- Email: [email protected]
Hereditary malignant melanoma: A unifying etiologic hypothesis
Hereditary Malignant Melanoma: A Unifying Etiologic Hypothesis Henry T. Lynch and Ramon M. Fusaro
ABSTRACT: The recent increase in incidence af malignant melanoma in many of the "sun belts" of the world has focused intense interest on its etiology. Although sunlight is an important factor, only a fraction of those individuals exposed to significantly large amounts of sunlight develop malignant melanoma. Genetic susceptibility undoubtedly plays an important etiologic role. We have put forth a multifaceted hypothesis, which we believe provides unification for the genetic-environmental interaction in the etiology of malignant melanoma. Genetic heterogeneity in the familial atypical multiple mole melanoma (FAMMM) syndrome and other single gene disorders that predispose to malignant melanoma are particularly stressed. The possibility of polygenic or multifactorial factors in the etiology of malignant melanoma remains to be elucidated. We propose a unifying hypothesis for the etiology of malignant melanoma (MM) (Fig. 1), which emphasizes that the occurrence of MM results from the interactions of two fundamental areas, namely, genetics and the environment [1, 2]. Logically, it embraces basic genetic principles of reduced gene penetrance, variable expressivity of phenotype, and genetic heterogeneity [1, 2]. Primary genetic perturbations (position effect, modifying genes), as well as extragenetic environmental interactions in consort with a plethora of mechanisms, including activated oncogenes, bear directly on melanoma carcinogenesis [3-8]. In accord with our hypothesis, we consider that the bulk of so called sporadic occurrences of melanoma must be scrutinized meticulously in consort with incomplete, as well as inaccurate, family histories, problems of nonpaternity, new germinal mutations, and occurrence of phenocopies [9]. The identification of hereditary forms of MM is not only complicated by a myriad of confounding variables but, moreover, it is significantly hampered by the limited knowledge of many physicians who attend these patients. If the spectrum of cutaneous phenotypic characteristics were easily definable, the diagnosis of hereditary melanoma would be more secure. However, variations in clinical expression of phenotype are extant and, thereby, require indepth clinical and genetic analysis of the patients and their family members.
From the Department of Preventive Medicine/Public Health (H.T.L.) and the Department of Dermatology (R.M.F.), Creighton University School of Medicine, and the Section of Dermatology(R.M.F.), Department of Internal Medicine, University of Nebraska Medical Center and The Hereditary Cancer Institute, Omaha, Nebraska. Address requests for reprints to Dr. Henry T. Lynch, Department of Preventive Medicine/ Public Health, Creighton University School of Medicine, California at 24th St., Omaha, NE 67178. Received February 27, 1985; accepted June 11, 1985.
301 © 1986 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Ave., New York. NY 10017
Cancer Genet Cytogenet20:301-304 (1986) 0165-4608/86/$03.50
•
phenocopJes
Q
0
• Q
0
i""'"
• ". " • " • •
Iz////I
new germinal mutations
SPORADIC
• 0
°
I
. i g
"~1 Q
/
/
I /
Po]ygenic
/
--..
~ /
~
Albinism XDP Retinoblastoma Werner's Syndrome IOM, site-specific Congenital nevocytic nevi Familial Nevus of Ota !Familial CMM, IOM, and other tumor spectrum in absence of FAMMM Other
/ ~
J
/ ~
..
Schematic diagram depicting the unifying etiologic hypothesis for the extant heterogeneity in hereditary MM. FAMMM, familial atypical multiple mole melanoma syndrome (suffixes indicate heterogeneous forms); XDP. xeroderma pigmentosum; IOM, intraocular malignant melanoma, site specific; CMM, cutaneous malignant melanoma.
