Melanoma and PD: New Findings Absolve Levodopa, Warrant Cancer Screening

February 1, 2007
Myra Partridge

The connection between Parkinson disease (PD) and melanoma is becoming increasingly apparent, leading some researchers to call for increased melanoma screening in the PD population. In addition, researchers are disproving previous theories that levodopa may be implicated in the link between melanoma and PD.

The connection between Parkinson disease(PD) and melanoma is becoming increasinglyapparent, leading some researchers to call forincreased melanoma screening in the PD population.In addition, researchers are disprovingprevious theories that levodopa may be implicatedin the link between melanoma and PD.

Patients with PD may be at more than twice the risk for the development of malignant melanoma than the general population, according to research by John M. Bertoni, MD, PhD, professor and chair of the Department of Neurology at Creighton University, Omaha, and colleagues.1 He presented the results of this research at the American Neurological Association's 131st Annual Meeting this past October in Chicago.

The study enrolled 2106 patients with idiopathic PD from 31 North American movement disorder centers. Prospective screening included a neurologic examination to obtain demographic information, medical history, and current medication use. Dermatologists then collected additional patient information on melanoma risk. Whole-body examinations and biopsies of suspicious lesions were performed.

Of these patients, 84.8% were receiving levodopa. Mean age of the patients was 68.6 years and mean duration of PD was 7.1 years. Of the 346 patients in whom suspicious pigmented lesions were identified, 20 in situ melanomas and 4 invasive melanomas were confirmed. Overall, the research team found that the prevalence of melanoma in patients with PD in North America is about 1.1%.

"We found that a lot of patients had melanoma of one kind or another," Bertoni said. Ninety-two patients (4.4%) had a new case or had previously received a diagnosis of melanoma.

Overall, the patients with PD and melanoma were found to have significantly more severe PD, be significantly older, and have significantly more melanoma risk factors. These risk factors included fair complexion, blue eyes, severe or blistering sunburn in childhood, sun sensitivity, freckles, blonde or red hair, other skin cancer in the past or present, inability to tan, 1 or more large or irregu- lar pigmented lesions, congenital moles, family history of melanoma, changes in moles, prior melanoma, immunosuppression, and whether the patient had undergone psoralen and ultraviolet A therapy or other radiotherapy.

Because of the results, Bertoni said he and fellow neurologists participating in this study are now warning patients with PD in this age group that they are at increased risk for melanoma. They recommend that patients undergo routine dermatologic examinations. "This was a gratifying study because most of the patients had a melanoma so minor that the biopsy we performed cured it," Bertoni added.

The prevalence of invasive melanoma in the 1692 US patients with PD included in this study was 2.2 times higher than age- and sex-matched populations derived from the national Surveillance Epidemiology and End Results (SEER) database, according to Bertoni. In addition, when patient data from his study were compared with patient data from the American Academy of Dermatology, risk of melanoma was found to be 8 times higher in Bertoni's group of patients.

The study also demonstrated no evidence of increased incidence of melanoma in patients receiving levodopa, compared with those not receiving it. However, Bertoni added that nearly 85% of the patients studied were treated with levodopa at some time. "So our study doesn't lay to rest the theory that a connection might exist between levodopa and melanoma," he said. Although he and coauthors concluded that increased vigilance for melanoma in patients with PD is warranted, Bertoni conceded that his study is limited and still leaves many unanswered questions. He added that the in- cidence of melanoma is under- reported in the SEER database.

LEVODOPA AND MELANOMA
A link between levodopa and melanoma was first drawn in 1972.2 Since that time, additional case reports also have supported the suspected association. However, a recent review of the literature by Zanetti and colleagues3 stated that there is still no good evidence that levodopa therapy for PD causes melanoma. After analyzing 36 case reports, case-control studies, and prospective cohort studies, the researchers stated that the proposed link is substantially inconsistent. While the authors reported that there is evidence of excess risk of melanoma in patients with PD, they attributed this risk to genetic predisposition. They also pointed out that animal studies did not display any carcinogenic effects of levodopa, which has been shown to be antineoplastic.

Bertoni agrees that the early literature on this subject can be misleading. "As with most research, people who think there may be an association report it, and those who think there is no association do not report it," he said. "So, you have to be careful. These case reports may be distorted."

