Kategorie: Zhoubné nádory plic a průdušek
Číslo abstraktu: 006
The development of personalized medicine with a focus on novel targeted therapies has supplanted the one-size-fits-all approach to the treatment of many cancers, including non-small cell lung cancer (NSCLC). Targeted therapies, if given to a patient subpopulation enriched by the presence of relevant molecular targets, can often abrogate cell signaling that perpetuates cancer progression. Critical targets activating pro-cancer pathways include,but are not limited to, epidermal growth factor receptor (EGFR),MET, vascular endothelial growth factor (VEGF), VEGF receptor (VEGFR), KRAS, HER2, EML4-ALK, PIK3CA, and insulin-like growth factor 1 receptor (IGF-1R). Some target-directed therapies, such as EGFR tyrosine kinase inhibitors and anti-VEGF monoclonal antibody, have already been approved for clinical use.
Others, such as those targeted to MET, VEGFR, EML4-ALK, and IGF-1R, are in clinical testing. This review describes molecular targets in NSCLC that are in development or being clinically applied and their implications for developing novel anticancer therapies for this previously refractory malignancy.
The treatment outcomes in advanced/metastatic non-small cell lung cancer (NSCLC) remain unsatisfactory. Standard-of-care palliative chemotherapeutic regimens only modestly prolong survival. (1, 2) Major discoveries in the molecular biology of human malignancies has led to successful application of targeted therapeutic strategies in several cancers. (3-6) Substantial progress has been made towards understanding the tumor biology of NSCLC during the last five years.
Epidermal receptor growth factor (EGFR) tyrosine kinase inhibitors (TKIs) (erlotinib) and antibody targeting vascular endothelial growth factor (VEGF) (bevacizumab) have been approved for clinical use. (7, 8) In addition, other therapies targeting EGFR, MET, VEGF receptor VEGFR), EML4-ALK, HER2, insulin-like growth factor 1 receptor (IGF-1R), and others are in clinical testing. (9-11) Furthermore, the first results of large-scale efforts investigating relevant molecular aberrations have been recently published(12, 13) This article delineates the role of molecular targets in NSCLC that are in development or being clinically applied.
EGFR is instrumental in activating the PBK/AKT/mTOR and RAS/RAF/MEK/MAPK signaling pathways in solid tumors (Figure 1). (14) These pathways are important in tumor cell growth, local invasion, angiogenesis, protein translation, autophagy, and cell metabolism. EGFR gene amplification is present in 30% to 60%(15-17) of NSCLC cases. Furthermore EGFR mutations in the tyrosine kinase domain are found in approximately 10% of Caucasian patients and in more than 30% of East Asian patients with NSCLC (Table 1). (18-21) EGFR has been clinically investigated for more than a decade as a potential target for anticancer therapy. Therapeutic strategies for targeting EGFR include TKIs and monoclonal antibodies.
EGFR tyrosine kinase inhibitors alone
Two principal EGFR TKIs, erlotinib and gefitinib,demonstrated promising clinical activity. (7, 22) Erlotinib, gained Food and Drug Administration (FDA) and European Medicines Agency (EMEA) approval based on the results of a pivotal international phase III trial, which demonstrated a survival advantage compared to placebo (6.7 months versus 4.7 months, hazard ratio [HR] 0.70, p < 0.001) in previously treated patients with advanced NSCLC (Table 2). (7) In a retrospective subanalysis, erlotinib yielded better results in women, never-smokers, and in patients with adenocarcinomas. However, there was a 2.1 month overall survival (OS) benefit in the same retrospective analysis, even in the subgroup of male smokers with squamous-cell carcinoma, who would theoretically be the most disadvantaged population.(23) Tissue samples were available from 24% of patients for EGFR mutation analysis. Patients who had tumors with classic EGFR-activating mutations had a hazard ratio (HR) for death of 0.65, but this was not statistically significant which may have been attributed to the small number of patients assessed, as only 19 patients had EGFR activating mutations.