Figure 1
:: iJ
FAMMM-1 "~ FAMMM-2|
t'~
Hereditary MM
303
The quest for valid diagnosis through these investigations is invariably not costeffective for either the patient or the practitioner in the usual setting of the clincian's office. Most clinicians cannot afford the time or personnel needed for these investigations. Additionally, there is a reluctance by many family members to consult a physician when they deem themselves "healthy." Some family members may even have a psychologic resistance to any investigatian, including even frank denial. The incidence of hereditary malignant melanoma is further complicated by enigmatic issues that obscure its true occurrence. The diagnosis of early cutaneous ma° lignant melanoma (CMM) in situ versus atypical dysplastic precursor nevi are issues that pathologists will continue to debate in direct proportion to the lack of clarity of definition [10, 11]. There is still dispute as to whether the majority of CMM cases arise de novo or from precursor melanocytic nevi that are banal or dysplastic [11]. The definition of classical precursor nevi in patients with the familial atypical multiple mole melanoma (FAMMM) syndrome is relatively easy, both clinically and histopathologically, thus signifying patients and their families with these stigmata who must receive attention of the highest possible priority, given their extraordinary need for cancer surveillance. However, as we extend the clinical and histologic criteria to their most minimal possible expression, the definition of a hereditary cancer-associated genodermatosis, such as the FAMMM, becomes proportionately obscure. Clinical and histologic experience with the FAMMM is still at an embryonic stage. Our experience [2] indicates that there is marked variation in all facets of phenotypes, both within and between families. Though still elusive etiologically, such observations in other cancer-associated genodermatoses have provided clues that were subsequently substantiated in support of heterogeneity. Thus, in Figure 1 we postulate heterogeneity for the FAMMM, as demonstrated by our attention to an allelic series of FAMMM genotypes. It is incumbent on investigators to study melanoma-prone families in sufficient depth to determine if patterns of dermatologic phenotype differ (i.e., FAMMM moles may have differing phenotypic characteristics within, as well as between, families). We need to elucidate if an allelic series of FAMMM genotypes exist that may even have unique biomarker determinants, not unlike complementation groups in xeroderma pigmentosum (XP) [12]. Researchers must also evaluate whether or not an integral tumor spectrum characterizes certain FAMMM genotypes or even other specific cancer syndromes. Xeroderma pigmentosum is more heterogeneous than originally perceived, as evidenced by the presence of at least eight subgroups that are based on the ability to repair DNA after ultraviolet (UV) irradiation of cultured skin fibroblasts. Cells deficient in excision repair (classic XP) have been classified as complementation groups A-G. At least one other group, the XP variant, exhibits defective postreplication repair. Capitalizing on this problem of heterogeneity, Lynch et al. [12] postulated that patients with XP of the complementation group C variety may be at greater risk for manifesting MM than other XP complementation groups. These observations also fit well with our unifying hypothesis in that they show that heterogeneity exists, even in such a rarely occurring hereditary disorder as XP. Furthermore, patients with a certain genetic subset (complementation group C) may be at increased susceptibility of MM. Albinism is another example of a hereditary disorder that predisposes to MM, and which is also heterogeneous. Whether or not patients with differing genetic variants of albinism have a greater susceptibility to MM remains elusive. [1]. In some of these disorders (XP, albinism, and hereditary retinoblastoma [13]), it is conceivable that they be influenced by the presence of variant FAMMM genotype(s) [1], which are now suspected to be in a significantly large part of our Cau-
304
1~ T. Lynch and R. M. ~'usar{~
casian population. In addition, differences in survival, as well as response to various therapeutic drugs and regimens, may be genotype dependent. Our hypothesis mandates e n v i r o n m e n t a l interaction with the basic h u m a n genetic pool. These e n v i r o n m e n t a l perturbations are influenced by the macromilieu of chemical, physical (particularly sunlight), and nutritional factors, and the micrornilieu of biologic agents, such oncogenes and other viruses of putative etiologic importance, which act in consort to modify the basic genetic expression of the hum a n disease [1]. Much work is yet to be accomplished before we can have an adequate compreh e n s i o n of m e l a n o m a genetics, inclusive of the FAMMM. Although we have stressed the importance of monogenic m e l a n o m a predisposing diseases, one must also consider the likelihood of polygenic or multifactorial etiology for a substantial proportion of the m a l a n o m a b u r d e n (Fig. 1). Failure to intensify our efforts in an eclectic m a n n e r in search of knowledge about m e l a n o m a genetics will merely perpetuate a narrow concept of hereditary MM [14].