Levodopa is a substrate of tyrosine hydroxylase, which is involved in the production of dopamine and melanin. For this reason, researchers have previously proposed that levodopa may have the potential to activate the development of malignant melanoma, explained Bala V. Manyam, MD, former professor and director of the Plummer Movement Disorders Center at Texas A&M Health Science Center College of Medicine in Temple. However, the interaction of levodopa metabolism on melanoma development would include multiple chemical steps, which have never been established, he explained.

While Manyam was a researcher at Texas A&M, he and a colleague examined the published hypothesis that chemical changes could affect dopamine and melanin. "Eventually, we just gave up," Manyam said. "There is nothing available in science to prove any kind of connection." Manyam also pointed out that dopamine in the brain is different from dopamine in the body.

"Can there be a relationship between PD and melanoma? It is very difficult to say," he said. "I cannot link the chemistry of PD with the chemistry of levodopa, so I think that the connection is coincidental." While he admits that this is just his personal opinion, he also points out that there are many diseases that go together, such as diabetes and high blood pressure, but correlations are hard to prove.

Manyam also has conducted a study of levodopa in patients with PD to determine whether melanoma developed in these patients.4 His team reviewed literature on 43 cases from previously reported studies and compared these patients' experiences with those of 11 patients who were treated at the Scott & White Clinic at Texas A&M. His team concluded that patients with PD treated with levodopa were not at higher risk for melanoma and stated that the occurrence of PD and melanoma in these patients was most likely coincidental.

PUT DOWN THE PDR
Manyam first became interested in this topic when working in his clinical practice with a patient who had PD and melanoma. While the Physicians' Desk Reference (PDR) lists melanoma as a contraindication for levodopa treatment in patients with PD, Manyam continued to treat his patient with levodopa in accordance with the patient's request.

"My patient said he'd rather risk getting melanoma again rather than suffer from the symptoms of PD," said Manyam. "Patients often look at their immediate misery rather than a long-term fear, so this patient wanted to continue taking the drug. I continued to prescribe the medication, and I took care of this patient for a very long time."

He said that although research is showing that levodopa may not cause melanoma, the PDR recommendations will not be changed in the near future by the PDR's publisher, the American Pharmaceutical Association. "This is a reflection of our litigious society," Manyam said. "When I recommended levodopa for my patient, I was violating the PDR recommendation. So if my patient had gotten melanoma and sued me, I may not have won [the case], but I still prescribed the medication because it was in the best interest of my patient."

Manyam hopes that physicians will join Bertoni in taking the bold step of treating patients with PD and a previous history of mela- noma with levodopa. "An average physician in a private practice will follow PDR recommendations and be afraid to give the drug, so many patients may be deprived of the proper treatment simply because of the physician's fear of getting sued. The irony is that physicians aren't able to care properly for some of their patients anymore because of recommendations like this," Manyam remarked.

SUPPORTING STUDIES
Additional epidemiologic studies also have shown that treatment with levodopa does not contribute to the increased risk of melanoma in patients with PD. Jørgen H. Olsen, MD, DMSc, professor and research director at the Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, led a recent study that concluded that melanoma in patients with PD is unrelated to levodopa treatment.5

Of 314 patients with PD, 45 had malignant melanoma and 97 had basal cell or squamous cell carcinoma; the additional 172 patients were controls. A retrospective review of medical records showed a 4- to 5-fold increase in risk of malignant melanoma development in those patients in whom idiopathic PD was diagnosed.

The report also showed that there was no relationship between a patient's cumulative dose of levodopa and the development of melanoma. Of the 284 patients in whom information on levodopa treatment was available, the mean dose intake before cancer diagnosis was as follows:

  • Patients with melanoma, 1495 g (range, 33 to 6194 g)
  • Patients with basal cell or squamous cell carcinoma, 1313 g (range, 11 to 13,533 g)
  • Control group, 1307 g (range, 0 to 7375 g)

Olsen and colleagues speculated that common factors might be involved in the development of melanoma and PD. That is, melanoma and PD might be part of the same disease syndrome or else additional genetic determinants may be at work, such as those that could increase the vulnerability of the skin to ultraviolet radiation. The authors also suggest that genetic factors could contribute to specific behaviors that have been associated with PD development, such as aversion to cigarette smoking, which has been shown to have a protective effect.

Earlier studies by Olsen and colleagues also have discounted a link between levodopa and melanoma development in patients with PD.6,7 A study of the same database of Danish patients showed that the presence of all carcinomas in patients with PD was slightly higher (11.94%) than in the general population (11.55%).6 Further analysis of the sites of presentation showed that cancer patterns in these patients were more complex; there was a greater number of melanomas and a lower number of smoking-related cancers in these patients.

