Insights Gained in the Role of EGFR in Cancer

In parallel to the clinical trials mentioned above and as the number of patients treated with these agents increased, a number of groups started to rationally seek factors that may be linked to the activity of the compounds. The first evidence came from clinical observations. It was known that female patients, patients of Asian origin, never -smokers and those with an adenocarcinoma type of NSCLC were the subgroups more likely to benefit from these agents. Subsequent molecular studies did not reveal the cause of this observation. This includes the link between receptor amplification and response to these agents, as well as the discovery of activating mutations of the egfr gene.

Abnormalities in egfr copy number are frequent in cancer. In a report that investigated egfr and EGFR expression (by fluorescent in situ hybridization [FISH] and immunohistochemistry [IHC], respectively) in 183 NSCLC patients, trisomy, polysomy and gene amplification were observed in 40 percent, 13 percent and 9 percent of the cases, respectively (50). EGFR over-expression was observed in 62 percent of the cases and correlated with increased gene copy number. Increased EGFR gene copy number detected by FISH is associated with improved survival after gefitinib therapy in patients with NSCLC (51). In this report, amplification or high polysomy of the egfr (documented in 33 of 102 patients) and high protein expression (observed in 58 of 98 patients) were significantly associated with better response (36% versus 3%, mean difference = 34%, 95% CI = 16.6 to 50.3; P < 0.001), disease control rate (67% versus 26%, mean difference = 40.6%, 95% CI = 21.5 to 59.7; P < 0.001), time to progression (9.0 versus 2.5 months, mean difference = 6.5 months, 95% CI = 2.8 to 10.3; P < 0.001), and survival (18.7 versus 7.0 months, mean difference = 11.7 months, 95% CI = 2.1 to 21.4; P = 0.03). Similar results regarding the correlation between egfr copy number and outcome were observed in a cohort of subjects with advanced bronchioalveolar carcinoma (BAC) (52). These two reports suggest FISH can be used to assess survival potential in patients treated with EGFR TKIs. In the latter subset of lung cancer subjects, no association was found between HER2 gene copy number and response or survival. Interestingly, in another report, increased HER2 copy number was also a solid marker of response to gefitinib therapy in a broader lung cancer population (53). Patients with HER2 FISH-positive tumors displayed increased expression of EGFR protein and gene gain. These findings highlight the relevance of the interplay between the HER family of receptors in the pathogenesis of cancer. In the univariate analysis of the NSCLC patients receiving erlotinib or placebo in the pivotal trial, survival was longer in the erlotinib group than in the placebo group when there was high EGFR expression (hazard ratio, 0.68; P = 0.02), or there was a high number of copies of egfr (hazard ratio, 0.44; P = 0.008) (54), but these correlations were not evident in the multivariate analysis.

Recent data have shown that mutations in the ATP-binding site of the egfr gene predict sensitivity of NSCLC patients to gefitinib (55, 56). In the report by Lynch, et al mutations were identified in the tyrosine kinase domain of the egfr gene in eight out of nine patients with gefitinib-responsive lung cancer, as compared with none of the seven patients with no response (P<0.001) (55). In the report by Paez, et al somatic mutations of the egfr gene were found in 15 of 58 unselected tumors from Japan and one of 61 from the United States (56). This phenomena was not agent- or family-specific, as it has been also documented in NSCLC patients treated with erlotinib (57), and in cell lines treated with the bifunctional (EGFR plus VEGFR2/KDR) inhibitor ZD6474 (58).

Mutations were either in-frame deletions or amino acid substitutions clustered around the ATP-binding pocket of the tyrosine kinase domain of EGFR. Remarkably, many of these deletions overlapped, sharing the deletion of four amino acids within exon 19. Other tumors had amino acid substitutions within exon 21, being particularly frequent and consistent in several reports the change from leucine to arginine at codon 858 (L858R). All mutations were heterozygous, and identical mutations were observed in multiple patients, suggesting an additive-specific gain of function. Matched normal tissue from available patients showed only the wild-type sequence, indicating that the mutations had arisen somatically during tumor formation. To further support the pathogenic role of mutations in determining the response of NSCLC to EGFR TKIs there are already reported cases where secondary mutations reverse an initial sensitivity to those agents (59).

The location of the mutations influences the sensitivity to EGFR inhibition. Gefitinib was more effective in patients with the deletion type of mutations than in patients with other mutations such as L858R (60). The response rate of patients with an exon 19 deletion and L858R were 84 percent and 71 percent, respectively, but only about half of the subjects bearing G719X had an objective response to gefitinib. In addition, patients with exon 19 deletions had a longer median survival after erlotinib or gefitinib than those with L858R (34 vs 8 months, respectively; P = 0.01) (61). In an analysis of erlotinib sensitivity using mutant constructs the order of sensitivity was exon 19 deletion = L858R > G719X > exon 20 insertion = wildtype, which is similar to the clinical observations so far (62).

