CME #111 Chemoprevention
Therapies for Non-Melanoma Skin Cancers

Skin & Aging is proud to bring you this latest installment in its CME series. The series, brought to you by HMP Communications, consists of regular CME activities that qualify you for two Category One Physician Credit Hours. As a reader of Skin & Aging, you aren’t required to pay a processing fee for this course. Non-melanoma skin cancers, the most common malignancies in the United States, are expected to increase in incidence as the population ages. Prevention is key and one area being studied now is chemoprevention. This month, Keren Baron, M.D., Meyer Horn, M.D., and Kenneth Gordon, M.D., examine the latest research and how chemoprevention can help prevent or slow down the disease state. At the end of this article, you’ll find a 10-question exam. Please mark your responses on the postage-paid postcard and return it to HMP Communications. About 1 month after the publication date, we’ll post this course on Skin & Aging’s Web site — www.skinandaging.com. I hope this CME series contributes to your clinical skills. Cordially, Steven R. Feldman, M.D., Ph.D. CME Editor, Skin & Aging How to obtain two Category One Physician Credits by reading this article Instructions: Physicians may receive two Category One Physician Credits by reading the article on p.50-61 and successfully answering the questions found on p. 62. Use the postage-paid card provided to submit your answers, or log on to www.skinandaging.com and respond electronically. Accreditation: HMP Communications, LLC, is accredited by the Accreditation Council for Continuing Medical Education to sponsor continuing medical education for physicians. Designation: HMP Communications, LLC, designates that this continuing education offering meets the criteria for two hours in Category One of the Physicians Recognition Award of the American Medical Association. Disclosure Policy: All faculty participating in Continuing Medical Education programs sponsored by HMP Communications, LLC, are expected to disclose to the audience any real or apparent conflicts of interest related to the content of their presentation. Grading: Answers to the CME exam will be graded by HMP Communications, LLC. Within 60 days, you will be advised that you have passed or failed the exam. A score of 70% or above will comprise a passing grade. A certificate will be awarded to participants who successfully complete the exam. Learning objectives: At the conclusion of this activity, the participant should be able to: • describe the concepts of primary and secondary chemoprevention and how they apply to the model of skin cancer. • delineate the mechanisms of carcinogenesis in the skin. • describe various agents that may have chemopreventative effects for skin cancer and integrate them into daily practice. This CME activity was planned and produced in accordance with ACCME essentials. Sponsored by HMP Communications, LLC. *All photos used with permission from Alex T. Hirsh, MD, and the Resident Photo Archives of Northwestern University Department of Dermatology. N on-melanoma skin cancers are the most common malignancies in the United States. As the population ages and the expected incidence of these cancers increase, preventing development of these malignancies becomes an important public health concern. One current area under intense study is the development of chemopreventive techniques — the use of pharmacologic or dietary intervention to prevent or retard the progression of a disease state. In cancer, chemoprevention may be accomplished by stopping the progression of premalignant cell populations or by reversing a malignant state. Skin cancer is an important model system for the study of chemoprevention. The ease with which clinical and histological changes may be followed in the skin is a great advantage over malignancies of other organ systems for a number of reasons. For one, actinic keratoses (AKs) are pre-malignant lesions that not only can progress to squamous cell carcinoma (SCC) but can also identify patients at risk to develop this cancer. Hereditary syndromes, such as xeroderma pigmentosum (XP) and basal cell nevus syndrome, also define populations at great risk for the development of non-melanoma skin cancer (NMSC). Finally, the primary carcinogen in skin cancer, ultraviolet radiation (UVR), is well studied in both humans and in animal models. Premalignant lesions are central to the study of chemoprevention. AKs are clones of atypical keratinocytes that are confined to the epidermis. Once these clones progress into the dermis, they are histologically indistinguishable from SCCs.1 It’s estimated that as many as 97% of SCCs may be associated with a contiguous AK2 and that individual AK lesions have a 0.075% to 0.096% chance per year of progressing to a SCC.3 With this well-defined pre-malignant lesion, SCC is likely the best candidate tumor for chemoprevention-related investigations. Basal cell carcinoma (BCC) is similar to SCC in that the highest risk is due to UVR exposure but there is no defined, pre-malignant lesion. The relationship of melanoma, other than lentigo maligna melanoma, to UVR is not as well delineated as for NMSC, but it’s likely that this relationship exists.4-6 