A trial testing gefitinib versus placebo in previously treated patients with advanced NSCLC showed no survival advantage despite the robust sample size (n = 1.692) (Table 2).(24) Median survival trended toward favoring gefitinib (5.6 months versus 5.1 months, p = 0.087), and time to treatment failure was longer in the gefitinib group than in the placebo group (3.0 versus 2.6 months, p = 0.0006) (Table 2). Preplanned subgroup analyses demonstrated a significantly longer survival in the gefitinib group than in the placebo group for patients who would be anticipated to have an increased rate of EGFR-activating mutations, including never-smokers (median survival 8.9 versus 6.1 months, p = 0.012), and in patients of East Asian origin (median survival 9.5 versus 5.5 months,p = 0.01). Similar to the erlotinib study, other retrospective efforts have attempted to identify possible predictors of better outcomes in NSCLC. (15) However, in this study, tissue samples were only available in 27 % of cases. EGFR mutations were detected in 12% of tested samples. In this study, a striking difference in the response rate (RR) to gefitinib was observed in patients harboring EGFR mutations compared to patients with wild-type EGFR (37.5 % versus 2.6 %). Unfortunately, the responding subgroup was too small to be evaluated for survival.
Why phase III studies showed improved OS only for erlotinib and not for gefitinib remains to be fully understood, particularly in view of various other positive study findings. While erlotinib may, in fact, be superior to gefitinib, the two agents have not been compared directly with each other, so that there are no data to support a true difference between their efficacies. Furthermore, the RRs (8.9 % for erlotinib and 8 % for gefitinib) were similar for the two agents in both randomized studies, and the median time to failure for gefitinib (3 months) was somewhat longer than the median progression-free survival (PFS) with erlotinib (2.2 months) (Table 2).(7, 24) A second possibility is that subsequent therapy had a role in OS. Both agents were associated with small, but statistically significant prolongations of PFS or time to treatment failure. A third possibility may reside in the diverse patient populations being treated. Simulations have demonstrated that if an agent works only in a small subpopulation of patients exhibiting a drug target, and not in populations that do not exhibit the target, then very small differences in the proportions of patients expressing the target can determine whether an agent will or will not have a significant impact on survival.(25) Additionally, patients treated on the erlotinib study demonstrated a higher median RR to prior therapy than patients treated on the gefitinib trial (38 % for erlotinib versus 18 % for gefitinib), which suggests the possibility of causative differences between the two study populations themselves. Importantly, erlotinib, unlike gefitinib, was administered at the maximum tolerated dose and was associated with a higher probability of patients developing a rash than gefitinib. It is well documented that the efficacy of EGFR TKIs correlates directly with rash severity.(26) Despite the fact that higher doses of gefitinib did not show better efficacy in earlier phase II trials,(27, 28)the possibility that gefitinib was dosed suboptimally in the phase III population cannot be completely disregarded.
The hypothesis that gefitinib is active in patients harboring EGFR mutations was tested in a small phase II trial.(29)In total, 30 EGFR mutation-positive patients were enrolled, including 22 with a poor Eastern Cooperative Oncology Group performance status of 3 or 4. The overall RR was 66% and the total disease control (complete response, partial response, stable disease) rate was 90 %. The performance status improvement rate was 79 %, and 68 % of the 22 patients improved from performance status 3 at baseline to performance status 1. Median PFS reached 6.5 months and median OS 17.8 months. These observations were confirmed in another phase II trial, which enrolled 31 patients with advanced NSCLC and underlying EGFR mutations. (30) Only two patients progressed on therapy, demonstrating an overall RR of 55 %, median PFS of 9.2 months, and projected OS of 17.5 months. Of the two patients who progressed, one had MET amplification, in keeping with the preclinical observation that MET amplification confers resistance to EGFR TKIs. (31) The Spanish Lung Cancer Group screened 2.105 chemonaive or previously treated patients with NSCLC and found EGFR mutations in 350 patients (16.6 %). Of these 350 patients, 217 were treated with erlotinib, yielding an overall RR of 70.6 %, median PFS of 14 months and median OS of 27 months. (19)
