The vast majority of the two to three million non-melanoma skin cancers and 132,000 melanoma skin cancers that, according to the World Health Organization, occur globally each year can be managed through excision. However, for patients with unresectable, recurrent, or advanced skin cancers, treatment options have been limited. Recent pharmacological developments are expanding the treatment options for advanced cutaneous malignancies, including recently approved and late-stage agents. At the same time, data indicate that established therapies like nonsteroidal anti-inflammatory drugs may eventually play a role in prevention.
Ipilimumab (Yervoy, Bristol-Myers Squib) is now available for the first- or second-line treatment of patients with malignant melanoma. It is an intravenously administered anti-cytotoxic T-lymphocyte antigen (CTLA)-4 monoclonal antibody. By blocking CTLA-4, the agent is thought to enhance immune response to tumors. Ipilimumab is administered over 90 minutes every three weeks for a total of four infusions.
Several press accounts indicated that ipilimumab is the first therapy shown to increase survival in melanoma patients. In the phase III international study of 676 patients with melanoma, those who received ipilimumab alone or in combination with the experimental tumor vaccine gp100 had an average survival time of 3.5 months longer than those who received the vaccine alone. All subjects in the trial had metastatic or inoperable disease and had ceased to respond to other melanoma therapies.
While survival is a key clinical goal for patients, it is also an appropriate endpoint for a clinical trial of an immunotherapy; quantifying response to immune-targeted therapy may be more nuanced than assessment of conventional therapies. Pennock, et al.1 note that patient response to ipilimumab therapy occurred in four ways:
1. response in baseline lesions;
2. a slow, steady decline in tumor burden;
3. response after an increase in tumor burden; and
4. response in index and new lesions accompanied by the appearance of other new lesions.
These authors properly acknowledge that the latter two response patterns “would be classified as progressive disease using conventional assessments.” However, data indicate that patients who continue ipilimumab therapy in those scenarios could continue to see treatment benefit. Therefore, careful consideration must be given to the notion of “non-response” and treatment withdrawal.
Pooled analysis of data from multiple trials suggests that HLA subtype has no impact on treatment response, as is expected given that ipilimumab targets CTLA-4. Because the phase III trials involved use of the gp100 vaccine, which was previously shown to bind only to HLA-A*0201, only HLAA* 0201 positive subjects were enrolled.2 Other trials of ipilimumab did not consider subjects' HLA sub-type as an inclusion or exclusion criteria, though it was noted. Analysis of data showed no difference in overall survival or immune-related or treatment-induced adverse events based on HLA sub-type.2
The latest data regarding ipilimumab show that use in combination with dacarbazine increased overall survival compared to dacarbazine monotherapy.3 For the study, 502 patients with previously untreated metastatic melanoma were randomized one-to-one to treatment with ipilimumab (10mg per kilogram) plus dacarbazine (850mg per square meter of body-surface area) or dacarbazine (850mg per square meter) plus placebo, at weeks 1, 4, 7, and 10, followed by dacarbazine alone every three weeks through week 22. At 22 weeks, patients with stable disease or an objective response and no dose-limiting toxic effects received ipilimumab or placebo every 12 weeks as maintenance therapy. Overall survival in the ipilimumab plus dacarbazine group was 11.2 months, significantly higher than in the dacarbazine alone group (9.1 months). One-, two-, and three-year survival rates were also significantly higher for combination therapy versus dacarbazine alone (47.3 percent vs. 36.3 percent; 28.5 percent vs. 17.9 percent; and 20.8 percent vs. 12.2 percent, respectively). It may be noted that the 11.2-month survival rate for ipilimumab plus dacarbazine in this trial is higher than the average 10 month survival seen in the ipilimumab phase III trials. Though a comparison of results may not be fair, there is at least a suggestion of benefit for the combination over ipilimumab alone.
Ipilimumab therapy may be associated with unusual and/or severe side effects, prompting FDA to require a Risk Evaluation and Mitigation Strategy (REMS) for the drug. Common side effects that can result from autoimmune reactions associated with ipilimumab use include fatigue, diarrhea, skin rash, endocrine deficiencies (gland or hormone), and inflammation of the intestines (colitis). Severe to fatal autoimmune reactions were seen in 12.9 percent of patients treated with ipilimumab. According to FDA, when severe side effects occurred, ipilimumab was stopped and corticosteroid treatment was started. Patients who did respond in some cases did not see any improvement for several weeks.