The designation of the dysplastic nevus as a pre-malignant lesion is controversial. Photocarcinogenesis The development of NMSC is a classic example of the multi-stage process of carcinogenesis. These stages are initiation, promotion and progression. Unlike many other cancers, NMSC has a common initiator and promoter, UVR. When UVB strikes the skin, DNA is damaged directly by photons absorbed by constituent skin components, and indirectly by free radical oxygen species. Photoproducts and the resultant pyrimidine base substitutions are formed. The hallmark photoproduct, the cyclobutane pyrimidine dimer (CPD), leads to abnormal kinks in DNA. Cytosine to thymidine (C to T) transition mutations result, and serve as “molecular fingerprints” for UVB-damaged DNA.7 Fingerprint mutations are crucial in that they enable us to follow the damaging affects of UVR into important areas of cell regulation,8 and will consequently allow for the formulation of directed chemoprevention. Characteristic UVA-induced photoproducts are not as well delineated, but may exist. UVR-related initiation is instigated by a DNA mutation in an area of the genome crucial to cell cycle regulation. The best known of these mutations occurs in the p53 tumor suppressor gene. Studies show that skin shielded from the sun reveals few p53 mutations, whereas chronically sun exposed sites strongly stain positive for p53 without clinically apparent AKs or SCCs.8 This abnormal p53 expression is consistent with the concept of “field cancerization,” which implies that tissues exposed to an initiator are abnormal and at risk for undergoing malignant transformation. Thus, AKs act as initiated clones of p53-mutated keratinocytes. As their cell growth and differentiation proceed unchecked, AKs also represent the active promotion stage. In situ and invasive SCC correspond to the progression stage.9 Although BCCs and melanoma do not follow this simple model of carcinogenesis, p53 mutations have been found in up to 50%10 and 13%11 of these cancers, respectively. Moreover, UVB fingerprint mutations are known to exist in the PTC (“patched”) gene locus of BCCs, an area coding for an important cell regulatory protein.12 This supports the action of UVB as an initiator in BCC as well. Another important factor in the development of skin cancer is the activity of the cutaneous immune system. Since UVR is ubiquitous, there is a need to eliminate potential neoplasms prior to their development. The cutaneous immune system can eliminate many tumors in early stages, before they become invasive. However, UVR itself induces marked immunosuppression and decreased immune surveillance,13-16 further potentiating its role as a tumor promoter. The importance of intact immune surveillance is demonstrated in patients with cutaneous immune suppression, such as a history of PUVA use17 or organ transplant,18 where increased incidences of NMSC have been observed. Chemoprevention Chemoprevention of skin cancer is accomplished when any step in the carcinogenic pathway is blocked. Inhibiting either initiation or promotion can be a successful strategy to prevent the development of invasive carcinoma. Chemoprevention can be further broken down into primary and secondary chemoprevention. Primary chemoprevention is blocking the formation of a first, invasive carcinoma while secondary chemoprevention is inhibition of future tumors in patients who have already had a skin cancer. Blocking Initiation The most logical method to prevent skin cancer is to avoid or inhibit the primary initiating agent, UVR. Patients can accomplish this with strict sun avoidance or by using blocking techniques such as UV-resistant clothing. These measures can prevent the initiation and promotion steps of photocarcinogenesis. However, these strategies are not generally employed in the United States. Chemical agents, such as those discussed below, have been developed to inhibit the initiation step in photocarcinogenesis. • Sunscreens. Sunscreens prevent carcinogenic UVR from interacting with keratinocytes. Chemical sunscreens work by absorbing UV, while physical agents, such as zinc oxide and titanium dioxide, block or reflect UV.19 Animal data suggest a decreased number of DNA mutations with use of sunscreen in mice. Moreover, UV-induced skin cancer in animal models is inhibited by application of sunscreens.20 In humans, long-term and consistent use of sunscreen has been shown to decrease the number of AK lesions in high-risk patients. Two controlled trials demonstrated a reduction in AK lesions in patients who were at high risk for these pre-malignant lesions after 121 or 222 years of sunscreen use. This lower incidence of AK lesions could lead to fewer potential SCCs over time. Moreover there is also data from a 4.5-year trial indicating that sunscreens, when used with beta-carotene, may have a significant effect in lowering the incidence of SCC.23 Sunscreen’s effect on the incidence of BCC or melanoma is unclear.24 Most commercially available sunscreens are comparatively more effective at limiting UVB than UVA penetration into the skin. However, as UVA exposure and its consequent cutaneous immunosuppression become clearer risk factors for NMSC, the protective effect of sunscreen has