In the first-line setting, Mok et al. (32) carried out a large phase III randomized trial comparing gefitinib with standard carboplatin/paclitaxel chemotherapy in chemonaive East Asian patients with advanced lung adenocarcinoma who had a limited history of smoking (never-smokers and ex-light-smokers). The trial was designed as a noninferiority study in which the patient population was enriched by inclusion criteria to increase the likelihood of response to gefitinib based on retrospective observations from earlier studies. (7, 24) More than 1.200 patients were randomized (Table 2). The trial exceeded its primary endpoint (noninferiority), showing in the intent-to-treat population a 26% risk reduction of disease progression with gefitinib (HR 0.74, 95 % CI 0.65-0.85,p < 0.001). Gefitinib also achieved a greater RR compared with carboplatin/paclitaxel (43 % versus 32.2%, p < 0.001). OS was comparable between the two arms, but the patients on gefitinib had a better quality of life as assessed using the FACT-L questionnaire. Tissue samples were available in 56.1 % of cases. EGFR was mutated in 59.7 % patients who had available tissue samples and those patients had a 71.2 % RR with gefitinib compared to 47.3 % for patients treated with chemotherapy. In contrast, only 1.1 % of patients with no EGFR mutation responded to gefitinib, whereas 23.5 % responded to chemotherapy. Similarly, PFS was longer in the gefitinib arm compared with chemotherapy in patients with EGFR mutations (HR 0.48, 95% CI 0.36-0.64, p < 0.001). The opposite was true for patients without mutations (HR 2.85, 95% CI 2.05-3.98, p < 0.001). On the basis of these data, gefitinib was approved by the EMEA as a first-line treatment for patients with advanced or metastatic NSCLC with underlying EGFR mutations. PFS and RR data based on specific EGFR mutations were used for post hoc analysis. (33) The subgroup of 66 patients with an exon 19 deletion had a significantly prolonged PFS following gefitinib treatment compared with chemotherapy, as reflected by a HR of 0.377 (95% CI 0.255-0.560). RR data showed that gefitinib produced a response in 84 % of patients and chemotherapy produced responses in 43.2 % of patients. In 64 patients who had an exon 21 L858R mutation, PFS was increased by gefitinib compared with carboplatin and paclitaxel, albeit with a less robust HR of 0.553 (95% CI 0.352-0.868) and a smaller difference in RR (gefitinib 60.9 % versus chemotherapy 53.2 %). FISH subanalysis showed that patients with a high EGFR copy number and wild-type EGFR did not benefit from gefitinib, but did benefit from chemotherapy. A limitation of these results was the small sample size (n = 55).
A smaller phase III trial assessing gefitinib versus carboplatin and paclitaxel randomly assigned 230 chemonaive patients with EGFR mutations to receive either gefitinib 250 mg daily or chemotherapy (Table 2).(34) The median PFS (10.8 months for gefitinib, 5.4 months for chemotherapy; p < 0.001) and overall RR results (73.7 % for gefitinib, 30.7 % for chemotherapy; p < 0.001) favored gefitinib. The median OS rates numerically, but not statistically, also favored gefitinib (30.5 months for gefitinib, 23.6 months for chemotherapy; p = 0.31).
Recently, similar outcomes were replicated with erlotinib, which was compared to carboplatin and gemcitabine given as first-line therapy in a randomized phase III trial on 165 patients with advanced/metastatic NSCLC and activating EGFR mutations (Table 2). (35) Patients treated with erlotinib demonstrated nearly tripled PFS of 13.1 months in comparison to median PFS of 4.6 months on chemotherapy (p < 0.0001). OS data are pending. Prospective and retrospective studies consistently showed overall better outcomes in patients with EGFR mutations treated both with EGFR TKIs and chemotherapy. (32, 34, 36, 37) EGFR mutations are predictive of a high RR and prolonged PFS in patients treated with anti-EGFR TKIs, which does not, however, translate into a benefit in OS.
