Management of Melanoma During Pregnancy

Sancy A. Leachman, MD, PhD; Ryan Jackson, HBS; Mark J. Eliason, MD; April A. Larson, MD; Jean L. Bolognia, MD

Disclosures

Dermatology Nursing. 2007;19(2):145-152,161. 

In This Article

Background and Literature Review

The incidence of cancer diagnosed during pregnancy has been estimated at 1:1,000 (Donegan, 1983). The most common malignancies diagnosed during pregnancy are the same as those seen in nonpregnant women of the same age group: carcinomas of the breast and cervix followed by melanoma and carcinomas of the thyroid, lung, and ovary (Wu et al., 2005). In 2004, 55,100 melanomas were diagnosed in the United States (Jemal, Tiwari et al., 2004) and the American Cancer Society (2005) has estimated 62,190 in 2006 (27,930 of whom will be women) (Jemal et al., 2006). Melanoma is currently the sixth most common cancer in women, and for all cancers, it ranks third in rate of increase in incidence (Jemal, Clegg et al., 2004). Maternal age at the time of pregnancy is on the rise, and the risk of melanoma is higher in older women. For example, from 1991 to 2001, the number of first births in women 35 to 39 years of age increased by 36%, and the number increased by 70% among women 40 to 44 years of age (Heffner, 2004). The incidence of melanoma per 100,000 person-years increases from 1.7 among 15 to 19-year-old Caucasian females to 17.1 in women ages 40 to 44 years. These data suggest that the rate of melanoma during pregnancy is likely to increase because: (a) melanoma incidence rates continue to increase in the general population, (b) pregnant women appear to get melanoma at the same rate as the general population, (c) melanoma occurs at a higher rate in older women, and (d) women are becoming pregnant later in life.

Having a large number of moles (or atypical moles) is a risk factor for melanoma (Slade et al., 1995) and although melanocytic nevi can be precursors of melanoma (Skender-Kalnenas, English, & Heenan, 1995), numerous nevi are primarily a phenotypic marker of skin at risk. Therefore it is important to determine whether pregnancy has either direct or indirect effects on melanocytes or nevi. Pigmentary changes, such as melasma and linea nigra, occur during pregnancy, demonstrating that pregnancy induces increased pigment production by melanocytes. In addition, nevi have been reported to change during pregnancy, growing larger, smaller, darker or more atypical in appearance. However, it is less clear that these changes lead to an increased predisposition to melanoma.

Based upon interviews, 10% to 30% of women reported changes in the color or size of their nevi during pregnancy (Foucar, Bentley, Laube, & Rosai, 1985; Sanchez, Figueroa, & Rodriguez, 1984). However, because participants in these studies did not undergo a clinical examination, it is unclear if the changing lesions were nevi or other non-melanocytic lesions such as seborrheic keratoses or skin tags. In contrast, in a prospective study of nevi during pregnancy, Pennoyer et al. (1997) examined pregnant women, and they documented the size of each nevus on the back (the body area least likely to expand during pregnancy). They found that only 8/129 (6%) of the nevi changed between the first and third trimesters, with half of the nevi becoming smaller. Of the three patients advised to have biopsies done during the study, only one had a lesion observed to have changed in size. It was not malignant. It is important to note that patients with a personal history of melanoma or atypical mole syndrome were excluded from this study. While few investigators have examined the relationship between other melanoma risk factors and changing nevi during pregnancy, Ellis (1991) suggested that patients with dysplastic nevi are more susceptible to changes in their nevi during pregnancy. It is possible that pregnancy does not typically lead to progression of nevi toward melanoma, but that in a subset of already predisposed women (for example, with a genetic predisposition toward nevus development or melanoma) it has a negative effect. Additional investigation of melanoma risk factors in women who develop melanoma during pregnancy is needed.

It has been hypothesized that hormonal factors directly or indirectly facilitate the formation and growth of melanoma. Research by Neifeld and Lippman (1980) demonstrated that 24 of 111 human melanoma tumors had estrogen-binding activity. Additional evidence suggestive of a hormonal influence on melanoma included the observations that melanoma is rarely seen before puberty (Sadoff, Winkley, & Tyson, 1973) and women have a better prognosis than men (White, 1959). However, other studies revealed that the apparent estrogen binding in melanoma cells may have been an artifact, caused by an interaction between the sex hormones and non-receptor proteins (McCarty et al., 1980; Zava & Goldhirsch, 1983). Ultimately, Flowers, Seigler, McCarty, Konrath, and McCarty (1987) and Lecavalier, From, and Gaid (1990) used more sensitive monoclonal antibody techniques to confirm a complete lack of estrogen receptor expression in human melanoma cells.

Interestingly, work done recently by Morvillo et al. (2002) not only confirmed a lack of estrogen receptors in several melanoma cell lines and tissue specimens of metastatic melanoma, but these investigators also found evidence for testosterone receptors that, when activated, caused cellular proliferation. These data suggest that hormonal responsiveness to estrogen, if present, is not occurring by direct activation of estrogen receptors on melanoma cells. This does not eliminate the possibility that other hormones (for example, testosterone or growth factors produced during pregnancy) could play a role in melanoma development or progression, or the possibility that sex hormones could have an indirect effect. Indeed, estrogen can activate testosterone receptors on melanoma cells and this activation can be inhibited by tamoxifen (Morvillo et al., 2002). Thus, our understanding of the influence of sex hormones on melanoma development and progression is incomplete.