The study also showed that the relative risk of malignant melanoma was increased before the diagnosis of PD, said Olsen. "This weakens the suggested hypothesis that malignant melanoma is caused by the levodopa treatment or other treatments in PD," he said. "Today, the combined body of evidence, I think, strongly supports the hypothesis that constituents of tobacco smoke, perhaps nicotine, inhibit or delay the development of PD. This results in a low prevalence of smokers among patients and a low relative risk of smoking-related cancers."

This observation indicates that drugs may be developed to prevent or postpone PD someday, Olsen continued. However, except for cancers of the skin and female breast, the relative risk in this study was somewhat decreased for other site-specific cancers, such as cancer sites unrelated to tobacco smoking.

"This indicates either a negative surveillance bias in this particular group of patients or an inborn protective mechanism in patients with PD against cancer," he said. "If it is assumed that apoptosis in the substantia nigra, leading to PD, is a manifestation of the apoptotic potential of a patient in general, the development of PD might be considered a marker for a person's ability to induce cell death by apoptosis-to provide protection against the effects of mutagen exposures from the environment. But this is entirely speculative."

Another study by Olsen demonstrated that the rate of malignant melanoma among patients with PD was 2 times higher than the rate in the general population.7 The findings support the results of Bertoni's study. Olsen's study also found that the rate of skin carcinoma in patients with PD was increased by 25%.

"The 25% increased risk of nonmelanoma skin cancer seen in our cohort study also seems to be real and unrelated to treatment for PD, as risks are increased in patients up to decades before the diagnosis of PD," said Olsen. "The findings of Bertoni and colleagues seem to be in perfect accordance with the findings of the Danish studies, which state that an increased risk for malignant melanoma that most likely is caused by other factors than treatment. All available information points to the theory that the increased rate of malignant melanoma in patients with PD is unrelated to the treatment."

Olsen plans to continue his research in this area. He has just received governmental funds for a large study on the causes of PD, such as environmental causes as well as gene-environment interactions, which will be undertaken on Danish patients.

ALS, PD, AND MELANOMA
Mortality attributable to amyotrophic lateral sclerosis (ALS) and PD is higher in patients with melanoma than in the general population, according to 2 recent studies.8,9

A total of 53 deaths attributed to ALS and 129 deaths attributed to PD were identified among 127,037 patients with melanoma in an Australian national cancer registry, according to Peter Baade, PhD, biostatistician and senior research fellow at the Viertel Centre for Research in Cancer Control, Queensland Cancer Fund, in Brisbane, Australia.8 Compared with the general population, the risk of death from ALS among patients with melanoma was 70% higher. In addition, patients with melanoma had a nearly 3-fold increased risk of dying of PD than did the general population.

Another study, by D. Michal Freedman, JD, MPH, PhD, an epidemiologist at the National Cancer Institute in Bethesda, Maryland, and colleagues also revealed a possible association between mela- noma and ALS and PD.9 Using 9 population-based cancer registries from the SEER database, the researchers examined data on 1.9 million persons in whom cancer was diagnosed between 1973 and 2000 and who survived at least 1 year. Patients with melanoma had significantly higher mortality rates attributable to ALS and PD.

"While these studies are exploratory-and they do not actually prove anything-the consistency of results across 2 different countries and environments shows evidence that an association exists between melanoma and each of the 2 neurodegenerative diseases," Baade said. "These studies use descriptive population-based data to better describe possible associations between ALS and PD and melanoma. It is hoped that this will direct further clinical and genetic research to increase the knowledge base of the causes of these neurodegenerative diseases, which is currently very limited."

Currently, the Cancer Council Australia and the Australasian College of Dermatologists do not recommend mass or population-based screening for melanoma, said Baade. However, they do recommend that general practitioners develop surveillance programs for patients at high risk.

"If there is an increased risk of melanoma among patients with ALS and PD, then it would seem that this cohort would satisfy the high-risk criteria, and surveillance programs would be appropriate," Baade said. "However, our study found an increased risk of [mortality associated with] ALS and PD among patients with melanoma. This is different from finding an increased mortality risk of melanoma in patients with ALS and PD."