EGFR tyrosine kinase inhibitors as maintenance therapy
EGFR TKIs are administered orally and have a favorable toxicity profile, which supports their possible role as NSCLC maintenance therapy. Erlotinib was tested as a maintenance therapy in unselected patients with NSCLC without progression after four cycles of platinum doublets(Table 2)38. Altogether, 1.949 patients were registered at the time of chemotherapy initiation and 889 were randomly assigned to receive either erlotinib or placebo. The trial met its primary endpoint, demonstrating a HR of 0.71 (p < 0.0001) favoring the erlotinib maintenance arm, however, the difference in median PFS compared to placebo was only 8.4 days (2.9 months versus 2.6 months). EGFR mutation analysis was performed in about 40 % of patients. Although patients with wild-type EGFR had a small PFS benefit from erlotinib (HR 0.78, p = 0.0185), the most striking difference was observed in a small patient subpopulation with EGFR mutations (approximately 10 % of the tested samples). Patients with EGFR mutations who were treated with erlotinib had a HR of 0.10 (p < 0.0001). The difference in median PFS for the whole study population translated into a survival advantage (12 months for erlotinib versus 11.1 months for placebo). Interestingly, the survival difference was maintained even in patients without EGFR mutations (HR 0.77, p = 0.0243). Although this study suggested that patients with EGFR mutations might derive a meaningful benefit from erlotinib; some patients with wild-type EGFR derived a small, albeit statistically significant, PFS and OS benefit from erlotinib maintenance therapy. Therefore complex genetic or epigenetic changes might be associated with response to anti-EGFR therapy in the absence of EGFR mutations. To date, efforts to identify the important changes have not been conclusive.(39)
EGFR tyrosine kinase inhibitors with chemotherapy
Erlotinib and gefitinib were each tested in combination with a standard platinum-containing front-line therapy in four large randomized trials (Table 2).(40-43) Unfortunately, there was no improvement in treatment outcomes in any of them. The disappointing results could be accounted for by various factors. First, gefitinib and erlotinib were examined in unselected patient populations. EGFR TKIs are potent drugs in a small proportion of selected NSCLC patients, but the overall success of these agents in an unselected patient population is overshadowed by the majority of patients who are resistant to them. Furthermore, crossover to second-line EGFR TKIs could have had a role. EGFR TKIs might antagonize chemotherapy effects by blocking cells in the G1 phase of the cell cycle, as well as interfere with platinum uptake into tumor cells, by decreasing membrane uptake transporter expression(44, 45) Since the effect on membrane transporters could last for several weeks or longer, EGFR TKIs could hypothetically impact negatively on the efficacy of chemotherapy given concurrently with an EGFR TKI, but also chemotherapy administered two to three months fter discontinuation of the targeted agent.(46)
Types of EGFR mutations
EGFR activating mutations occur in the region of the ATP-binding pocket of the tyrosine kinase domain. (47) EGFR mutations target four exons (18-21) (Table 3). (48, 49) The most prevalent mutations associated with sensitivity to EGFR TKIs are in-frame deletions of exon 19 (44 %) and L858R substitutions in exon 21 (41 %).(50) Nucleotide substitutions in exon 18 and in-frame deletions of exon 20 are less frequent (5 % and < 1 %, respectively). Despite dramatic responses to EGFR TKIs, most patients eventually develop disease progression in less than a year. (34) A single secondary mutation, T790M in exon 20, is found in about 50 % of patients relapsing after initial response to EGFR TKIs. (49, 51) Much less prevalent is a D761Y secondary mutation of exon 19 associated with resistance to EGFR TKIs in fewer than 1 % of patients who initially responded to EGFR TKIs.
EGFR monoclonal antibodies
Cetuximab is a chimeric monoclonal antibody directed against the extracellular domain of EGFR,
which has activity in a broad spectrum of tumor types, including lung cancer.(52-54)
Promising results from randomized phase II trials led to initiation of a study that compared cisplatin and vinorelbine with or without cetuximab in 1.125 patients with advanced EGFR-overexpressing NSCLC (Table 2).(55) Although this trial did not show any difference in PFS, it met its primary endpoint, demonstrating a modest survival gain of 1.2 months for cetuximab containing arm (11.3 months versus 10.1 months, HR 0.871, 95% CI 0.762-0.996, p = 0.044). The RR was marginally better in the cetuximab arm (36 % versus 29 %). Whether this result is clinically relevant is debatable; however, patients treated with cetuximab who developed a skin rash within the first 3 weeks of treatment achieved an impressive 15-month median survival versus 8.8 months in patients with no rash. (56) Whether this side-effect reflects the higher efficacy of cetuximab, or whether patients with skin rash have a better prognosis regardless of a specific treatment regimen, remains to be seen.