Another promising drug for advanced melanoma is vemurafenib (Roche/Genentech). The drug specifically targets melanomas that are positive for the BRAF V600 mutation and is being developed in conjunction with a diagnostic for BRAF status. Results of the phase III BRIM3 study, a global, randomised, open-label, controlled, multicenter trial that compared vemurafenib to dacarbazine chemotherapy in 675 patients with previously untreated BRAF V600 mutation-positive, unresected locally advanced or metastatic melanoma, were recently announced by Roche. The risk of death was reduced by 63 percent for subjects treated with vemurafenib compared to those who received chemotherapy (hazard ratio [HR]=0.37, p<0.0001). The risk of disease progression was reduced by 74 percent in the vemurafenib cohort compared to the chemotherapy cohort. The safety profile of vemurafenib was consistent with previous clinical studies.
The response rate (tumour shrinkage) was 48.4 percent for vemurafenib, versus 5.5 percent for dacarbazine. Six-month survival rate was 84 percent for vemurafenib, compared to 64 percent for chemotherapy.
The most common Grade 3 or higher adverse events were keratoacanthomas, rash, joint pain, sensitivity to the sun and fatigue. Squamous cell carcinoma was reported in 12 percent of patients; lesions were removed, and patients continued with treatment.
Advanced Basal Cell Carcinoma Therapy
Mutations of the patched homologue 1 (PTCH 1) and smoothened homologue (SMO) genes that encode the hedgehog pathway have been associated with BCC. SMO homologue genes are shown to promote growth of BCCs directly. PTCH usually inhibits SMO, so mutation in PTCH encourages BCC.4,5
The oral small-molecule hedgehog inhibitor vismodegib (previously GDC-0449, Genentech) has provided promising results in treatment of advanced BCC. In an earlier trial, 33 patients with metastatic or locally advanced BCC were randomized to receive oral GDC-0449 150mg/day (17 patients), 270mg/day (15 patients), and 540mg/day (one patient) for a median duration of 9.8 months.6 Of the 33 patients, 18 had an objective response to GDC-0449: seven according to assessment on imaging, 11 on physical examination (one patient rated on both). Two patients had a complete response, and 16 had a partial response. Of the remaining patients, only four had progression of disease, and 15 patients remained stable. Fatigue, hyponatremia, muscle spasm, or atrial fibrillation (one case) deemed possibly associated with treatment were reported by six patients. One incident of asymptomatic hyponatremia was judged to be unrelated to GDC-0449.
Results of a larger phase II trial presented at the Seventh European Association of Dermato- Oncology (EADO) Congress offer additional positive evidence for the hedgehog inhibitor. The ERIVANCE BCC trial is an international, singlearm, multi-center, two-cohort, open-label study that enrolled 104 patients with advanced BCC, including locally advanced BCC (laBCC, n=71) and metastatic BCC (mBCC, n=33). laBCC patients had lesions that were inappropriate for surgery (inoperable, or for whom surgery would result in substantial deformity) and for which radiotherapy was unsuccessful or contraindicated. mBCC was defined as BCC that had spread to other parts of the body, including the lymph nodes, lung, bones and/or internal organs. Subjects received 150mg vismodegib orally, once daily, until disease progression or intolerable toxicity.
The primary endpoint of overall response rate was 43 percent in the laBCC cohort, and 30 percent in mBCC, as assessed by independent review. Based on study investigators' assessments, the overall response rate for laBCC was 60 percent and for mBCC was 46 percent (secondary endpoint). The median duration of progression-free survival by independent review was 9.5 months for both metastatic and locally advanced BCC patients.
In 75 percent of patients, vismodegib was shown to shrink tumours, heal visible lesions, or prevent tumor growth. The most common adverse events included muscle spasms, hair loss, altered taste sensation, weight loss, fatigue, nausea, decreased appetite and diarrhea. Serious adverse events (SAEs) were observed in 26 patients (25 percent), however these were considered to be related to vismodegib treatment in only four subjects (four percent). Although fatal events were reported in seven patients (seven percent), none were considered by investigators to be related to treatment with vismodegib. Pre-existing diseases or symptoms were related to the presumed causes of death.
