Capsaicin, the pungent alkaloid in jalapeños and other chile peppers that makes them hot, drives prostate cancer cells to kill themselves off, according to studies published in the March 15 issue of Cancer Research. See:
http://www.psa-rising.com/eatingwell/peppers_hot.htm

Some commentators reacted by warning that hot peppers may cause gastric cancer. Is this true? How strong is the association?

A study conducted at a Texas Veterans Administration hospital in 1988, published in the JAMA, injected about an ounce of jalapeno pepper directly into the stomachs of volunteers. Follow-up observation showed no damage to their stomach linings. But this did not amount to chronic exposure, and anti-cancer versus cancer-causing effects of capsaicin are still controversial. A couple of years ago Mexico National Institute of Public Health found higher rates of gastric cancer in people who ate the equivalent of 9-25 jalapeno peppers a day compared to people who no more than 3 a day. They found no evidence that bacterioa known to be associated with some types of stomach were causing this increased risk.

A Berkeley neurobiology lab has addressed the question:
Could capsaicin have chemoprotective properties?:

Excessive ingestion of hot peppers has been considered to be a risk factor for gastric and hepatic cancers. Chili and more specifically capsaicin consumption has been shown, however, to be protective against stomach cancer (Surh et al. 1998). In Mexico, where people heavily consume peppers, the frequency of gastric cancer is relatively low (Surh et al. 1998). In the United States, moreover, the rate of stomach cancer incidence has been declining for the past twenty years despite increased chili pepper consumption (Surh et al. 1998). Human gastric cancer may be caused by a number of factors: salty, smoked, or pickled foods, cigarette smoking, insufficient fruit and vegetable intake, and Helicobacter pylori. Capsaicin seems to have some inhibitory effects on carcinogenesis induced by chemical carcinogens such as vinyl carbamate (Surh et al. 1998). Treatment with capsaicin before exposure to gaseous chemicals such as sulfur dioxide and nitrogen dioxide has been shown to protect against free radical-induced pulmonary damage in rats (Surh et al. 1998). Capsaicin is also observed to inhibit inflammatory responses related to tumor promotion (Surh et al. 1998).

All those references are to a 1998 article by Young-Joon Surh, Ph.D. In 2002, Dr. Surh wrote an editorial for the Journal of the National Cancer Institute about the latest research on capsaicum: More Than Spice: Capsaicin in Hot Chili Peppers Makes Tumor Cells Commit Suicide. He stated that “The role of capsaicin in carcinogenic processes is quite controversial. Although some investigators suspect that capsaicin is a carcinogen, co-carcinogen, or tumor promoter, others have reported that it has chemopreventive and chemotherapeutic effects.” After a detailed look at the current state of knowledge of how it affected cells, he concluded:

In summary, the oxidative stress that is stimulated by vanilloid treatment of SCC cells is primarily of mitochondrial origin and contributes to the death of these cells by apoptosis. The exact molecular milieu that characterizes elevated oxidative stress caused by vanilloid treatment is not clear and requires further investigation. It is possible that a high metabolic rate or increased oxygen utilization as a direct or indirect consequence of aberrant electron flow in the mitochondrial respiratory system of malignant cells results in the increased production of oxidants, which may overwhelm cellular antioxidant protections and lead to apoptosis. However, it can be argued that such inappropriate ROS generation may also have a deleterious effect on nonmalignant cells as well, and if this is the case, the vanilloids would be metabolic poisons rather than valuable candidates for use in preventive therapy for skin cancer or other cutaneous disorders.

The small amount of research done on the specific problem of whether capsaicin causes gastric cancer has turned up opposite results in Mexico (e.g. Int J Cancer. 2003 Aug 20) and Korea (World J Gastroenterol. 2005 Oct 28; ). One difference is between study of cancer in populations (Mexico) versus study of effects of the chemical on tumor cells (Korea).

Findings on Capsaicin pepper, cancer and US ethnic cuisine (Med Hypotheses. 2002 Oct) do give grounds for caution:

The ‘hot’ sensation produced by exposure to pepper is apparently due to two natural carcinogens: capsaicin in chili type peppers and safrole in black/white pepper. There are four cookeries in the United States that are noted for their high pepper content: Mexican-American, Cajun, white Creole, and black Creole. Each is largely confined to a single ethnic-cultural group which is concentrated in some counties. By use of county population and mortality data, significantly higher rates for stomach and liver cancer were found in counties inhabited by these four ethnic-cultural groups than in matched control counties. This involved both sexes. The cancer increase was dependent on the concentration of these groups in a county. These results strengthen and extend an earlier case-control study which found odds ratios above 5 for the stomach cancer association with capsaicin pepper. It is further evidence that capsaicin is a human carcinogen.