to be reconsidered. A new rating system for the potential effectiveness of sunscreens has been suggested to include immune protection factors, or IPF, and mutation protection factor, or MPF.25 Also, establishing standardized, effective, yet practical, UVA sunscreen testing methods to provide labeling that is understandable to consumers was recommended by the American Academy of Dermatology (AAD) consensus conference on UV protection of sunscreens in February 2000. • Laundering Methods and Dyeing. Protective clothing has been a long-time armor in the war against UV damage. A comparison study of white cotton fabrics treated with (a) water only, (b) detergent only, (c) detergent-UV absorber, and (d) dyes, revealed a 407% increased UV protection factor after five treatments with detergent-UV absorber, 544% after blue dye and 212% after yellow dye.26 Encourage all of your patients to wear dark-colored clothing in anticipation of sun exposure. Overall, the best protection is achieved with dry, loose-fitting and tightly woven fabrics of wool and/or polyester. Agents that Inhibit Promotion and Progression While agents that block UVR may inhibit progression of skin cancer as well as initiation, other agents may exert their chemopreventative effects primarily by inhibiting promotion and progression of the disease. The use of these types of agents, as outlined below, becomes particularly important in patients at high-risk for development of skin cancers, such as those with solid organ transplants on chronic immunosuppression, XP and basal cell nevus syndrome. Systemic Retinoids. These are the best-studied group of compounds for skin cancer chemoprevention. Their mechanism of action includes the promotion of normal maturation and differentiation of keratinocytes, and reversal of malignant transformation.27 This may occur through the inhibition of the AP-1 signal transduction pathway and/or suppression of COX-2 expression.28,29 Additionally, retinoids may decrease UV transmission through the skin by increasing epidermal thickness via up-regulation of epidermal growth factor receptors.30 They may act as immunoregulatory agents31 and exhibit anti-angiogenic32,33 and pro-apoptotic actions.34 Much of the data on retinoids and their role in chemoprevention is divided into risk-category of study subjects. • Moderate-Risk Individuals. Sys-temic retinoids are beneficial in reversing photoaging and UV damage,35 and the literature supports using these agents to prevent cutaneous cancers. The successful administration of retinoids in several randomized chemoprevention trials suggests that these agents have a significant effect in reversing pre-malignant skin lesions.36 The Southwest Skin Cancer Prevention Group (SKICAP) study is the first Phase III skin cancer prevention trial to demonstrate a decrease in the incidence of SCC. In one arm of the study, patients with a history of 10 or more AKs and no more than one SCC or BCC were given 25,000 units of retinol or placebo daily for 5 years. Retinol significantly reduced the 5-year probability of developing a new SCC in these moderate-risk individuals.37 There was no effect on the incidence of BCCs. • High-Risk Individuals. The use of systemic retinoids in patients at higher risk for NMSC secondary to immunosuppression or defective DNA repair has been well studied. Acitretin provided a significant chemoprophylactic effect for NMSCs for up to 4 years in renal transplant patients.38 Similarly, etretinate was found to diminish SCC risk in a similar group of patients.39 Five patients with XP used isotretinoin and showed an average reduction in NMSC incidence of 63%.40 Importantly, in all of these studies, the effects of systemic retinoids disappeared upon discontinuation of these drugs.41 These encouraging results are contrary to the secondary prevention arm of the SCC/BCC SKICAP study. This group, which included higher-risk individuals with a history of at least four prior NMSCs, showed no significant effect in reduction of NMSC incidence with 25,000 units of retinol, 5 mg to 10 mg of isotretinoin or placebo daily for 3 years.42 These findings may be explained by the much lower doses of retinoids tested in the SKICAP study as compared with administered dosages for XP or renal transplant patients. The results of the AK arm versus the SCC/BCC arm of the SKICAP study also suggest that retinol may be more effective in the relatively early stages of skin carcinogenesis. The systemic retinoids of the future will ideally possess all of the chemopreventative properties and fewer of the mucocutaneous side effects of our current agents. You can increase the efficacy of retinoids by combining these medications with other drugs for a synergistic effect. For example, isotretinoin has been shown to increase the efficacy of 5-FU in the treatment of AKs.43 We have reported on a patient we treated with interferon-a (alpha) and isotretinoin — a standard therapy for in-operable head and neck SCC — who had a significant reduction in lesions of AK and Bowen’s disease.44 More importantly, this patient showed no new lesions after the discontinuation of this therapy. To date, three people have been treated with this protocol with two of the three showing marked regression