was calculated and used in the analysis.
RESULTS
Myeloma Patients
There were 79 MM patients in 39 families. Forty-one of the MM patients (52%) were women. The mean age at diagnosis for the 68 patients with a known age at diagnosis was 61.5 years, and the median age was 62 years. Three of the families had four members with MM (families 1, 2, and 7), seven families had three members with MM (families 3 to 6 and 16 to 18), 15 families had two members with MM (families 8 to 15 and 19 to 25), and 13 families had only one member with MM (families 26 to 28 and 30 to 39). One of the families had no members with MM (family 29); however, two individuals in that family had Waldenstrom's macroglobulinemia. Finally, one of the individuals in family 18 (patient II-1) developed amyloidosis.
The 17 multipatient, multigeneration families included 46 MM patients. Twenty-six of these patients (57%) were women, and the mean age at diagnosis was 63.3 years. The eight multipatient, single-generation families included 19 MM patients. Ten of these patients (53%) were women, and the mean age at diagnosis was 59 years. Differences between these two groups in sex and age at diagnosis were not significant.
Other cancers were observed in some MM patients. Two MM patients had a history of colon cancer, and one MM-affected patient also had prostate cancer. Three MM patients had breast cancer, and one MM patient had malignant melanoma. The most common cancer types among the other family members were lymphoma and leukemia (28 patients), breast cancer (32 patients), colon cancer (18 patients), and pancreatic cancer (10 patients).
Sex Linkage
All possible combinations of MM-affected, parent-child pairs were observed in these pedigrees (mother-daughter, mother-son, father-daughter, and father-son). Among the MM-affected sibling pairs, all possible sex combinations were also observed (Table 1). In the parent-child pairs, there was a marginally significant sex correlation. Male patients more often had an affected son, whereas females more often had an affected daughter (Fisher's exact test, P < .05).
Anticipation
After exclusion of several families that had been included in a previous study of anticipation by one of the authors,11 11 of the families in this study had MM patients in two successive generations with known ages at diagnosis. In the first generation, the mean age was 75 years (range, 63 to 84 years), whereas in the second generation, the mean age was only 61 years (range, 51 to 77 years), which is significantly different (P < .003). There was only one family that had a third-generation patient (family 7). This patient was not included in the analysis, but we note that this third generation patient was 9 years younger at diagnosis than her affected parent.
DISCUSSION
The mode of transmission of a rare genetic trait may be reflected in the pattern of occurrence in pedigrees. X-linked recessive traits are reflected by affected grandfather-grandson pairs and brother-brother pairs, whereas X-linked dominant traits are reflected in an absence of father-son pairs. Autosomal dominant traits are reflected in frequent parent-child pairs, whereas for autosomal recessive traits, affected sibling pairs, including brother-sister pairs, are typical. Incomplete penetrance of the trait can obscure these patterns, and these same patterns can be associated with chance occurrence, familial environmental exposures, and multifactorial inheritance. We reviewed these 39 pedigrees to develop hypotheses about the mode of transmission of familial MM. We found that 17 of the 39 families had MM-affected members in multiple generations, and thus, we hypothesize that a dominantly inherited trait plays a role in this phenomenon. There were only eight families with multiple MM-affected members in a single generation, and the MM patients in these families did not differ in sex ratio or age at diagnosis from the MM patients in the multigeneration families. Thus, we hypothesize that a recessively inherited trait may play a smaller role, if any. There were no cases suggesting a role for an X-linked trait. Our finding of a sex correlation in parent-child pairs has no obvious implications for mode of inheritance. Overall, 52% of our MM patients were women, which is equivalent to or slightly more than in the general population in which 47% of MM patients are women.13 This finding argues against the phenotype being sex limited or sex influenced.
The families in this study do not reflect the reported high incidence of MM in African American families.4,12,14 There are only three families of African American descent in this study (families 2, 16, and 19). Aside from these, all the families are white, which includes one family of known Ashkenazi descent (family 17). We think it is unlikely that this reflects any predominance of whites among familial MM, but rather, it reflects ascertainment bias.