Prespecified subanalysis showed that cetuximab provides bene fit irrespective of histology. Biomarker analysis reflecting EGFR copy number on FISH and KRAS mutation showed no significant predictive value for any of the agents assessed, although the HR favoring cetuximab was slightly better in patients with EGFR amplification on FISH.(17)
A similar trial was carried out in the USA in nonselected patients with stage IIIB/IV NSCLC, which randomized 676 patients (Table 2). (57) This study did not meet its primary endpoint, as the difference in PFS of 0.16 months did not reach statistical significance despite a better RR (25.7 % versus 17.2 %) in the experimental arm. OS curves were overlapping. Tissue samples for biomarker subanalysis were available for 225 (33 %) patients. EGFR expression, EGFR copy number, EGFR mutations, and KRAS mutations did not influence treatment outcomes. (58)
The only study suggesting a prognostic or predictive value associated with EGFR FISH status was a phase II trial, SWOG 0342, which randomized 229 patients to carboplatin and paclitaxel with concurrent or sequential cetuximab.(16) Biomarker data were available for 76 patients. Patients with high EGFR copy numbers had a longer PFS and OS in a retrospective analysis. Whether this finding reflects a prognostic or predictive value of EGFR status is difficult to assess, as both treatment arms contained cetuximab. Biomarker data in studies using cetuximab consistently showed that, unlike in colorectal cancer(59) , mutant KRAS does not predict NSCLC resistance to cetuximab. In addition, neither EGFR mutations nor EGFR amplification predicted a favorable response to cetuximab, suggesting that monoclonal antibodies have mechanisms of action that differ from anti-EGFR TKIs. Cetuximab might interact with other pathways that have not yet been described and might also alter non-kinase functions of EGFR, as demonstrated in the preclinical setting. (60) Additional biomarker studies are needed to predict the benefit of cetuximab in NSCLC. Another anti-EGFR monoclonal antibody, fully human panitumumab, is currently being tested in a randomized phase II clinical trial.
The MET oncogene encoding a transmembrane receptor with tyrosine kinase activity was identified nearly three decades ago. (61) The MET receptor is activated upon binding of the hepatocyte growth factor (HGF), which is secreted by mesenchymal cells. In cancer, MET is involved in complex prosurvival mechanisms, including downstream activation of the RAS/RAF/MEK and PI3K/AKT pathways (Figure 1).
MET amplification has been reported in 17% of patients with NSCLC and was associated with a dismal prognosis and resistance to EGFR TKIs.(30,31,62) Activating mutations in the tyrosine kinase domain of MET are observed in fewer than 3% of patients (Table 1).(63-65) Preclinical models have shown that MET is involved in resistance to angiogenesis inhibitors. (66) Antiangiogenic drugs induce HIF-1a expression, which in turn triggers transcription of multiple genes, including MET. (67) This association provides a rationale for combining MET inhibitors with EGFR TKIs or antiagiogenic drugs.