REFERENCES 1. Lynch HT, Fusaro RM (1982): Cancer-Associated Genodermatoses. New York: Van Nostrand Reinhold. 2. Lynch I-IT, Fusaro RM, Danes BS, Kimberling WJ, Lynch JF (1983): A review of hereditary malignant melanoma, including biomarkers in familial atypical multiple mole melanoma syndrome. Cancer Genet Cytogenet 8:325-358. 3. Lynch HT, Fusaro RM, Pester J, Albano WA, Lynch JF (1983): Phenotypic variation in the familial atypical multiple mole melanoma syndrome (FAMMM). J Med Genet 29:25-29. 4. Lynch HT, Frichot BC, Lynch J~ (1978): Familial atypical multiple mole melanoma syndrome. J Med Genet 15:352-356. 5. Lynch HT, Krush AJ (1968): Hereditary and malignant melanoma: Implications for early cancer detection. Can Med Assoc J 99:17-21. 6. Lynch HT, Fusaro RM, Pester J, Lynch JF (1980): Familial atypical multiple mole melanoma (FAMMM) syndrome: Genetic heterogeneity and malignant melanoma. Br J Cancer 42:58-70. 7. Lynch HT, Frichot BC, Lynch PM, Lynch J, Guirgis HA (1975): Familial studies of malignant melanoma and associated cancer. Surg Gyn Obstet 141:517-522. 8. Lynch HT, Fusaro RM, Pester J, Oosterhuis JA, Went LN, Rumke P, Neering H, Lynch JF (1981): Tumor spectrum in the FAMMM syndrome. Br J Cancer 44:553-560. 9. Lynch HT, Fusaro RM (1984): The NIH consensus report on precursors to malignant melanoma: A difference of opinion. J AM Med Assoc 252:2872-2873. 10. Lynch HT, Fusaro RM (1984): The NIH consensus report on "precursors to malignant melanoma": A different perspective. Am J Dermatopathol 6(suppl 1):177-179. 11. Ackerman AB, Mihara I (1985): Dysplasia, dysplastic melanocytes, dysplastic nevi, the dysplastic nevus syndrome, and the relation between dysplastic nevi and malignant mel~ anoma. Hum Pathol 16:87-91. 12. Lynch HT, Fusaro RM, Johnson JA (1983): Xeroderma pigmentosum: Complementation group C and malignant melanoma. Arch Dermatol 120:175-179. 13. Strong LC, Herson J, Haas C, Elder K, Chakraborty R, Weiss KM, Majumder P (1984): Cancer mortality in relatives of retinoblastoma patients. J Natl Cancer Inst 73:303-311. 14. Consensus Conference (1984): Precursors to malignant melanoma. ] Am Med Assoc 251:1864-1866.
ABSTRACT: The recent increase in incidence af malignant melanoma in many of the "sun belts" of the world has focused intense interest on its etiology. Although sunlight is an important factor, only a fraction of those individuals exposed to significantly large amounts of sunlight develop malignant melanoma. Genetic susceptibility undoubtedly plays an important etiologic role. We have put forth a multifaceted hypothesis, which we believe provides unification for the genetic-environmental interaction in the etiology of malignant melanoma. Genetic heterogeneity in the familial atypical multiple mole melanoma (FAMMM) syndrome and other single gene disorders that predispose to malignant melanoma are particularly stressed. The possibility of polygenic or multifactorial factors in the etiology of malignant melanoma remains to be elucidated. We propose a unifying hypothesis for the etiology of malignant melanoma (MM) (Fig. 1), which emphasizes that the occurrence of MM results from the interactions of two fundamental areas, namely, genetics and the environment [1, 2]. Logically, it embraces basic genetic principles of reduced gene penetrance, variable expressivity of phenotype, and genetic heterogeneity [1, 2]. Primary genetic perturbations (position effect, modifying genes), as well as extragenetic environmental interactions in consort with a plethora of mechanisms, including activated oncogenes, bear directly on melanoma carcinogenesis [3-8]. In accord with our hypothesis, we consider that the bulk of so called sporadic occurrences of melanoma must be scrutinized meticulously in consort with incomplete, as well as inaccurate, family histories, problems of nonpaternity, new germinal mutations, and occurrence of phenocopies [9]. The identification of hereditary forms of MM is not only complicated by a myriad of confounding variables but, moreover, it is significantly hampered by the limited knowledge of many physicians who attend these patients. If the spectrum of cutaneous phenotypic characteristics were easily definable, the diagnosis of hereditary melanoma would be more secure. However, variations in clinical expression of phenotype are extant and, thereby, require indepth clinical and genetic analysis of the patients and their family members.
