Is there a need to take ethnicity into account when considering lung cancer treatment?

Global burden of lung cancer

Lung cancer is the most common cancer in the world. In 2002, the estimated number of new lung cancer cases was 1.35 million (0.97 million in men and 0.39 million in women), representing 12.4% of all new cancers in that year. Lung cancer is also the most common cause of death from cancer, with an estimated 1.18 million deaths (0.85 million deaths among men and 0.33 million deaths among women) in 2002, accounting for 17.6% of all cancer deaths that year. Patients with distant metastases at the time of their initial diagnosis account for approximately 40% of all patients with lung cancer. Thus, one of the most important approaches to reducing the lung cancer burden is to develop more effective anticancer agents [1].

Little attention has been paid to the role of ethnicity in the treatment of lung cancer, except for the various incidences and risks of lung cancer among ethnic groups [2]. Now is, however, an appropriate time to consider how ethnic differences may affect lung cancer treatment, since global drug-development efforts are becoming more common and our scientific knowledge of how different ethnic groups respond to drugs is developing steadily.

Global drug development

The problematic issues that underlie recent drug development include the prolonged timeline, the high costs involved in terms of financial, patient and professional resources, and the increasing complexity of clinical trials involving molecular-targeted agents [3]. Historically, regional regulatory authorities often requested extensive duplication of clinical evaluations for each compound, owing to concerns that ethnic differences might affect the medication’s safety and efficacy, resulting in an unnecessary waste of drug-development resources. One effort to improve the efficiency of this process has been the start of internationally harmonized drug development. The International Conference on the Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) was launched in 1990 to facilitate the registration of new agents. The ICH-E5 (R1) guideline, Ethnic Factors in the Acceptability of Foreign Clinical Data, published in 1998, addresses the issue of ethnic factors in the acceptability of foreign clinical data [101]. In this guideline, ethnic factors are defined as intrinsic factors relating to genetic and physiologic characteristics and extrinsic factors are those relating to the cultural and environmental characteristics of a population.

A bridging study is defined as a study that provides pharmacodynamic or clinical data on efficacy, safety, dosage and dose regimens in a new region and that will allow the extrapolation of foreign clinical data to the new population. The sensitivity of an agent to ethnic factors helps to determine the need for, and nature of, the bridging study. Anticancer agents are considered to be sensitive to ethnic factors because, generally, cytotoxic agents have a steep pharmacodynamic (effect–concentration) curve for both efficacy and safety within the range of the recommended dose. Consequently, even a small change in the dose can result in a large change in the effect. Molecular-targeted agents, however, have a relatively flat pharmacodynamic curve for both efficacy and safety, allowing these agents to have wider therapeutic dose ranges. This is one reason for the recent increase in the adoption of global clinical trials for new anticancer agents that involve trial institutions on more than one continent. This strategy enables adequate sample sizes to be obtained in a relatively short time period and eliminates the need for redundant clinical trials with similar objectives to be conducted in different countries. Definitions of the pharmacokinetics, pharmacodynamics and dose response of the drug made during the early clinical phases can also help to determine the need for bridging studies and global clinical trials.

Polymorphisms of drug-metabolizing enzymes & transporters

A comparison of Phase III trials for the same drug combinations against lung cancer in different countries shows a great diversity in hematological toxicity [4]. For example, a combination of carboplatin and paclitaxel, which is widely used in the treatment of non-small-cell lung cancer (NSCLC) worldwide with similar doses and schedules, produced grades 3–4 neutropenia and febrile neutropenia in 88 and 16%, respectively, of patients in a Japanese trial but in only 6–65 and 2–4%, respectively, of patients in American trials. A combination of cisplatin and irinotecan and that of cisplatin and etoposide for extensive small-cell lung cancer were also more toxic among Japanese patients than among American patients: grades 3–4 neutropenia was noted in 65 and 32%, respectively, of patients who received cisplatin and irinotecan, and in 92 and 65%, respectively, of patients who received cisplatin and etoposide [5].