The MET pathway could be inhibited by monoclonal antibodies against HGF, monoclonal antibodies against a MET receptor, or by MET TKIs. A recent phase I trial with a fully human anti-HGF monoclonal antibody AMG 102 in combination with the antiVEGF monoclonal antibody bevacizumab in patients with advanced solid tumors demonstrated encouraging activity in NSCLC patients. (68) An anti-MET receptor monoclonal antibody, PRO143966 (MetMab), is currently being evaluated in a randomized phase II study in combination with erlotinib. Early results suggest possible benefit in patients with IHC overexpression of MET. (69) ARQ 197 is a selective, non-ATP competitive inhibitor of MET kinase. A phase II randomized trial compared ARQ 197 in combination with erlotinib to placebo and erlotinib in chemotherapy pretreated, but EGFR TKI-naive, patients with advanced NSCLC (Table 2). The trial did not meet its primary endpoint as the increase in median PFS in the experimental arm was not statistically significant (3.8 months versus 2.3 months; p = 0.23); however, when adjusted for prespecified factors such as histology, KRAS, and EGFR mutations, there was a significant 32% risk reduction of disease progression in the experimental arm (HR 0.68, p < 0.05). (70) Patients with nonsquamous histology, wild-type EGFR, and mutated KRAS fared best on the experimental arm. XL184 is an oral inhibitor of MET, VEGFR, RET, KIT and TIE-2. Preliminary data from an ongoing phase Ib/II study evaluating the safety and efficacy of XL184 in combination with erlotinib in patients with advanced NSCLC produced encouraging results in patients with an EGFR T790M mutation (indicating resistance to EGFR TKIs) and in patients with MET amplification. (71)
VEGF AND VEGFR
Bevacizumab, a monoclonal antibody against VEGF, increased survival in NSCLC patients when added to standard carboplatin and paclitaxel chemotherapy. (8) The pivotal phase III study (ECOG 4599) randomized patients who had advanced NSCLC with a nonsquamous histology. Adding bevacizumab led to a longer OS of 12.3 versus 10.3 months, a longer PFS of 6.2 versus 4.5 months, and a higher RR of 35% versus 15% (Table 2). Based on this study, bevacizumab gained FDA and EMEA approval as first-line therapy for advanced nonsquamous NSCLC. The AVAiL (Avastin in Lung Cancer) study evaluated the regimen of cisplatin and gemcitabine with placebo or in combination with one of two doses of bevacizumab (7.5 mg/kg or 15 mg/kg every 3 weeks). A statistically significant, but clinically minimal difference of 0.4-0.6 months (p = 0.03, and p = 0.003) in PFS (primary endpoint) was seen when bevacizumab was added. (72) No dose-effect relationship in the bevacizumab arms was observed. Nevertheless, the chemotherapy arm demonstrated a better than expected PFS of 6.2 months. Updated results revealed no significant difference in OS (13.1, 13.6, and 13.4 months, respectively; (Table 2). (73) More than 60% of patients treated in this trial received some post-protocol therapy, which may explain the lack of survival benefit. Data on post-protocol therapies from the pivotal ECOG 4599 study are not available, which precludes drawing definitive conclusions. An additional conundrum is the lack of validated biomarkers that can identify which patients are likely to benefit from the addition of bevacizumab. High blood pressure has been associated with prolonged PFS and OS when bevacizumab was added to chemotherapy in a retrospective analysis of ECOG 4599. (74) This needs to be confirmed by other studies with bevacizumab in NSCLC, including further insight on the downstream effects of VEGF suppression and hypertension.
VEGFR tyrosine kinase inhibitors
Sunitinib, and sorafenib are TKIs that target multiple kinases, including the VEGFR, and they have been approved for therapy in metastatic renal-cell carcinoma, gastrointestinal stromal cancer (GIST) and hepatocellular carcinoma. (75-77) Both drugs showed promising results in phase II trials in NSCLC, (78, 79) although a pivotal phase III trial in advanced NSCLC patients demonstrated no advantage when sorafenib was added to carboplatin and paclitaxel (Table 2). (80) Currently, sorafenib is being compared with placebo in the third-line and fourth-line therapy of NSCLC. The trial should be completed by 2011.
A randomized phase II trial tested a VEGFR/EGFR/RET inhibitor, vandetanib, as a monotherapy or in combination with paclitaxel and carboplatin compared with paclitaxel and carboplatin in 181 chemonaive, advanced NSCLC patients (Table 2). (81) The vandetanib monotherapy arm was stopped early since it was less effective than chemotherapy. No survival difference was observed between the two remaining treatment arms; however, vandetanib added to chemotherapy delayed progression by one week. Vandetanib was also tested as second-line therapy in combination with docetaxel in a large phase III trial, which randomized 1.391 patients to either docetaxel and vandetanib or docetaxel and placebo (Table 2). (82) Early results from this trial showed a significant difference in RR favoring vandetanib (17% versus 10%, p < 0.001). The trial met its primary endpoint of PFS, favoring the experimental arm (4 months versus 3.2 months, HR 0.79, 95% CI 0.70-0.90, p < 0.001). The OS did not significantly differ among treatment arms (10.6 months for docetaxel and vandetanib versus 10 months with docetaxel and placebo, HR 0.91, 95% CI 0.78-1.07, p = 0.196). Assessment of EGFR mutational analysis suffered from the relatively small sample size.