In those initial, retrospective and non-consecutive analyses mutations were more prevalent in female patients with adenocarcinoma histology, and in Asian ethnic backgrounds. The report by Kosaka, et al confirmed in a systematic manner what had been described in the anecdotal initial series of NSCLC patients (63). Egfr mutations were not related to age or clinical stage, but there was a strong positive correlation between female gender, non-smoking status, adenocarcinoma subtype, and high degree of differentiation to mutation presence. Across all reports, independently of ethnic origin, egfr mutations appear almost exclusively in adenocarcinomas. It is relevant to note that as opposed to Western patterns, adenocarcinoma accounts for the majority of the NSCLC cases in Japan - as much as 70 percent in a series of resected cases (64). The actual difference in incidence of mutations between Japanese and American populations may, in part, arise from different ethiopathogenic factors mostly evidenced by profoundly dissimilar tobacco consumption, especially in women. Spontaneous mutation occurrence in predisposed histologic glandular cell subtypes, as opposed to carcinogen-induced in epithelial cells, may be behind these differential patterns.

In one of the first reports to gain further insight on the mechanistic basis of this observation, cell lines were transfected with such mutations, and mutant strains showed equivalent sensitivity to gefitinib concentrations 10-fold lower than parental cell lines (55). Differences in EGFR phosphorylation were noted and, as in trans-fection-induced mutated cell lines EGFR Tyr1068 phosphorylation, was more intense and also had a longer duration. These results may indicate that the mutations lower the threshold of efficacy for TKIs and thus render the EGFR susceptible to lower (clinically achievable) drug concentrations, which are suboptimal to efficaciously inhibit the receptor in the patients bearing the wild-type phenotype. As mentioned bellow, this may have explained the results of some of the clinical trials. Several reports indicate that the occurrence of EGFR mutations is an early event in carcinogenesis. Particularly, a study that analyzed mutation-positive and -negative cancers and normal adjacent mucosa showed that egfr mutations identical to the tumors were detected in the normal respiratory epithelium in 9 of 21 (43 percent) patients with mutant adenocarcinomas, but in none of the 16 patients without tumor mutations (65). The finding of mutations being more frequent in normal epithelium within tumor (43 percent) than in adjacent sites (24 percent) suggests a localized field effect phenomenon. In a small report in Japanese patients, egfr mutations were found in 12 of 19 (63 percent) of brain metastases of patients with NSCLC (66).

The same types of mutations were found in those where both primary and metastatic tissue were available, suggesting that mutation occurrence precedes systemic spread and supporting an early appearance.

A seminal report generated transgenic mice with inducible expression in type II pneumocytes of two common egfr mutants seen in human lung cancer (67). Both transgenic lines developed lung adenocarcinoma after sustained egfr mutant expression, confirming their oncogenic potential. Importantly, maintenance of these lung tumors was dependent on continued expression of the EGFR mutants and treatment with small molecule inhibitors (erlotinib or HKI-272), as well as prolonged treatment with a humanized anti-hEGFR antibody (cetuximab) which led to dramatic tumor regression. However, the pathogenic role of these mutations and its impact in downstream pathways is not completely understood. A report by Sordella, et al has shed some light in this issue, as it analyzed the differences in EGFR phosphorylation patterns in the five possible sites of the intracellular domain of the EGFR comparing mutated and wild-type NSCLC cell lines (68). Y1045 and Y1173 showed no differences, Y992 and Y1068 were more activated in mutated vs. wild-type, and Y845 was more activated in missense mutations vs. wild-type or deletion mutations. ERK status was equal in mutated vs. wild-type cell lines, probably because this signal is usually transduced via Y1173 to ras and then ERK. Phosphorylation of both Akt and STAT5 was higher in mutated vs. wild-type, as they are linked to Y992 and Y1068. These results suggest that 1) the downstream will ultimately depend on the mutation type, and 2) Akt status has questionable predictive value per se, as it fluctuates depending on the type of mutation phenotype present. It would be more informative to determine the actual subtype of EGFR phosphorylation, instead, to put into perspective the downstream scenario. In concordance to the prior data, Conde, et al determined in an analysis of the genetic and histological features of NSCLC patients that the mammalian target of rapamycin (mTOR) pathway was significantly more activated in both egfr and Kras mutants than in their wild-type counterparts (69). EGFR mutations tended to be associated with increased numbers of CA repeats and increased egfr gene copy numbers, but not with EGFR and caveolin-1 mRNA over-expression (70). In summary, it is increasingly evident that egfr mutations are oncogenic, appear early in tumorigenesis, are associated with specific signalling signatures, and induce a phenomenon of oncogene addiction that render the strains bearing them particularly sensitive to EGFR targeted therapies.