CONFOUNDING REPORTS
While most new studies show that there may be no link between melanoma and PD and levodopa, some researchers continue to link the treatment with the diseases. A 2006 French case report concluded that the development of a high number of melanomas in a patient raises suspicions about levodopa as a precursor in melanin synthesis.10 A male patient treated with levodopa and a dopa decarboxylase inhibitor for PD developed 22 cutaneous melanomas over a 4-year period. A genetic analysis of the patient demonstrated a CDKN2A mutation with loss of p16 activity. This supports the conclusion that multiple melanomas may be associated with genetic predisposition, according to the researchers. They suggest that screening for predisposing genetic factors should be increased in patients who acquire melanoma while being treated with levodopa.

Bertoni said that while most recent studies show that melanoma is most likely not associated with levodopa treatment, no definitive study has proved this theory. "But I've been around long enough to see that the things we believe might be disproved someday," said Bertoni. "It's like taking 2 steps forward and 1 step back. Let's see what the next study shows."

Olsen agreed that more studies are necessary to determine whether levodopa effects the development of melanoma in patients with PD. "If we assume that the increased risk of malignant melanoma in patients with PD is caused by treatment of this disease with levodopa, one would expect that the incidence of malignant melanoma would be increased in patients only after they begin treatment for PD," said Olsen. "However, if the increased risk is caused by environmental or genetic factors common to both diseases, we expect the association between PD and malignant melanoma would be bidirectional, so that the risk for malignant melanoma also would be increased before a diagnosis of PD."

Overall, research on the cause of PD has been less extensively investigated when compared with other chronic diseases, such as cancer and cardiovascular diseases, said Olsen. "This will probably begin to change due to the continuous increase in life expectancy in most industrialized countries, which has led to a rapid increase in the proportion of families that eventually becomes affected by the disease," he said. "In combination with increasing wealth, this has resulted in demands for a better understanding of the causes of PD and safer, more effective treatments."

At this point, research may be too preliminary for physicians to send all patients with PD to the dermatologist's door, said Man- yam. "We cause a lot of unnecessary anxiety among our patients," he said. "When I approach patients about this, I explain that melanoma screening may be beneficial, but the role of a physician is to heal, not to frighten. I don't like to subject my patients to all sorts of tests that may not be necessary."

But Bertoni feels the research his team has collected warrants serious consideration by patients and neurologists. "Suspicious skin lesions were found in about 25% of the patients in our study, and over 6% of all patients were found to have some form of skin cancer," he said. "Therefore some of us feel it may be advisable to have patients with PD see a dermatologist."

REFERENCES1. Bertoni JM, Arlette JP, Fernandez HH, et al. Parkinsons disease and melanoma: an epidemiologic evaluation. Presented at: American Neurological Association 131st Annual Meeting; October 10, 2006; Chicago.
2. Skibba JL, Pinckley J, Gilbert EF, Johnson RO. Multiple primary melanoma following administration of levodopa. Arch Pathol. 1972;93:556-561.
3. Zanetti R, Loria D, Rosso S. Melanoma, Parkinson's disease and levodopa: causal or spurious link? A review of the literature. Melanoma Res. 2006;16:201-206.
4. Fiala KH, Whetteckey J, Manyam BV. Malignant melanoma and levodopa in Parkinson's disease: causality or coincidence? Parkinsonism Relat Disord. 2003;9:321-327.
5. Olsen JH, Tangerud K, Wermuth L, et al. Treatment with levodopa and risk for malignant melanoma. Mov Disord. In press.
6. Olsen JH, Friis S, Frederiksen K. Malignant melanoma and other types of cancer preceding Parkinson disease. Epidemiology. 2006;17:582-587.
7. Olsen JH, Friis S, Frederiksen K, et al. Atypical cancer pattern in patients with Parkinson's disease. Br J Cancer. 2005;92:201-205.
8. Baade PD, Fritschi L, Freedman DM. Mortality due to amyotrophic lateral sclerosis and Parkinson's disease among melanoma patients. Neuroepidemiology. 2006;28:16-20.
9. Freedman DM, Travis LB, Gridley G, Kuncl RW. Amyotrophic lateral sclerosis mortality in 1.9 million US cancer survivors. Neuroepidemiology. 2005;25:176-180.
10. Templier I, Charles J, Combe MC, et al. CDKN2A gene mutation and loss of p16 protein activity in a patient on levodopa presenting sporadic multiple primary melanoma [in French]. Ann Dermatol Venereol. 2006;133:777-780.