Other antiangiogenic treatments, such as VEGFR inhibitors (cediranib, motesanib, pazopanib) and
vasculature disrupting agents are being tested in numerous phase III studies.
EML4-ALK fusion is a rare abnormality detected in approximately 5-7% of patients with NSCLC( Table 1), (83, 84) a frequency that doubled (13%) in a population of patients with at least two of the following characteristics: female gender, Asian ethnicity, never or light smoking history, and adenocarcinoma histology. (85) Activated ALK, similar to EGFR/HER2, might constitutively switch on the RAS/RAF signaling pathway (Figure 1). Patients with EML4-ALK are more likely to be light or never-smokers, similar to patients with EGFR mutations, although there is an association with younger age (median age about 50 years) and, per some authors, with male gender. (85) EML4-ALK is associated with wild-type EGFR and wild-type KRAS. (83) Patients with EML4-ALK fusion are resistant to EGFR TKIs. The response to platinum-based chemotherapy does not, however, seem to be affected by the presence of the EML4-ALK fusion gene.
Patients with the EML4-ALK fusion demonstrated an extraordinary response to the MET and ALK inhibitor crizotinib in a phase I/II trial. (86)Initially, in the dose-escalation portion of this trial,a response was seen in a patient with NSCLC and an underlying ELM4-ALK fusion. In the phase II cohort, 82 NSCLC patients with ELM4-ALK fusion were assessed. Most of them (59 %) had been heavily pretreated with at least two prior regimens. Most patients never smoked (never-smokers 76 %, ex-smokers 23 %, current smokers 1 %) and all but three had adenocarcinomas. In total, 47 out of 82 patients (57 %) experienced objective responses and PFS at 6 months was 72%. Only six patients showed disease progression. This trial reiterated the importance of incorporating prospective molecular profiling into the selection criteria for early-phase clinical trials examining targeted therapies. The phase III clinical trial comparing crizotinib with chemotherapy in a second-line setting is currently underway.
Activating mutations in codons 12 and 13 of the KRAS oncogene occur in about 10-30 % of lung adenocarcinomas(17, 37, 58, 87, 88) KRAS mutations are usually mutually exclusive to EGFR mutations and occur early in the development of smoking-related carcinomas (Table 1).(89) Although it might be assumed that patients with KRAS-mutated carcinomas (mostly adenocarcinomas) would be resistant to EGFR TKIs or monoclonal antibodies, similar to patients with metastatic colorectal cancer and mutated KRAS, subanalyses of studies contradict this hypothesis in NSCLC and showed similar survival profiles after gefitinib or cetuximab treatment, irrespective of KRAS status.(17, 37, 58)Generally, in NSCLC, mutant KRAS seems to be associated with lower response rates to anti-EGFR therapies than wild-type EGFR.(15, 31,90-93) Because no available drug blocks KRAS directly, studies are evaluating other potential targets in the RAS/RAF/ MEK pathway that function downstream of RAS (Figure 1). Apart from sorafenib, which is a relatively weak RAF inhibitor, encouraging preclinical data suggest a potential benefit from MEK inhibitors. (94-96) These compounds are currently in the early stages of clinical research.