Typically, hereditary cancer occurs at an earlier age than sporadic cancers of the same type. Thus, it is notable that the MM patients in our families were younger at diagnosis than MM patients in the general population, where the median age is 71 years.13 However, the process through which these families were ascertained could have been biased in favor of younger patients, so the hypothesis that familial MM occurs at an earlier age than sporadic MM will need to be evaluated in population-based studies.
Our results provide new evidence for anticipation in familial MM, which is reflected in a lower age at diagnosis in successive generations. Two prior studies15,11 also noted a decreasing age at MM diagnosis in successive generations, a finding that suggested anticipation. Past observations of anticipation have often been attributed to ascertainment bias, but this phenomenon is also characteristic of certain types of genetic mutations, such as those involving unstable trinucleotide repeats. Thus, we hypothesize that this type of mutation underlies some part of familial MM.
A diagnosis of MM also carries a familial risk for other types of cancer. A case-control study by Brown et al12 showed a significantly increased risk of MM associated with a family history of any hematologic cancer (odds ratio = 1.7). Eriksson and Hallberg16 investigated the family histories of cancer in patients with hematologic malignancies from Sweden and found an increased risk of MM in persons with first-degree relatives with any nonhematologic cancer (relative risk [RR] = 1.21). In particular, the risk for MM increased with the occurrence of prostate cancer (RR = 3.11) or a brain tumor (RR = 6.61) in relatives.16 Evidence for an increased risk of MM in the relatives of carriers of BRCA1 and BRCA2 mutations has also been reported.17 Furthermore, Dilworth et al18 described a familial melanoma family in which a germline mutation of the CDKN2A (p16) gene was identified in a family member with MM. To determine whether the CDKN2A mutation was responsible for MM, these authors looked for loss of heterozygosity and found that the wild-type CDKN2A allele was lost in the malignant plasma cells. They suggested that germline mutations of CDKN2A may confer an increased susceptibility to MM, in addition to the well-know predisposition to melanoma and pancreatic cancer.
Observations like these suggest that familial MM might occur in the context of other well-known or currently unknown hereditary cancer syndromes involving cancers of other types. In the well-known hereditary cancer syndromes, such as those caused by defective BRCA1, BRCA2, APC, and mismatch-repair genes, there is an elevated risk for more than one type of cancer. Thus, we reviewed our 39 families for patterns suggesting the association of other types of cancer with familial MM.
Some of the MM families we report herein have a high frequency of other cancers, particularly lymphoma and leukemia and certain solid tumors. The most common cancer types seen were lymphoma and leukemia, breast cancer, colon cancer, and pancreatic cancer. However, many of these families were ascertained in research settings favoring the ascertainment of families with other cancer occurrences, so these patterns may be attributable to ascertainment bias. Because of the possible link with the CDKN2A mutation mentioned earlier, 10 MM-affected patients and two MGUS-affected patients in eight of our families (families 3, 8, 11, 12, 18, 20, 33, and 37), including several families with members with pancreatic cancer, were tested for CDKN2A gene mutations, but all patients were negative (D. Hogg, unpublished data). In addition, because of a family history of breast cancer, a MM-affected member of family 7 was tested for mutations in BRCA1 and BRCA2, and a pathogenic BRCA2 mutation was identified.19 The large number of breast cancer patients in these families, coupled with the identification of a BRCA2 mutation in a familial MM and breast cancer family, suggests the hypothesis that a subset of familial MM may be accounted for by BRCA2 mutations. The large number of hematologic cancers other than MM in our families suggests the hypothesis that another subset of familial MM may be accounted for by a syndrome that includes different types of hematologic cancer.