From the Department of Preventive Medicine/Public Health (H.T.L.) and the Department of Dermatology (R.M.F.), Creighton University School of Medicine, and the Section of Dermatology(R.M.F.), Department of Internal Medicine, University of Nebraska Medical Center and The Hereditary Cancer Institute, Omaha, Nebraska. Address requests for reprints to Dr. Henry T. Lynch, Department of Preventive Medicine/ Public Health, Creighton University School of Medicine, California at 24th St., Omaha, NE 67178. Received February 27, 1985; accepted June 11, 1985.
301 © 1986 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Ave., New York. NY 10017
Cancer Genet Cytogenet20:301-304 (1986) 0165-4608/86/$03.50
•
phenocopJes
Q
0
• Q
0
i""'"
• ". " • " • •
Iz////I
new germinal mutations
SPORADIC
• 0
°
I
. i g
"~1 Q
/
/
I /
Po]ygenic
/
--..
~ /
~
Albinism XDP Retinoblastoma Werner's Syndrome IOM, site-specific Congenital nevocytic nevi Familial Nevus of Ota !Familial CMM, IOM, and other tumor spectrum in absence of FAMMM Other
/ ~
J
/ ~
..
Schematic diagram depicting the unifying etiologic hypothesis for the extant heterogeneity in hereditary MM. FAMMM, familial atypical multiple mole melanoma syndrome (suffixes indicate heterogeneous forms); XDP. xeroderma pigmentosum; IOM, intraocular malignant melanoma, site specific; CMM, cutaneous malignant melanoma.
Figure 1
:: iJ
FAMMM-1 "~ FAMMM-2|
t'~
Hereditary MM
303
The quest for valid diagnosis through these investigations is invariably not costeffective for either the patient or the practitioner in the usual setting of the clincian's office. Most clinicians cannot afford the time or personnel needed for these investigations. Additionally, there is a reluctance by many family members to consult a physician when they deem themselves "healthy." Some family members may even have a psychologic resistance to any investigatian, including even frank denial. The incidence of hereditary malignant melanoma is further complicated by enigmatic issues that obscure its true occurrence. The diagnosis of early cutaneous ma° lignant melanoma (CMM) in situ versus atypical dysplastic precursor nevi are issues that pathologists will continue to debate in direct proportion to the lack of clarity of definition [10, 11]. There is still dispute as to whether the majority of CMM cases arise de novo or from precursor melanocytic nevi that are banal or dysplastic [11]. The definition of classical precursor nevi in patients with the familial atypical multiple mole melanoma (FAMMM) syndrome is relatively easy, both clinically and histopathologically, thus signifying patients and their families with these stigmata who must receive attention of the highest possible priority, given their extraordinary need for cancer surveillance. However, as we extend the clinical and histologic criteria to their most minimal possible expression, the definition of a hereditary cancer-associated genodermatosis, such as the FAMMM, becomes proportionately obscure. Clinical and histologic experience with the FAMMM is still at an embryonic stage. Our experience [2] indicates that there is marked variation in all facets of phenotypes, both within and between families. Though still elusive etiologically, such observations in other cancer-associated genodermatoses have provided clues that were subsequently substantiated in support of heterogeneity. Thus, in Figure 1 we postulate heterogeneity for the FAMMM, as demonstrated by our attention to an allelic series of FAMMM genotypes. It is incumbent on investigators to study melanoma-prone families in sufficient depth to determine if patterns of dermatologic phenotype differ (i.e., FAMMM moles may have differing phenotypic characteristics within, as well as between, families). We need to elucidate if an allelic series of FAMMM genotypes exist that may even have unique biomarker determinants, not unlike complementation groups in xeroderma pigmentosum (XP) [12]. Researchers must also evaluate whether or not an integral tumor spectrum characterizes certain FAMMM genotypes or even other specific cancer syndromes. Xeroderma pigmentosum is more heterogeneous than originally perceived, as evidenced by the presence of at least eight subgroups that are based on the ability to repair DNA after ultraviolet (UV) irradiation of cultured skin fibroblasts. Cells deficient in excision repair (classic XP) have been classified as complementation groups