Another way of examining the effects of pregnancy hormones on melanoma is to determine melanoma risk in patients taking oral contraceptives or hormone replacement therapy. A systematic review of case-control studies by Pfahlberg, Hassan, Willie, Lausen, and Gefeller (1997) came to the conclusion that the use of oral contraceptives did not increase the risk of melanoma. Similarly studies by MacKie (1999) and Durvasula, Ahmed, Vashisht, and Studd (2002) did not find an elevated risk of melanoma among women on hormone replacement therapy. While some historical and observational data suggested that hormones played a role in melanoma progression, these more recent studies do not support this assertion. Based on these data, there is little reason to recommend avoiding oral contraceptive or hormone replacement therapy in the routine melanoma patient. However, as is always the case in medicine, there will be the occasional patient with an unusual history (for example, multiple primary cutaneous melanomas arising during pregnancy or therapy for infertility) where additional debate of the relative pros and cons will be necessary.

In addition to hormonal changes, the immune system is also altered during pregnancy. These immunologic adaptations, felt to be essential for safe fetal development, have been explored by researchers as a potential explanation for the development of malignancies during pregnancy. A review by Luppi (2003) hypothesized that many of the changes detected in leukocyte populations during pregnancy were consistent with a simultaneous suppression of the adaptive arm of the immune system and enhancement in innate immunity. This alteration of the T-cell mediated adaptive immunity is accomplished by monocytes that display signals which increase T-cell tolerance. When T-cell activity is suppressed in this manner, it may subsequently decrease tumor surveillance and allow tumor growth and progression. However, Luppi does acknow ledge that several other studies of the changes in distribution and number of maternal peripheral blood leukocytes found contradictory results; some found an increase, some no change, and some a de crease in CD8+ cells, and no change or a decrease in the CD4+ cells (Fiddes, O'Reilly, & Cetrulo, 1986; Kuhnert, Strohmeier, Stegmuller, & Halberstadt, 1998; Matthiesen, Berg, Ernerudh, & Skogh, 1995; Sabahi, Rola-Plesczcynski, O'Connell, & Frenkel, 1995). The role of immunologic tolerance induced by the pregnant state in tumors that develop during pregnancy remains poorly understood.

In the 1950s, Pack and Scharnagel (1951) published a case series of pregnant patients with melanoma with very poor clinical outcomes. These observations, as well as those of other clinicians at the time, led to the hypothesis that pregnancy itself negatively influenced prognosis in a patient with melanoma. This perception led to the recommendation of the sterilization for women who were diagnosed with melanoma during a past pregnancy (Byrd & McGanity, 1954). More recently, retrospective analyses have compared the overall survival, disease-free interval (DFI), tumor thickness, stage of diagnosis, ulceration, and frequency of lymph node involvement in pregnant and age-matched melanoma patients (see Table 1 ). It is important to note that the vast majority of patients with melanoma included in these studies had early, localized disease (stage I and II by the 2001 AJCC guidelines) (see Table 2 ). Therefore, the conclusions from these studies are not necessarily applicable to metastatic disease.

With regard to prognostic indicators of melanoma, in particular Breslow depth and DFI, there was considerable variability in the conclusions reached by these studies (see Table 1 ). For example some studies found that pregnant patients had significantly thicker lesions, (Mackie, Bufalino, Morabito, Sutherland, & Cascinelli, 1991; Reintgen, McCarty, Vollmer, Cox, & Seigler, 1985), while others reported only a trend (statistically nonsignificant) toward thicker lesions in pregnant patients (Daryanani et al., 2003; McManamny, Moss, Pocock, & Briggs, 1989; Wong, Sterns, Kopald, Nizze, & Morton, 1989). In addition, there was a disagreement as to whether a difference existed between the DFI in pregnant versus nonpregnant patients. However, all agreed that there was no difference in 5-year survival when the two groups were compared.

One Russian study by Trapeznikov, Khasanow, and Iavorskii (1987) [reviewed by Kjems and Krag (1993)] reported a statistically significant decrease in 10-year survival for pregnant melanoma patients (26%) compared with matched controls (43%). One hypothesis advanced was that nonstatistically significant trends toward a poorer survival for pregnant melanoma patients seen at a followup of 5 years become significant with longer followup periods (Antonelli, Dotters, Katz, & Kuller, 1996). However, a large retrospective cohort study of 185 Swedish pregnant melanoma patients with a followup period of up to 42.9 years (Lens et al., 2004) reported no difference in the overall survival of pregnant women compared to nonpregnant women. Also, Lens et al. (2004) reported that pregnancy occurring after a diagnosis of melanoma was not associated with an increased risk of death, as has been shown by a number of studies (see Table 1 ).

In summary, although case series in the 1950s and 1960s early studies reported more advanced lesions with a worse survival in pregnant patients with melanoma, recent, larger case-controlled studies with longer followup periods have not observed a difference in survival. Collectively, the literature fails to demonstrate that stage I-II melanoma behaves more aggressively in pregnant patients. It is important to note that even in the larger studies with longer followup periods, very few patients had metastatic disease. In the O'Meara, Cress, Xing, Danielsen, and Smith (2005) study, 14/412 had regional spread and 6/412 had distant or remote spread at the time of diagnosis. Hence, these conclusions do not necessarily apply to the prognosis of patients with stage III or IV disease.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....