In three of the families (families 18, 26, and 29), four individuals were diagnosed with MGUS, which is believed to be a precursor of MM.5,20-22 However, most members of our families have not been evaluated for MGUS. Grosbois et al15 identified 15 families with multiple cases of MM, and three of the families also had members with MGUS. Shoenfeld et al23 found an increased incidence of immunoglobulin abnormalities among the healthy relatives of MM patients, thereby postulating genetic susceptibility to MM. MGUS is an age-related phenomenon and is present in approximately 10% of individuals in their tenth decade.2 Kuehl and Bergsagel2 indicate that MGUS progresses to MM at a rate of 1% per year, and progression of MGUS to MM cannot be prevented.10 Horwitz et al24 postulated that certain patients may have a hereditary predisposition to MGUS, thereby increasing their risk for developing MM. We hypothesize that MGUS may have a higher prevalence among the close relatives of familial MM patients than in the general population.
Our observations about familial MM are tenuous because of the relatively small number of known MM families and their nonsystematic ascertainment. Case-control studies have established that MM is familial,12 but large genetic-epidemiologic studies are needed to determine whether this familiality is caused by genetic factors, to describe the inheritance pattern(s) accurately, to establish the role of environmental exposures, and to answer other important questions. However, in parallel with this work, in-depth studies of multipatient families may also provide critical information, including the discovery of heritable genetic abnormalities associated with MM susceptibility. To advance this latter effort, we propose to organize a much larger MM family resource. This can best be accomplished by a consortium of collaborators committed to recruiting MM families for study. Therefore, we propose an international consortium to study familial MM, and we invite all interested colleagues to participate (contact Henry T. Lynch at htlynch@creighton.edu).
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We acknowledge the technical assistance of Tami Richardson-Nelson, BGS, who prepared the pedigrees of the 39 families.
NOTES
Supported by revenue from Nebraska cigarette taxes awarded to Creighton University and the University of Nebraska Medical Center by the Nebraska Department of Health and Human Services. Support was also provided through National Institutes of Health grant Nos. 1U01 CA86389 and CA36727.
The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the state of Nebraska or the Nebraska Department of Health and Human Services.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Foerster J, Paraskevas F: Multiple myeloma, in Lee RL, Foerster J, Lukens J, et al (eds): Wintrobe's Clinical Hematology. Philadelphia, PA, Lippincott Williams & Wilkins, 1999, pp 2631-2680
2. Kuehl WM, Bergsagel PL: Multiple myeloma: Evolving genetic events and host interactions. Nat Rev Cancer 2:175-187, 2002
3. Jemal A, Tiwari RC, Murray T, et al: Cancer statistics, 2004. CA Cancer J Clin 54:8-29, 2004
4. Malpas JS, Bergsagel DE, Kyle R, et al: Multiple Myeloma: Biology and Management. Oxford, United Kingdom, Oxford University Press, 1998
5. Morgan GJ, Davies FE, Linet M: Myeloma aetiology and epidemiology. Biomed Pharmacother 56:223-234, 2002
6. Kyle RA, Finkelstein S, Elveback LR, et al: Incidence of monoclonal proteins in a Minnesota community with a cluster of multiple myeloma. Blood 40:719-724, 1972
7. Ende M: Multiple myeloma: A cluster in Virginia Va Med 106:115-116, 1979
8. Kyle RA, Heath CW, Carbone P: Multiple myeloma in spouses. Arch Intern Med 127:944-946, 1971
9. Kyle RA, Greipp PR: Multiple myeloma: Houses and spouses. Cancer 51:735-739, 1983
10. Lynch HT, Sanger WG, Pirruccello S, et al: Familial multiple myeloma: A family study and review of the literature. J Natl Cancer Inst 93:1479-1483, 2001
11. Deshpande HA, Hu XP, Marino P, et al: Anticipation in familial plasma cell dyscrasias. Br J Haematol 103:696-703, 1998
12. Brown LM, Linet MS, Greenberg RS, et al: Multiple myeloma and family history of cancer among blacks and whites in the U.S. Cancer 85:2385-2390, 1999
13. Ries LAG, Eisner MP, Kosary CL, et al (eds): SEER cancer statistics review, 1975-2000. http://seer.cancer.gov/csr/1975_2000, 2003