These ethnic differences in toxicity can be explained by varying activities of drug-metabolizing enzymes and transporters. The best-studied example is that of the effects of polymorphisms in the UDP-glucuronosyltransferase (UGT)1A1 gene, which is involved in irinotecan metabolism. Grade 4 irinotecan-induced neutropenia was noted in 40–57% of the patients who were homozygous for UGT1A1*28, a polymorphic variant in the promoter region of the UGT1A1 gene; whereas, grade 4 neutropenia was only observed in 15% or less of the patients with wild-type alleles [6]. The UGT1A1*6 allele is another polymorphism at exon 1 that is associated with a defective glucuronidating function. Among the patients who received irinotecan monotherapy, grade 3–4 neutropenia developed in 50% of the patients with UGT1A1*6 homozygosity but in only 3% of the patients who had no UGT1A1*6 alleles [7]. In 177 Japanese patients treated with irinotecan-including chemotherapy, a homozygous or double heterozygous genotype for UGT1A1*6 and UGT1A1*28 (*6/*6, *28/*28 or *6/*28) was significantly associated with grade 3–4 neutropenia [8]. The frequencies of these UGT1A1 variant alleles differ among ethnic groups: the frequencies of the UGT1A1*28 allele in African–American, Caucasian and East-Asian populations are 0.45, 0.39 and 0.097–0.14, respectively; while the frequencies of the UGT1A1*6 allele in these three ethnic groups are zero, 0.007 and 0.15–0.21, respectively [7–10]. Thus, the UGT1A1*28 allele is more common in African–American and Caucasian individuals, while the UGT1A1*6 allele is exclusively found in the East-Asian population.

Associations between polymorphisms in transporter genes and drug clearance or toxicity have also been reported. These associations include a decreased docetaxel clearance in patients with a homozygous C1236T polymorphism in the ABCB1 gene [11], an increased risk of grade 3 irinotecan-induced diarrhea in patients with a homozygous C3435T polymorphism in the ABCB1 gene [5] and an increased gemcitabine-associated hematological toxicity in patients with a SLC28A1 plus 1561 G>A polymorphism [12]. These results, however, are all preliminary, and large-scale confirmatory studies are needed.

Efficacy of EGF receptor tyrosine kinase inhibitors among ethnic groups

EGF receptor (EGFR), a cell membrane receptor with tyrosine kinase activity, is expressed in most patients with NSCLC and plays a role in cellular proliferation, inhibition of apoptosis, angiogenesis, metastatic potential and chemoresistance. Gefitinib and erlotinib, selective EGFR tyrosine kinase inhibitors, can cause dramatic tumor shrinkage in some patients with NSCLC. Two Phase II studies of gefitinib in patients with advanced NSCLC who had been previously treated with platinum-based chemotherapy showed an increased response rate in never-smokers, women and those with an adenocarcinoma histology, compared with patients who did not have these characteristics. In addition, the response rate was 28% in Japanese patients but only 9–12% in patients of other ethnicities [13,14]. A randomized Phase III trial of gefitinib versus a placebo in previously treated NSCLC patients failed to show any survival benefit of gefitinib in the overall population but showed a favorable effect with regards to survival in an Asian population [15]. A similar association between objective responses and ethnicity was observed in Phase II studies of erlotinib monotherapy for previously treated advanced NSCLC; the response rate was 28% for Japanese patients and 12% for American patients [16,17]. A Phase III trial of erlotinib or a placebo in the same patient population confirmed that the response rate in Asian patients was higher than that observed in patients of other ethnicities (28 vs 10%; p = 0.02) [18]. These results of Phase II and III trials consistently demonstrated that EGFR tyrosine kinase inhibitors were more effective in Asian patients than in patients of other ethnicities.

In 2004, lung cancers with activating mutations of the EGFR gene were reported to be highly responsive to gefitinib [19,20]. The percentage of patients with EGFR mutations in different subgroups of patients almost paralleled the percentage of responders to gefitinib or erlotinib in the corresponding subgroups; the EGFR mutations were detected in approximately 30% of NSCLC tissues from Asian patients but in only 7–8% of NSCLC tissues from other ethnic populations. The EGFR mutations are predominantly present in exons 18–21, the first four exons encoding the tyrosine kinase domain. According to data from compiled publications, approximately 48% of the mutations were deletions, with over 20 variant types in exon 19, 43% were the point mutation L858R in exon 21, 4% were in-frame insertions in exon 20 and 3% were the point mutation G719X in exon 18 [21]. The distribution of the mutation types among different ethnic groups has not been evaluated in a sufficient number of patients. Responses to EGFR tyrosine kinase inhibitors differed among the mutational types: the response rate was highest in patients who had tumors with exon 19 deletions (81%), followed by those with tumors with L858R (71%) and G719X (56%) mutations. None of the tumors with an in-frame insertion in exon 20 responded to EGFR tyrosine kinase inhibitors [21].