The initial reports were retrospective and, therefore, could not address the prevalence of egfr mutations in the general population of cancer patients. Pao and Miller reviewed this and the results are summarized in Table 10-3 (71). A consecutive series of 277 Japanese patients with NSCLC has shown a prevalence of mutations of 40 percent that were associated with female and non-smoker status, and adeno-carcinoma subtype (63). A relevant aspect of this report is that egfr mutations were never found along with Kras mutations, and were more prevalent in non-smokers. In the trial that compared erlotinib with a placebo 177 samples were analyzed for egfr mutations, and a mutation incidence of 22 percent was documented (54). Finally, in a recently reported analysis of 860 consecutive NSCLC Italian patients

Table 10-3 Incidence of egfr mutations in various subgroups of NSCLC (71)

No. of tumors No. of tumors with Positive for EGFR Characteristic evaluated EGFR mutation mutation (%)

Table 10-3 Incidence of egfr mutations in various subgroups of NSCLC (71)

No. of tumors No. of tumors with Positive for EGFR Characteristic evaluated EGFR mutation mutation (%)

Never-smokers

181

92

50.8

Smokers

434

39

9.0

Women

216

81

37.5

Men

422

55

13.0

Adenocarcinoma

453

142

31.3

Non-adenocarcinoma

306

7

2.3

East Asian

419

122

29.1

Non-East Asian

340

27

7.9

United States

262

25

9.5

Total

759

149

19.6

a global egfr mutation incidence of 4.5 percent was found (72). No mutations in 454 squamous carcinomas and 31 large cell carcinomas investigated were documented, and 39 were found in the series of 375 adenocarcinomas. Again egfr mutations and Kras mutations were mutually exclusive. Bearing in mind that Kras serves as a downstream mediator for EGFR, the authors of the Italian report speculate that the mutually exclusive presence of egfr and Kras mutations may respond to an evolutionary paradigm where activating mutations in egfr are redundant if a mutation in Kras is already present (and vice versa). This may also help explain the striking inverse relationship of tobacco consumption and incidence of egfr mutations observed by this and other groups (73); it can be speculated that smoking tends to induce mutations in Kras that somehow prevent or make unnecessary other function-acquiring genetic changes. In addition, this downstream event seems to render EGFR-targeted therapy inefficacious, adding predictive value to its evaluation.

Few reports have addressed the independent prognostic value of egfr mutations. Egfr mutations were detected in 13 percent of 274 tumors of previously untreated patients with advanced NSCLC in the Phase III study that randomly assigned to carboplatin and paclitaxel with erlotinib or placebo. Mutation presence was associated with longer survival, irrespective of treatment (P < .001) (74). Whether this is directly related to the egfr mutation per se, or a consequence of the absence of Kras mutation, is unknown. Among erlotinib-treated patients, egfr mutations were associated with improved response rate (P < .05) and there was a trend toward an erlotinib benefit on time to progression (P = .092), but not improved survival (P = .96). In contrast, the Japanese report on 277 patients and the follow-up analysis of the gefitinib-treated subjects showed that whereas in patients that had not received the drug the mutational status had no significant prognostic value, the analysis of the patients that had received gefitinib revealed that the presence of the mutation had predictive value for increased survival (60, 63).

In an analysis of 90 NSCLC patients treated with gefitinib the response rate in the 17 patients harboring an egfr mutation was 65 percent in contrast to 13.7 percent in patients without mutation (P < .001) (75). Moreover, these 17 patients with EGFR mutation had significantly prolonged time to progression (21.7 v 1.8 months;

P < .001) and overall survival (30.5 v 6.6 months; P < .001) compared with the remaining 73 patients without mutation. In a recent report in 69 Korean NSCLC patients treated with gefitinib that analyzed the predictive value of several genetic and histologic parameters, there were no responders among carriers of Kras mutations that included two cases with concomitant egfr mutations (76); egfr mutation presence was the only factor with predictive value in multivariate analysis. Other reports confirmed the predictive value of egfr mutations to TKIs in NSCLC patients, particularly of Asian origin (70, 77, 78). In the clinical trial that compared erlotinib with a placebo for NSCLC 325 samples were analyzed for EGFR expression and 177 samples were analyzed for egfr mutations (54). In contrast with other series in the multivariate analyses, adenocarcinoma (P=0.01), never having smoked (P<0.001), and expression of EGFR (P=0.03) were associated with an objective response, but survival after treatment with erlotinib was not influenced by the status of EGFR expression, the number of egfr copies, or egfr mutations (although EGFR expression and gene copy number appeared to be predictive in the univariate analysis). However, several methodological criticisms can be raised, including that mutational analysis was conducted in less than 25 percent of randomized patients, and that there is no indication that the sequencing was repeated in those positive cases (which may account for the high incidence of non-reported mutation types).

Beating The Butt On Your Own

Beating The Butt On Your Own

Need To Stop Smoking? Are You Willing To Follow My Powerful Strategies To Stop Smoking And Vividly Transform Your Life Today? Proven Tips, Tools and Tactics To Stop Smoking And Live An Awesome Life You Always Wanted.

Get My Free Ebook


Post a comment