14. Cohen HJ, Crawford J, Rao MK, et al: Racial differences in the prevalence of monoclonal gammopathy in a community-based sample of the elderly. Am J Med 104:439-444, 1998
15. Grosbois B, Jego P, Attal M, et al: Familial multiple myeloma: Report of fifteen families. Br J Haematol 105:768-770, 1999
16. Eriksson M, Hallberg B: Familial occurrence of hematologic malignancies and other diseases in multiple myeloma: A case-control study. Cancer Causes Control 3:63-67, 1992
17. Struewing JP, Hartge P, Wacholder S, et al: The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 336:1401-1408, 1997
18. Dilworth D, Liu L, Stewart K, et al: Germline CDKN2A mutation implicated in predisposition to multiple myeloma. Blood 95:1869-1871, 2000
19. Sobol H, Vey N, Sauvan R, et al: Familial multiple myeloma: A family study and review of the literature. J Natl Cancer Inst 94:461-463, 2002
20. Blade J, Lopez-Guillermo A, Rozman C, et al: Malignant transformation and life expectancy in monoclonal gammopathy of undetermined significance. Br J Haematol 81:391-394, 1992
21. Pasqualetti P, Casale R: Risk of malignant transformation in patients with monoclonal gammopathy of undetermined significance. Biomed Pharmacother 51:74-78, 1997
22. Kyle RA: "Benign" monoclonal gammopathy: After 20 to 35 years of follow-up. Mayo Clin Proc 68:26-36, 1993
23. Shoenfeld Y, Berliner S, Shaklai M, et al: Familial multiple myeloma: A review of thirty-seven families. Postgrad Med J 58:12-16, 1982
24. Horwitz LJ, Levy RN, Rosner F: Multiple myeloma in three siblings. Arch Intern Med 145:1449-1450, 1985(Henry T. Lynch, Patrice W)
University of Nebraska Medical Center, Omaha, NE
Weill Medical College of Cornell University
Our Lady of Mercy Cancer Center, New York Medical College, New York, NY
University of Chicago, Chicago, IL
Institut Paoli-Calmettes, Marseille, France
University of Toronto, Toronto, Ontario, Canada
ABSTRACT
PATIENTS AND METHODS: This observational study consisted of 39 families with multiple cases of MM or related disorders from four collaborating research centers. Each center followed its usual family study method. Probands were interviewed, and, when possible, cancers were verified by medical records and pathology review. A working pedigree was compiled on each family.
RESULTS: Seventeen families had affected members in two or more generations, and eight families had two or more affected members in a single generation. Four families had two or more members with plasma cell dyscrasias, with or without a single case of MM. In the remaining 10 families, a single MM case occurred with a family history of other cancers. Other cancers observed in family members included hematologic malignancies and solid tumors. In families with MM in multiple generations, there was a decrease in the age at MM diagnosis in successive generations.
CONCLUSION: The study of familial MM may provide insights into the pathogenesis and, ultimately, the control and prevention of MM and related disorders. Population-based epidemiologic studies are crucial, but because of the rarity of familial MM, a concerted case-finding approach may also be fruitful. Therefore, we propose an international consortium to study familial MM, and we invite all interested colleagues to participate.
INTRODUCTION
The estimated number of new MM patients in the United States is 15,270 for 2004, and the estimated mortality is 11,070 individuals.3 These figures demonstrate the dismal survival rate of patients with MM. MM constitutes approximately 10% to 15% of all hematologic malignancies and approximately 1% of all forms of cancer, and it accounts for approximately 2% of all cancer deaths and close to 20% of the deaths caused by hematologic malignancies.4 It is a disease of middle-aged and elderly persons, with approximately 40% of patients being less than 60 years of age at the time of diagnosis. Only 2% of cases occur in persons less than 40 years old.5
Among MM patients, a small but unknown fraction are familial. Chance occurrence, familial environmental exposures such as pesticides or solvents,5 and inherited factors may contribute to these familial clusters. Reports of community clusters of MM6,7 and connubial occurrences of MM8,9 suggest a role for environmental factors in the etiology of familial MM.