The mechanisms responsible for the high frequency of EGFR mutations in Asian patient populations are puzzling and have not been elucidated. First, the majority of somatic mutations in lung cancer tissues, such as typical mutations in the p53 tumor-suppressor gene, are strongly associated with exposure to smoking-related carcinogens [22]. EGFR mutations, however, are more common in nonsmokers than in smokers. A case–control study showed that smoking history was not associated with the incidence of EGFR-mutated NSCLC, in sharp contrast to the strong association between smoking history and the incidence of EGFR wild-type NSCLC [23]. Smoking status also influences the spectrum of p53 mutation types [22], but has little influence on the type of EGFR mutation, especially among women [24]. Second, genetic instability induced by the loss of DNA-repair function can increase somatic mutations in many unrelated genes, but no mechanisms that promote mutations in a specific gene have been proposed. No reports have suggested that the frequency of any other mutations in EGFR-mutated tumor tissues may be higher than that in tumor tissues with wild-type EGFR. Furthermore, EGFR mutations have been identified exclusively, without any simultaneous mutations in established oncogenes such as K-RAS and ERBB2[21]. Thus, the occurrence of mutations in the EGFR gene cannot be attributed to the genetic instability that is widely observed in tumor tissues. Third, the repeat length of the CA simple sequence repeat 1, located at intron 1 of the EGFR gene, varies according to ethnicity, with Asian individuals tending to have longer repeats than Americans [25]. Polymorphisms in the regulatory sequence of a gene may change the expression of the gene but are unlikely to directly induce a mutation in the gene.

The currently available data on risk factors for the development of EGFR-mutated lung cancer are limited. A case–control study of the effect of dietary factors on the risk of NSCLC showed that soybean products had a protective association against EGFR-mutated, but not EGFR wild-type, NSCLC; whereas, green–yellow vegetables reduced the risk of EGFR wild-type, but not EGFR-mutated, NSCLC [26]. Another case–control study showed that a longer reproductive lifespan, associated with a higher endogenous estrogen exposure, was a risk factor for EGFR-mutated NSCLC [23]. However, polymorphisms in genes that encode enzymes involved in estrogen biosynthesis and metabolism are unlikely to be major genetic modifiers of the prevalence of EGFR-mutated NSCLC [27].

Conclusion

Ethnicity is an important factor to be considered in the introduction of new drugs. The best example of this principle can be found in the clinical development of gefitinib. This drug was evaluated in two large randomized Phase III trials conducted in more than 2000 patients from the USA and Europe [28,29]. Both trials, however, failed to show a survival benefit with respect to the addition of gefitinib to conventional chemotherapy. Another large-scale Phase III trial of gefitinib versus a placebo in 1692 chemotherapy-refractory NSCLC patients also showed no survival benefit of gefitinib [15]. If these trials were to be reviewed today, their negative results would seem reasonable because the frequency of EGFR mutations in the tumor tissues of those patient populations is, at most, only 10%. On the other hand, the Iressa Pan-Asia Study was a Phase III study of gefitinib versus a combination of carboplatin and paclitaxel administered as a first-line treatment for advanced NSCLC in a clinically selected patient population whose tumors had an adenocarcinoma histology and who were either lifetime never-smokers or long-term ex-light smokers (i.e., had ceased smoking for at least 15 years and up to 10 pack-years exposure). The study enrolled 1217 patients from nine countries in Asia between March 2006 and October 2007, and demonstrated a superior progression-free survival (the trial’s primary end point) for gefitinib versus carboplatin and paclitaxel chemotherapy, with a hazard ratio of 0.74 (95% confidence interval: 0.65–0.85; p < 0.0001) [30]. This study clearly indicated that gefitinib is indispensable to the treatment of advanced NSCLC in Asia.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.