NSAIDs for Chemoprevention
Several recent studies have suggested that use of nonsteroidal anti-inflammatory drugs (NSAIDs) may be associated with at least a modest reduction in an individual's risk for developing melanoma and nonmelanoma skin cancer. Despite some less positive findings, the accumulating evidence indicates a possible influence of NSAIDs on skin cancer risk and suggest a need for further study. There is as yet, no clear clinical application of these findings for skin cancer prevention, but clinicians should be aware of recent findings and anticipate future developments.
In one of the more promising recent studies of NSAID use and skin cancer incidence, Tori, et al. found NSAID use to be associated with a modestly reduced risk of SCC.7 Aspirin use provided the greatest reduction in risk, with the greatest benefit seen among patients taking the drug for six years or less. NSAID use seemed to reduce the risk of SCC tumors with p53 or PTCH mutations, specifically. This same study identified a slightly reduced risk of BCC associated with the use of aspirin or paracetamol (acetaminophen; an analgesic but not a true NSAID), but little or no association between NSAID use and BCC risk overall. The study involved 1,484 participants. Of these, 535 had SCC, 487 had BCC, and 462 were controls.
Another study investigated the potential utility of the cyclooxygenase-2 (COX-2) inhibitor celecoxib as a chemopreventive agent for actinic keratoses (AK) and NMSC.8 COX-2 is known to be expressed in AKs. The double-blind, placebo-controlled, randomized trial involved 240 subjects ages 37-87 diagnosed with 10-40 AKs. Patients were randomly assigned to receive 200mg of celecoxib or placebo orally twice daily for nine months. There was no difference in the incidence of actinic keratoses between the two groups at nine months. However, at 11 months patients in the treatment arm had fewer NMSCs than did controls. The reduction in risk persisted even after adjusting for age, sex, Fitzpatrick skin type, history of AK at randomization, NMSC history, and patient time on study. The rate of both SCC and BCC was reduced among treated patients compared to placebo controls. Although COX-2 inhibitors generally have been linked to an increased risk of cardiovascular events, in this trial the rates of serious and cardiovascular adverse events were similar in both arms.
In another study, use of NSAIDs for shorter durations (less than five years) was associated with an increased time to a first SCC or BCC, while more long-term NSAID use (five years or longer) was found to be less protective against NMSC formation.9 The analysis was performed for subjects enrolled in the SKICAP-AK trial, which was primarily focused on ascertaining the effects of retinol therapy on NMSC incidence. For BCC, specifically, aspirin showed enhanced protection relative to non-aspirin NSAIDs in this study. Otherwise, there were no strong associations between NSAIDs and specific skin cancer types.
Data link longer-term (more than five years) aspirin use with a significant reduction in melanoma risk compared to infrequent NSAID use or non-use.10 The protective benefit was observed after adjustment for confounders in the cohort of 400 melanoma patients and 600 age- and gender-matched, community-based controls. The same study investigated any effect of statin use on melanoma risk and found none.
In 2009, Joosse, et al. reported that continuous use of low-dose aspirin was associated with a reduced incidence of cutaneous melanoma in women but not in men.11 Their conclusions were based on a case-control study of individuals in the Dutch PHARMO pharmacy database and a national pathology database. Cases of primary cutaneous melanoma diagnosed between 1991 and 2004 in individuals 18 or older with three years of pre-diagnosis PHARMO data were analyzed. Age-, sex-, and geographical region-matched controls were selected. A total of 1,318 cases of melanoma and 6,786 controls analyzed. The only association between NSAID use and melanoma risk in these patients was a significant decrease—relative to controls and to aspirin non-users or non-continuous users—in melanoma risk for women who had continuously used low-dose aspirin. Similar trends were not seen in men.
An earlier study involving 63,809 men and women who had been enrolled in the Vitamins and Lifestyle (VITAL) cohort, however, found no association between NSAID use and melanoma incidence after adjusting for melanoma risk factors and indications for NSAID use.12 Analysis looked specifically at nonaspirin NSAIDs, various doses of aspirin, and non- NSAID use and failed to uncover any associations with melanoma. For their analysis, researchers also looked for any effect of NSAID use on tumor invasion, tumor thickness, or risk of metastasis and found none.