Our interest in familial MM was stimulated by our initial study of a single family10 (family 18, Fig 1) that contained a sibship of seven persons in whom three had histologically verified MM. Two other siblings had monoclonal gammopathy of undetermined significance (MGUS), as determined by immunoelectrophoresis of serum and urine, and one of these siblings was recently diagnosed with amyloidosis and has died. In addition, other relatives had acute lymphocytic leukemia, melanoma, and prostate cancer. These remarkable observations suggested the potential significance of inherited genetic factors in MM. Therefore, we organized this study of a larger series of familial MM to generate hypotheses about the mode of genetic transmission for MM. We had previously observed evidence of anticipation,11 which is the tendency in multigenerational families of earlier disease onset or greater disease severity in later generations. Therefore, we also aimed to study the phenomenon of anticipation in this larger collection of families.
PATIENTS AND METHODS
Each center followed its usual family study method. Thus, there was no uniformity in family ascertainment or data collection. Whenever possible, the probands from these centers were interviewed regarding details about their family histories of cancer as well as their personal medical histories. Particular attention was given to documentation of the diagnosis of MM when present in a proband and the proband's history of other types of cancer. When possible, cancers in other family members were also verified by medical records and pathology review, and a working pedigree was then compiled.
Analyses were performed using SAS/STAT for Windows (SAS Institute, Cary, NC). Pedigrees were classified by the number of MM cases observed and the number of generations in which they occurred. MM patients from the multipatient, multigeneration families were compared with the MM patients from the multipatient, single-generation families using the {chi}2 test of independence (for sex ratio) and analysis of variance (for age at diagnosis). Evidence for anticipation was assessed in families with members with MM in two or more successive generations using the paired t test. In families in which there were multiple patients with MM in a generation, the mean age of those patients was calculated and used in the analysis.
RESULTS
Myeloma Patients
There were 79 MM patients in 39 families. Forty-one of the MM patients (52%) were women. The mean age at diagnosis for the 68 patients with a known age at diagnosis was 61.5 years, and the median age was 62 years. Three of the families had four members with MM (families 1, 2, and 7), seven families had three members with MM (families 3 to 6 and 16 to 18), 15 families had two members with MM (families 8 to 15 and 19 to 25), and 13 families had only one member with MM (families 26 to 28 and 30 to 39). One of the families had no members with MM (family 29); however, two individuals in that family had Waldenstrom's macroglobulinemia. Finally, one of the individuals in family 18 (patient II-1) developed amyloidosis.
The 17 multipatient, multigeneration families included 46 MM patients. Twenty-six of these patients (57%) were women, and the mean age at diagnosis was 63.3 years. The eight multipatient, single-generation families included 19 MM patients. Ten of these patients (53%) were women, and the mean age at diagnosis was 59 years. Differences between these two groups in sex and age at diagnosis were not significant.
Other cancers were observed in some MM patients. Two MM patients had a history of colon cancer, and one MM-affected patient also had prostate cancer. Three MM patients had breast cancer, and one MM patient had malignant melanoma. The most common cancer types among the other family members were lymphoma and leukemia (28 patients), breast cancer (32 patients), colon cancer (18 patients), and pancreatic cancer (10 patients).
Sex Linkage
All possible combinations of MM-affected, parent-child pairs were observed in these pedigrees (mother-daughter, mother-son, father-daughter, and father-son). Among the MM-affected sibling pairs, all possible sex combinations were also observed (Table 1). In the parent-child pairs, there was a marginally significant sex correlation. Male patients more often had an affected son, whereas females more often had an affected daughter (Fisher's exact test, P < .05).