A-G. At least one other group, the XP variant, exhibits defective postreplication repair. Capitalizing on this problem of heterogeneity, Lynch et al. [12] postulated that patients with XP of the complementation group C variety may be at greater risk for manifesting MM than other XP complementation groups. These observations also fit well with our unifying hypothesis in that they show that heterogeneity exists, even in such a rarely occurring hereditary disorder as XP. Furthermore, patients with a certain genetic subset (complementation group C) may be at increased susceptibility of MM. Albinism is another example of a hereditary disorder that predisposes to MM, and which is also heterogeneous. Whether or not patients with differing genetic variants of albinism have a greater susceptibility to MM remains elusive. [1]. In some of these disorders (XP, albinism, and hereditary retinoblastoma [13]), it is conceivable that they be influenced by the presence of variant FAMMM genotype(s) [1], which are now suspected to be in a significantly large part of our Cau-
304
1~ T. Lynch and R. M. ~'usar{~
casian population. In addition, differences in survival, as well as response to various therapeutic drugs and regimens, may be genotype dependent. Our hypothesis mandates e n v i r o n m e n t a l interaction with the basic h u m a n genetic pool. These e n v i r o n m e n t a l perturbations are influenced by the macromilieu of chemical, physical (particularly sunlight), and nutritional factors, and the micrornilieu of biologic agents, such oncogenes and other viruses of putative etiologic importance, which act in consort to modify the basic genetic expression of the hum a n disease [1]. Much work is yet to be accomplished before we can have an adequate compreh e n s i o n of m e l a n o m a genetics, inclusive of the FAMMM. Although we have stressed the importance of monogenic m e l a n o m a predisposing diseases, one must also consider the likelihood of polygenic or multifactorial etiology for a substantial proportion of the m a l a n o m a b u r d e n (Fig. 1). Failure to intensify our efforts in an eclectic m a n n e r in search of knowledge about m e l a n o m a genetics will merely perpetuate a narrow concept of hereditary MM [14].
REFERENCES 1. Lynch HT, Fusaro RM (1982): Cancer-Associated Genodermatoses. New York: Van Nostrand Reinhold. 2. Lynch I-IT, Fusaro RM, Danes BS, Kimberling WJ, Lynch JF (1983): A review of hereditary malignant melanoma, including biomarkers in familial atypical multiple mole melanoma syndrome. Cancer Genet Cytogenet 8:325-358. 3. Lynch HT, Fusaro RM, Pester J, Albano WA, Lynch JF (1983): Phenotypic variation in the familial atypical multiple mole melanoma syndrome (FAMMM). J Med Genet 29:25-29. 4. Lynch HT, Frichot BC, Lynch J~ (1978): Familial atypical multiple mole melanoma syndrome. J Med Genet 15:352-356. 5. Lynch HT, Krush AJ (1968): Hereditary and malignant melanoma: Implications for early cancer detection. Can Med Assoc J 99:17-21. 6. Lynch HT, Fusaro RM, Pester J, Lynch JF (1980): Familial atypical multiple mole melanoma (FAMMM) syndrome: Genetic heterogeneity and malignant melanoma. Br J Cancer 42:58-70. 7. Lynch HT, Frichot BC, Lynch PM, Lynch J, Guirgis HA (1975): Familial studies of malignant melanoma and associated cancer. Surg Gyn Obstet 141:517-522. 8. Lynch HT, Fusaro RM, Pester J, Oosterhuis JA, Went LN, Rumke P, Neering H, Lynch JF (1981): Tumor spectrum in the FAMMM syndrome. Br J Cancer 44:553-560. 9. Lynch HT, Fusaro RM (1984): The NIH consensus report on precursors to malignant melanoma: A difference of opinion. J AM Med Assoc 252:2872-2873. 10. Lynch HT, Fusaro RM (1984): The NIH consensus report on "precursors to malignant melanoma": A different perspective. Am J Dermatopathol 6(suppl 1):177-179. 11. Ackerman AB, Mihara I (1985): Dysplasia, dysplastic melanocytes, dysplastic nevi, the dysplastic nevus syndrome, and the relation between dysplastic nevi and malignant mel~ anoma. Hum Pathol 16:87-91. 12. Lynch HT, Fusaro RM, Johnson JA (1983): Xeroderma pigmentosum: Complementation group C and malignant melanoma. Arch Dermatol 120:175-179. 13. Strong LC, Herson J, Haas C, Elder K, Chakraborty R, Weiss KM, Majumder P (1984): Cancer mortality in relatives of retinoblastoma patients. J Natl Cancer Inst 73:303-311. 14. Consensus Conference (1984): Precursors to malignant melanoma. ] Am Med Assoc 251:1864-1866.