The availability of ipilimumab for the treatment of advanced melanoma provides a much needed new option for a disease that has had a dearth of treatment options. Promising findings for vemurafenib suggest that the treatment field may continue to grow. Though many cases of advanced disease are referred for management, dermatologists as the primary diagnosticians must be aware of rapid developments in this realm. The possibility of an oral agent to manage advanced BCC provides new possibilities for long-term clinical management of a disease associated with significant morbidity and mortality.
Findings so far regarding NSAID use and skin cancer chemoprevention are intriguing but not yet clinically applicable for most patients. NSAIDs may be an appropriate option for the prevention of skin cancer in certain high-risk patients with a history of cutaneous malignancies. Given the relatively low risk associated with NSAID use, daily aspirin intake with the goal of tertiary prevention of skin cancer may be reasonable. This is not unlike the long-term use of aspirin in patients at risk for cardiovascular disease. Larger studies, perhaps prospective trials, are needed to further elucidate the relationship between NSAID use and prevention of melanoma or NMSC. The early evidence suggests that treatment may provide relatively modest effects, and that these may be more pronounced in certain sub-populations (melanoma for women; or certain genotypes of NMSC). It also remains to be seen whether targeting COX-2 specifically may be more beneficial than targeting COX-1 and -2, as in the case of aspirin.
The contrasting apparent effects of therapeutic duration on level of protection are also challenging. So far, the evidence seems to suggest that protection against NMSC is greater in the short-term, while protection against melanoma increases with prolonged use.
Dr. Wolfe has no relevant disclosures.
Jonathan Wolfe, MD is Clinical Assistant Professor of Dermatology at the University of Pennsylvania in Philadelphia, PA where he is on the staff of the Pigmented Lesion CLinic. He is in private practice, in Plymouth Meeting, PA.
- Pennock GK, Waterfield W, Wolchok JD. Patient Responses to Ipilimumab, a Novel Immunopotentiator for Metastatic Melanoma: How Different are these From Conventional Treatment Responses? Am J Clin Oncol. 2011 Feb 17. [Epub
- Wolchok JD, Weber JS, Hamid O, et al. Ipilimumab efficacy and safety in patients with advanced melanoma: a retrospective analysis of HLA subtype from four trials. Cancer Immun. 2010 Oct 20;10:9.
- Robert C, Thomas L, Bondarenko I, eta l. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med. 2011 Jun 30;364(26):2517-26.
- Dierks C. GDC-0449--targeting the hedgehog signaling pathway. Recent Results Cancer Res. 2010;184:235-8.
- Stone DM, Hynes M, Armanini M, et al. The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature. 1996 Nov 14;384(6605):129-34.
- Von Hoff DD, LoRusso PM, Rudin CM, et al. Inhibition of the hedgehog pathway in advanced basal-cell carcinoma. N Engl J Med. 2009 Sep 17;361(12):1164-72.
- Torti DC, Christensen BC, Storm CA, et al. Analgesic and nonsteroidal anti-inflammatory use in relation to nonmelanoma skin cancer: A population-based case-control study. J Am Acad Dermatol. 2011 Apr 27, e-pub.
- Elmets CA, Viner JL, Pentland AP, et al. Chemoprevention of nonmelanoma skin cancer with celecoxib: a randomized, double-blind, placebo-controlled trial. J Natl Cancer Inst. 2010 Dec 15;102(24):1835-44.
- Clouser MC, Roe DJ, Foote JA, Harris RB. Effect of non-steroidal anti-inflammatory drugs on non-melanoma skin cancer incidence in the SKICAP-AK trial. Pharmacoepidemiol Drug Saf. 2009 Apr;18(4):276-83.
- Curiel-Lewandrowski C, Nijsten T, Gomez ML, et al. Long-Term Use of Nonsteroidal Antiinflammatory Drugs Decreases the Risk of Cutaneous Melanoma: Results of a United States Case-Control Study. J Invest Dermatol. 2011 Mar 10. Epub
- Joosse A, Koomen ER, Casparie MK, et al. Non-steroidal anti-inflammatory drugs and melanoma risk: large Dutch population-based case-control study. J Invest Dermatol. 2009 Nov;129(11):2620-7.
- Asgari MM, Maruti SS, White E. A large cohort study of nonsteroidal anti-inflammatory drug use and melanoma incidence. J Natl Cancer Inst. 2008 Jul 2;100(13):967-71.