Anticipation
After exclusion of several families that had been included in a previous study of anticipation by one of the authors,11 11 of the families in this study had MM patients in two successive generations with known ages at diagnosis. In the first generation, the mean age was 75 years (range, 63 to 84 years), whereas in the second generation, the mean age was only 61 years (range, 51 to 77 years), which is significantly different (P < .003). There was only one family that had a third-generation patient (family 7). This patient was not included in the analysis, but we note that this third generation patient was 9 years younger at diagnosis than her affected parent.
DISCUSSION
The mode of transmission of a rare genetic trait may be reflected in the pattern of occurrence in pedigrees. X-linked recessive traits are reflected by affected grandfather-grandson pairs and brother-brother pairs, whereas X-linked dominant traits are reflected in an absence of father-son pairs. Autosomal dominant traits are reflected in frequent parent-child pairs, whereas for autosomal recessive traits, affected sibling pairs, including brother-sister pairs, are typical. Incomplete penetrance of the trait can obscure these patterns, and these same patterns can be associated with chance occurrence, familial environmental exposures, and multifactorial inheritance. We reviewed these 39 pedigrees to develop hypotheses about the mode of transmission of familial MM. We found that 17 of the 39 families had MM-affected members in multiple generations, and thus, we hypothesize that a dominantly inherited trait plays a role in this phenomenon. There were only eight families with multiple MM-affected members in a single generation, and the MM patients in these families did not differ in sex ratio or age at diagnosis from the MM patients in the multigeneration families. Thus, we hypothesize that a recessively inherited trait may play a smaller role, if any. There were no cases suggesting a role for an X-linked trait. Our finding of a sex correlation in parent-child pairs has no obvious implications for mode of inheritance. Overall, 52% of our MM patients were women, which is equivalent to or slightly more than in the general population in which 47% of MM patients are women.13 This finding argues against the phenotype being sex limited or sex influenced.
The families in this study do not reflect the reported high incidence of MM in African American families.4,12,14 There are only three families of African American descent in this study (families 2, 16, and 19). Aside from these, all the families are white, which includes one family of known Ashkenazi descent (family 17). We think it is unlikely that this reflects any predominance of whites among familial MM, but rather, it reflects ascertainment bias.
Typically, hereditary cancer occurs at an earlier age than sporadic cancers of the same type. Thus, it is notable that the MM patients in our families were younger at diagnosis than MM patients in the general population, where the median age is 71 years.13 However, the process through which these families were ascertained could have been biased in favor of younger patients, so the hypothesis that familial MM occurs at an earlier age than sporadic MM will need to be evaluated in population-based studies.
Our results provide new evidence for anticipation in familial MM, which is reflected in a lower age at diagnosis in successive generations. Two prior studies15,11 also noted a decreasing age at MM diagnosis in successive generations, a finding that suggested anticipation. Past observations of anticipation have often been attributed to ascertainment bias, but this phenomenon is also characteristic of certain types of genetic mutations, such as those involving unstable trinucleotide repeats. Thus, we hypothesize that this type of mutation underlies some part of familial MM.
A diagnosis of MM also carries a familial risk for other types of cancer. A case-control study by Brown et al12 showed a significantly increased risk of MM associated with a family history of any hematologic cancer (odds ratio = 1.7). Eriksson and Hallberg16 investigated the family histories of cancer in patients with hematologic malignancies from Sweden and found an increased risk of MM in persons with first-degree relatives with any nonhematologic cancer (relative risk [RR] = 1.21). In particular, the risk for MM increased with the occurrence of prostate cancer (RR = 3.11) or a brain tumor (RR = 6.61) in relatives.16 Evidence for an increased risk of MM in the relatives of carriers of BRCA1 and BRCA2 mutations has also been reported.17 Furthermore, Dilworth et al18 described a familial melanoma family in which a germline mutation of the CDKN2A (p16) gene was identified in a family member with MM. To determine whether the CDKN2A mutation was responsible for MM, these authors looked for loss of heterozygosity and found that the wild-type CDKN2A allele was lost in the malignant plasma cells. They suggested that germline mutations of CDKN2A may confer an increased susceptibility to MM, in addition to the well-know predisposition to melanoma and pancreatic cancer.
