Heterocyclic amines (HAAs) are suspected carcinogens that are formed in meat when it is cooked at high temperature for long durations. These compounds require metabolic activation by CYP1A2 and N-acetyltransferase (NAT) 2 or NAT1 before they can bind to DNA. It has been hypothesized that well-done meat increases the risk of colorectal cancer (CRC), especially in individuals with the rapid phenotype for CYP1A2 and NAT2. This association may be particularly strong in smokers because smoking is known to induce CYP1A2. We conducted a population-based case-control study on Oahu, Hawaii to specifically test this hypothesis. An in-person interview assessed the diet and preference for well-done red meat of 349 patients with CRC and 467 population controls. A urine collection after caffeine challenge and a blood collection were used to assess phenotype for CYP1A2 and NAT2 and genotype for NAT2 and NAT1, respectively. No statistically significant main effect association with CRC was found for red meat intake, preference for well-done red meat, the NAT2 rapid genotype, the CYP1A2 rapid phenotype or the NAT1*10 allele. However, in ever-smokers, preference for well-done red meat was associated with an 8.8-fold increased risk of CRC (95% confidence interval, 1.7-44.9) among subjects with the NAT2 and CYP1A2 rapid phenotypes, compared with smokers with low NAT2 and CYP1A2 activities who preferred their red meat rare or medium. No similar association was found in never-smokers, and there was no increased risk for well-done meat among smokers with a rapid phenotype for only one of these enzymes or for smokers with both rapid phenotypes who did not prefer their red meat well-done. These data provide additional support to the hypothesis that exposure to carcinogens (presumably HAAs) through consumption of well-done meat increases the risk of CRC, particularly in individuals who are genetically susceptible (as determined by a rapid phenotype for both NAT2 and CYP1A2) and suggest that smoking, by inducing CYP1A2, facilitates this effect.