Commentary: Current Status of Retinoid Chemoprevention of Lung Cancer

The lack of effective treatments for advanced cancer has been one of the motivating forces in the field of cancer prevention. Nowhere is this approach more justified than in primary pulmonary neoplasms. In this article, Benner and his M.D. Anderson colleagues review trials investigating the prevention of lung cancer utilizing retinoids and carotenoids.

As the authors discuss, trials suggest that retinoids, especially 13-cis-retinoic acid (Accutane), may decrease the incidence of second primaries in patients with head and neck cancers. They also touch upon trials of naturally occurring retinoids, such as retinyl palmitate, as adjuvant therapy for patients with resected lung cancer. They briefly discuss the role of intermediate end points, and the many issues surrounding the interpretation as well as conduct of trials using these end points.

While this article provides a brief overview of the field of lung cancer chemoprevention, it does not emphasize many of the complex issues relevant to lung cancer chemoprevention and intermediate end point trials. Also, there is no discussion of new retinoids now entering clinical trials or the potential
for discovering new retinoids and
carotenoids.

Complex Issues

In general, lung cancer prevention trials have centered on three distinct populations:

1. The general populations at risk because of tobacco abuse or asbestos exposure

2. High-risk smokers with documented bronchial dysplasia or metaplasia

3. Patients with a previous diagnosis of head and neck or lung cancer.

The first group has been the focus of the large, lengthy primary prevention trials. The second population has been a prime target for intermediate end point trials, and the third group has been earmarked for both intermediate end point trials and trials studying second primary cancers.

When designing a chemoprevention trial, the choice of a study population is governed by many issues. Trials involving healthy populations can be justified only if the agents being studied are extremely safe. In these populations, the only retinoids and carotenoids studied thus far have been retinol, retinyl palmitate, and beta-carotene. All of these agents have been available as over-the-counter supplements for many years, and are presumed safe.

High-risk smokers with bronchial dysplasia or metaplasia and patients with head and neck or lung cancer have been the target populations for more toxic retinoids, such as 13-cis-retinoic acid and newer agents. Although 13-cis-retinoic acid has shown efficacy in preventing second primaries in head and neck cancer patients, it is unlikely to be useful as a primary prevention agent because of its long-term toxicity. The relationship between an agent's toxicity and its potential use as a chemoprevention agent in populations of varying risks needs to be stressed in any discussion of chemoprevention.

Because of the lack of agents known to be safe with chronic dosing, and the costs and statistical requirements of conducting primary prevention trials, attention has shifted to the study of intermediate end points. It is hoped that these biologic changes will be reliable predictors of cancer incidence. Benner et al have pointed out some difficulties in evaluating these trials, citing the sputum metaplasia/dysplasia trials.

The first publication using this end point was a single-arm French trial, which suggested efficacy for the synthetic retinoid etretinate (Tegison). However, a later randomized controlled trial conducted by the M.D. Anderson group showed no apparent effect.

Results of these trials underscore the necessity of the randomized placebo-control design when the natural history of the end points under study has not been defined. In addition, none of the currently available intermediate end points has been shown to predict lung cancer incidence, or to be a reliable surrogate end point for determining the efficacy of a prevention agent. Numerous ongoing and planned trials are now testing newer markers, such as oncogene and tumor-suppressor gene expression and mutation, proliferation indices, and changes in cell surface proteins.

New Retinoids and Carotenoids

The development of newer retinoids and carotenoids is another area worthy of discussion. Currently, 4-N-hydroxyphenyl retinamide (4-HPR) is undergoing investigation in Europe as a breast cancer chemoprevention agent [1]. This retinoid may have applicability in patients with lung cancer. One motivation for the development of this agent (and other synthetic retinoids) has been its apparent clinical safety and high efficacy in in vitro trials. In the aforementioned European breast cancer trial, minimal toxicity was reported at 42 months [2], and little additional toxicity has been seen with follow-up of over 5 years. Hence, 4-HPR may have potential value as a primary prevention agent as well as an agent for patients at high risk for second primaries.

Trans-retinoic acid and potentially 9-cis-retinoic acid (a biologically active metabolite of trans-retinoic acid) are undergoing evaluation, and may also have clinical applicability in selected populations at high risk for lung cancer. As the authors discuss, retinoids are also being tested in combination with other agents, for example, the combination of retinyl palmitate and N-acetylcysteine is being studied in the ongoing Euroscan trial.

Although carotenoid clinical trials are frequently considered in parallel with retinoid trials, these agents are not interchangeable. One biologic activity of beta-carotene derives from its transformation to retinol. However, this agent can function as an electron-scavenging antioxidant, and has actions independent of its retinoid activity [3]. Other dietary carotenoids are also being described, and epidemiologic studies are exploring the relationship between their serum concentration and lung cancer incidence. The study of carotenoids will likely introduce other potential prophylactic agents into the realm of clinical trials.

A Cautionary Note

Benner and colleagues point out that the Finland smokers trial, which studied beta-carotene (20 mg) and alpha-tocopherol (50 mg), has added an appropriate cautionary note to the field. This trial found an 18% increase in the incidence of lung cancer in the groups taking beta-carotene. Although many agents studied as cancer prevention agents are nutritional components, it must be remembered that when given in supraphysiologic doses, they must be considered as drugs with unknown long-term side effects. It is unclear whether these agents, taken on a daily basis for long periods, are safe or a potential health hazard.

Many agents being tested, even those that have been presumed safe, such as beta-carotene, have never been studied in long-term clinical trials. Carefully conducted trials will be needed to determine whether any of the agents that appear so promising in epidemiologic and in vitro trials is effective in decreasing cancer incidence. Determining long-term toxicity will be as crucial as determining efficacy. Like the study of cytotoxic chemotherapy for advanced cancer, this field will have successes and failures along the way.

References:

1. Costa A, Formelli F, Chiesa F, et al: Prospects of chemoprevention of human cancers with the synthetic retinoid fenretinide. Cancer Res (suppl) 54:2032s-2037s, 1994.

2. Rotmensz N, De Palo G, Formelli F, et al: Long-term tolerability of fenretinide (4-HPR) in breast cancer patients. Eur J Cancer 27:9, 1127-1131, 1991.

3. Gerster H: Anticarcinogenic effect of common carotenoids. Int J Vitam Nutr Res 63:93-121, 1993.