Observations like these suggest that familial MM might occur in the context of other well-known or currently unknown hereditary cancer syndromes involving cancers of other types. In the well-known hereditary cancer syndromes, such as those caused by defective BRCA1, BRCA2, APC, and mismatch-repair genes, there is an elevated risk for more than one type of cancer. Thus, we reviewed our 39 families for patterns suggesting the association of other types of cancer with familial MM.
Some of the MM families we report herein have a high frequency of other cancers, particularly lymphoma and leukemia and certain solid tumors. The most common cancer types seen were lymphoma and leukemia, breast cancer, colon cancer, and pancreatic cancer. However, many of these families were ascertained in research settings favoring the ascertainment of families with other cancer occurrences, so these patterns may be attributable to ascertainment bias. Because of the possible link with the CDKN2A mutation mentioned earlier, 10 MM-affected patients and two MGUS-affected patients in eight of our families (families 3, 8, 11, 12, 18, 20, 33, and 37), including several families with members with pancreatic cancer, were tested for CDKN2A gene mutations, but all patients were negative (D. Hogg, unpublished data). In addition, because of a family history of breast cancer, a MM-affected member of family 7 was tested for mutations in BRCA1 and BRCA2, and a pathogenic BRCA2 mutation was identified.19 The large number of breast cancer patients in these families, coupled with the identification of a BRCA2 mutation in a familial MM and breast cancer family, suggests the hypothesis that a subset of familial MM may be accounted for by BRCA2 mutations. The large number of hematologic cancers other than MM in our families suggests the hypothesis that another subset of familial MM may be accounted for by a syndrome that includes different types of hematologic cancer.
In three of the families (families 18, 26, and 29), four individuals were diagnosed with MGUS, which is believed to be a precursor of MM.5,20-22 However, most members of our families have not been evaluated for MGUS. Grosbois et al15 identified 15 families with multiple cases of MM, and three of the families also had members with MGUS. Shoenfeld et al23 found an increased incidence of immunoglobulin abnormalities among the healthy relatives of MM patients, thereby postulating genetic susceptibility to MM. MGUS is an age-related phenomenon and is present in approximately 10% of individuals in their tenth decade.2 Kuehl and Bergsagel2 indicate that MGUS progresses to MM at a rate of 1% per year, and progression of MGUS to MM cannot be prevented.10 Horwitz et al24 postulated that certain patients may have a hereditary predisposition to MGUS, thereby increasing their risk for developing MM. We hypothesize that MGUS may have a higher prevalence among the close relatives of familial MM patients than in the general population.
Our observations about familial MM are tenuous because of the relatively small number of known MM families and their nonsystematic ascertainment. Case-control studies have established that MM is familial,12 but large genetic-epidemiologic studies are needed to determine whether this familiality is caused by genetic factors, to describe the inheritance pattern(s) accurately, to establish the role of environmental exposures, and to answer other important questions. However, in parallel with this work, in-depth studies of multipatient families may also provide critical information, including the discovery of heritable genetic abnormalities associated with MM susceptibility. To advance this latter effort, we propose to organize a much larger MM family resource. This can best be accomplished by a consortium of collaborators committed to recruiting MM families for study. Therefore, we propose an international consortium to study familial MM, and we invite all interested colleagues to participate (contact Henry T. Lynch at htlynch@creighton.edu).
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We acknowledge the technical assistance of Tami Richardson-Nelson, BGS, who prepared the pedigrees of the 39 families.
NOTES
Supported by revenue from Nebraska cigarette taxes awarded to Creighton University and the University of Nebraska Medical Center by the Nebraska Department of Health and Human Services. Support was also provided through National Institutes of Health grant Nos. 1U01 CA86389 and CA36727.