in lesions and all three showing a decrease in the number of new lesions for at least 1 year after the treatment was stopped (Gordon, unpublished data). The effects of retinoids in high-risk patients with BCC are less clear. Studies suggest that retinoid chemoprevention for BCC may only be appropriate for very high-risk patient groups, as patient populations with only moderate numbers of BCCs show no significant differences in BCC numbers with treatment.41,45 Topical Retinoids. Topical treatment with retinoids in humans has been shown to increase the number of Langerhans cells in the skin.46 Tretinoin has been thoroughly studied for treatment of photodamage with demonstration of clear benefit.47 Topical retinoids are also effective in decreasing the number of AKs with minimal toxicity.48 There have been no studies to date proving topical retinoids prevent skin cancer. In fact, a limited number of animal studies suggest that topical tretinoin enhances UV-induced skin tumor production under laboratory-defined conditions.49-51 For this reason, caution should be exercised in recommending topical retinoids in patients with ongoing significant sun exposure. No data are available on the newer synthetic topical retinoids. Cyclooxygenase Inhibition. The activation of the inducible form of cyclooxygenase, COX-2, is thought to be an important promotional event in AKs and SCCs. UVB exposure significantly up-regulates COX-2, and COX-2 protein expression is enhanced in epidermal cancer cells.52 Non-specific COX suppression with indomethacin has been shown to reduce UV-induced skin tumors by 78%.53 Specific inhibition of COX-2 with celecoxib demonstrated a comparable 89% reduction in skin tumor yields in mice exposed to carcinogenic UV. Mice given oral indomethacin daily during photocarcinogenesis had increased probability of remaining skin tumor-free and had a lower average tumor multiplicity. This pathway is also likely to explain how the NSAID, diclofenac sodium gel (Solaraze) exerts its beneficial effect on AKs. Like many of the compounds described above, however, no significant data are available for determining whether COX-2 specific inhibitors and non-specific COX inhibitors such as aspirin will actually decrease the incidence of NMSC in humans. A recent study suggests that the protective effects of retinoids and antioxidants on skin carcinogenesis may involve COX-2 expression.29 Antioxidants These substances are thought to exert their effects through the reduction and/or scavenging of reactive oxygen species that may damage cellular DNA and cause varying degrees of immune suppression. While data in animal models have supported enthusiasm over such agents for chemoprevention, human data and clinical experience is usually lacking. • Pseudocatalase. Catalase is an important free radical scavenger. Topically applied pseudocatalase (a non-polar, bis-manganese complex) may act as a substitute for catalase in the skin. Patients with XP are believed to be at greater risk from the effects of reactive oxygen species and represent an extreme case of low antioxidant defense capacity.54,55 To date, one XP patient treated with pseudocatalase cream had a significant reduction in NMSCs.56 • Dietary Substances. A number of dietary substances with purported antioxidant effects have been identified for their potential chemopreventative effects, including resveratrol, curcumin, silymarin, ginger, and diallyl sulfide.57 Though there have been encouraging results from animal studies, the chemopreventative effects of antioxidant dietary substance in humans are either inconclusive or unavailable. For example, oral beta-carotene supplementation has been extensively studied in humans and found to have no significant effect on reducing skin cancers.23,58,59 For more potential dietary antioxidants see “Dietary Substances with Purported Antioxidant Effects” on p. 56. • Green and Black Teas have received notoriety in recent years as potential antioxidant chemopreventative agents.60 The oral administration of both black and green teas to mice exposed to photocarcinogenic doses of UVB was associated with a decreased incidence of skin tumors.61,62 Black tea appeared to provide more protection against UVB induced tumors than green tea.60 More recently, a study in humans suggests that green tea polyphenols (GTPs), specifically (-)-epigallocatechin-3-gallate (EGCG) and (-)-epicatechin-3-gallate, effectively inhibit the erythema response, reduce the number of sunburn cells, protect epidermal Langerhans cells from UV damage, and reduce the DNA damage associated with UV radiation.63 Destruction of AKs It is common practice for dermatologists to view the destruction of the pre-malignant lesions of SCCs as a means to prevent the development of invasive carcinoma. While data clearly suggest that many SCCs develop from AKs, there is no strong data to support the view that destruction of AKs can act as a preventative procedure. In fact, the large-scale studies that would be necessary to establish whether destruction of AKs could have such an effect have not been performed. However, in other areas of medicine, destruction of premalignant lesions has been shown to be an effective method to prevent invasive carcinoma. For