The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the state of Nebraska or the Nebraska Department of Health and Human Services.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Foerster J, Paraskevas F: Multiple myeloma, in Lee RL, Foerster J, Lukens J, et al (eds): Wintrobe's Clinical Hematology. Philadelphia, PA, Lippincott Williams & Wilkins, 1999, pp 2631-2680
2. Kuehl WM, Bergsagel PL: Multiple myeloma: Evolving genetic events and host interactions. Nat Rev Cancer 2:175-187, 2002
3. Jemal A, Tiwari RC, Murray T, et al: Cancer statistics, 2004. CA Cancer J Clin 54:8-29, 2004
4. Malpas JS, Bergsagel DE, Kyle R, et al: Multiple Myeloma: Biology and Management. Oxford, United Kingdom, Oxford University Press, 1998
5. Morgan GJ, Davies FE, Linet M: Myeloma aetiology and epidemiology. Biomed Pharmacother 56:223-234, 2002
6. Kyle RA, Finkelstein S, Elveback LR, et al: Incidence of monoclonal proteins in a Minnesota community with a cluster of multiple myeloma. Blood 40:719-724, 1972
7. Ende M: Multiple myeloma: A cluster in Virginia Va Med 106:115-116, 1979
8. Kyle RA, Heath CW, Carbone P: Multiple myeloma in spouses. Arch Intern Med 127:944-946, 1971
9. Kyle RA, Greipp PR: Multiple myeloma: Houses and spouses. Cancer 51:735-739, 1983
10. Lynch HT, Sanger WG, Pirruccello S, et al: Familial multiple myeloma: A family study and review of the literature. J Natl Cancer Inst 93:1479-1483, 2001
11. Deshpande HA, Hu XP, Marino P, et al: Anticipation in familial plasma cell dyscrasias. Br J Haematol 103:696-703, 1998
12. Brown LM, Linet MS, Greenberg RS, et al: Multiple myeloma and family history of cancer among blacks and whites in the U.S. Cancer 85:2385-2390, 1999
13. Ries LAG, Eisner MP, Kosary CL, et al (eds): SEER cancer statistics review, 1975-2000. http://seer.cancer.gov/csr/1975_2000, 2003
14. Cohen HJ, Crawford J, Rao MK, et al: Racial differences in the prevalence of monoclonal gammopathy in a community-based sample of the elderly. Am J Med 104:439-444, 1998
15. Grosbois B, Jego P, Attal M, et al: Familial multiple myeloma: Report of fifteen families. Br J Haematol 105:768-770, 1999
16. Eriksson M, Hallberg B: Familial occurrence of hematologic malignancies and other diseases in multiple myeloma: A case-control study. Cancer Causes Control 3:63-67, 1992
17. Struewing JP, Hartge P, Wacholder S, et al: The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 336:1401-1408, 1997
18. Dilworth D, Liu L, Stewart K, et al: Germline CDKN2A mutation implicated in predisposition to multiple myeloma. Blood 95:1869-1871, 2000
19. Sobol H, Vey N, Sauvan R, et al: Familial multiple myeloma: A family study and review of the literature. J Natl Cancer Inst 94:461-463, 2002
20. Blade J, Lopez-Guillermo A, Rozman C, et al: Malignant transformation and life expectancy in monoclonal gammopathy of undetermined significance. Br J Haematol 81:391-394, 1992
21. Pasqualetti P, Casale R: Risk of malignant transformation in patients with monoclonal gammopathy of undetermined significance. Biomed Pharmacother 51:74-78, 1997
22. Kyle RA: "Benign" monoclonal gammopathy: After 20 to 35 years of follow-up. Mayo Clin Proc 68:26-36, 1993
23. Shoenfeld Y, Berliner S, Shaklai M, et al: Familial multiple myeloma: A review of thirty-seven families. Postgrad Med J 58:12-16, 1982
24. Horwitz LJ, Levy RN, Rosner F: Multiple myeloma in three siblings. Arch Intern Med 145:1449-1450, 1985(Henry T. Lynch, Patrice W)