example, removing adenomatous polyps clearly protects against the development of colon cancer. Thus, the practice of eliminating of AKs is a reasonable, though unproven, method of preventing SCC. By far the most common method of elimination is physical or chemical destruction (See “Physically Destructive Methods in the Treatment of AKs” above). A number of agents developed to chemically destroy AK lesions are available and many others are under investigation. Here’s a look at what we have available. • 5-fluorouracil (5-FU) is the most common form of chemical destruction, even though this method is limited by patient discomfort and consequent low patient compliance. Rates of up to 93% elimination of clinical AK lesions have been reported.64 The combination of topical 5-FU and 0.05% tretinoin cream appears to enhance the efficacy of 5-FU,65 and 5-FU with oral isotretinoin has been shown to be highly effective for widespread AKs on photodamaged skin.43 The results of a 32-month study of trichloroacetic acid (TCA)-chemical peel versus 5-FU showed negligible difference between these two methods in the rate of recurrences or improvement of actinically damaged skin.66 Intermittent chemical peels with TCA or alpha-hydroxy acids may be particularly useful for less adherent patients. • Photodynamic therapy (PDT) refers to the topical application of aminolevulinic acid (ALA) in conjunction with light. The endogenous photosensitizer, protoporphyrin IX, preferentially accumulates in dysplastic cells resulting in tissue-specific phototoxicity with light.67 The use of 20% ALA solution with a non-laser fluorescent blue light source found 85% clearance of AKs after two treatments.68 In a comparison study of 5-FU and PDT in the treatment of AKs, a single treatment with PDT was as effective in clearing AKs as 3 weeks of twice-daily 5-FU. Both methods showed a mean reduction in lesion area of 70% or more.69 • Diclofenac sodium gel (Solaraze) is the first drug to contain the non-steroidal anti-inflammatory drug (NSAID) diclofenac sodium in a 3% topical formulation with hyaluronan to receive regulatory approval in the United States for the treating AKs. Early data from animal models demonstrated inhibition of angiogenesis and induction of neovascular regression in inflammatory tissue after being given the drug.70 One randomized, controlled trial demonstrated that 47% of patients receiving active treatment with 3% diclofenac sodium gel twice daily were completely clear of AKs at follow-up (30 days after a 90-day treatment period) as compared to 19% of patients in the placebo group.71 Destruction of AKs – Agents Under Study • Imiquimod (Aldara) is an immune response modifier approved to treat condyloma acu-minata. Imiquimod affects cell-mediated immunity by inducing interferon-alpha (INF-alpha), INF-gamma and interleukin 12.72 A study of six case reports of AKs treated with imiquimod 5% cream three times per week for 6 to 8 weeks demonstrated successful clearing of all lesions, without histological evidence of persistent AKs or recurrences reported during follow-up.73 • Alpha-2-(Difluoromethyl)-dl-orni-thine (DFMO) is an irreversible inhibitor of ornithine decarboxylase. DFMO has been shown to suppress skin carcinogenesis in mice after oral or topical administration. A study of 48 patients using a topical hydrophilic ointment formulation with 10% DFMO for 6 months showed a reduced number of AKs in high-risk participants, warranting its additional study as a skin cancer chemopreventative agent.74 A Phase II dose-ranging study of topical DFMO is being planned. • T4 endonuclease V (Dimericine) is a DNA repair enzyme that is delivered in liposomes in the form of a topical cream. It is thought to act via two mechanisms. Immediately, T4 endonuclease V removes DNA dimers, primarily cyclobutane pyrimidine type. In the long-term, it may restore p53 gene function and exert a lasting chemopreventative effect. This is in contrast to retinoids. This compound is reported to significantly decrease the development of AKs and BCCs in patients with XP.75 There is no available data regarding a decrease in the development of SCC or melanoma. Presently, the FDA is reviewing approval for XP patients. A second clinical trial will be conducted soon with T4 endonuclease V on at-risk patients. Future Treatments The future holds hope for developments in skin cancer chemoprevention. Newer broad-spectrum sunscreens and increased patient awareness about UVR may have a significant impact on cutaneous malignancy incidence. Many anti-oxidant compounds are deserving of large scale controlled studies in humans to assess their efficacy. Retinoids, both oral and topical, may still represent the most promising of the chemopreventative agents, and new advances in receptor-specific retinoids will maximize their therapeutic benefit while minimizing the adverse effects. DNA micro-array analysis will eventually lead to the discovery of new genes and the formulation of more directed therapies. Gene therapy for patients with defects in DNA repair may soon become a reality. Recent advances in both our understanding of skin cancer and chemoprevention are poised to significantly impact the incidence and morbidity of cutaneous neoplasms.