The HER2 oncogene is a member of the EGFR family, which in addition includes EGFR, HER3 and HER4. (97) These members encode transmembrane receptors that drive and regulate cellular functions of cell proliferation. (98) The HER2 receptor does not have a known ligand and is putatively activated by homodimerization with other HER2 receptors or by heterodimerization, preferentially with either EGFR or HER3.(99) The role of the HER4 receptor in cancer is less defined.(100)HER2 and HER3 heterodimers seem to be more potent than HER2 homodimers.(101) HER2 putatively activates the PI3K/AKT/mTOR and RAS/RAF/MEK pathways. (98)
HER2 overexpression and amplification have been found in diverse tumors, including breast, gastric and salivary gland cancers.(102-104)In breast and gastric cancer, HER2 overexpression or HER2 amplification indicate sensitivity to the anti-HER2 humanized monoclonal antibody trastuzumab.(105, 106)In NSCLC, HER2 amplification is found in 2-23 % of patients, however, these patients do not seem to benefit from anti-HER2 monoclonal antibodies (trastuzumab) or HER2 TKIs (lapatinib).(107-110) Interestingly, HER2 mutations in exon 20 of the tyrosine kinase domain were found in 3-10 % of lung adenocarcinomas (Table 1).(112) The most frequent mutation is YVMA 776-779 ins, which was associated with constitutive activation of HER2 kinase, leading to phosphorylation of downstream effectors such as AKT and MEK.)(113) In the same preclinical experiments, cell lines with mutated HER2 were also sensitive to the HER2 tyrosine kinase inhibitor lapatinib. In NSCLC HER2, EGFR, and KRAS mutations seem to be mutually exclusive. However, a small proportion of patients has coexisting HER2 and EGFR mutations. (112, 114) Patients with HER2 mutations are resistant to anti-EGFR TKIs irrespective of their EGFR mutation status as pro-growth signals are executed through HER2 kinase. (113) HER2 receptor provides an attractive target for anticancer therapy in patients with activating mutations. Several clinical trials investigating the efficacy of anti-HER2 antibodies (trastuzumab, MGAH22) and HER2 inhibitors (PF-00299804, BIIW 2992) are currently underway.
The PI3K/AKT/mTOR pathway is activated in many cancers,including NSCLC. This signaling pathway plays an important role in the early stages of lung cancer development.(115)
Molecular aberrations such as amplification or mutations of the p110a subunit,PIK3CA, can lead to the activation of the pathway. PIK3CA amplification has been reported in about 12-17% of patients with NSCLC. (116, 117) Mutations in the helical or kinase domain of PIK3CA were reported in 2% to 13% of NSCLC patients (Table 1). (117-119) Numerous drugs interfere with this pathway, including PI3K, AKT and mTOR inhibitors. Some of them,such as rapalogs inhibiting the mTOR Complex 1 (mTORC1), temsirolimus and everolimus, are already approved by the FDA and EMEA for other indications. (120, 121) Preclinical data and early clinical experiments suggest that coexisting KRAS and PIK3CA mutations may be associated with resistance to PI3tyAKT/mTOR axis inhibitors. (119, 122) Importantly, these agents might be active even in the absence of a PIK3CA mutation owing to frequent alterations at other levels of the PI3K/AKT/mTOR pathway, such as PTEN loss, AKT mutations, and other alterations. In addition, activation of PI3K/AKT/mTOR pathway may be involved in resistance to EGFR TKIs. (123)
BRAF mutations in the G-loop (exon 11) or activation segment (exon 15) of its kinase domain have been reported in 3% of patients with NSCLC.(124) BRAF mutations in lung cancer are early events in lung cancer tumorigenesis and they are qualitatively different from V600E BRAF mutations in malignant melanoma, which may have implications for BRAF inhibitor efficacy. (125) Preclinical data suggested that BRAF mutations might predict sensitivity of NSCLC cells to MEK inhibitors (Figure 1). (126) This hypothesis is now being tested in early clinical trials, with several MEK inhibitors currently in clinical development.
The insulin-like growth factor 1 receptor (IGF-1R) is a promising novel therapeutic target in NSCLC.(127) In operable NSCLC, an increased copy number of the IGF-1R gene was associated with a better prognosis(128) At the cellular level, IGF-1R is activated upon binding of insulin-like growth factor (IGF). (129) IGF-1R signaling involves the activation of various intracellular signaling pathways, including the RAS/RAF/MAP kinase and PI3K/AKT pathways (Figure 1).
A randomized phase II trial demonstrated a better RR when figitumumab, a monoclonal antibody against IGF-1R, was added to standard carboplatin and paclitaxel chemotherapy (54 % versus 42 %, p < 0.001). (130) The RRs were particularly encouraging in tumors with a squamous histology, supporting the hypothesis that deregulation of the IGF-1R pathway may be common in squamous histology NSCLC. (131) Unfortunately, a subsequent phase III trial comparing figitumumab plus chemotherapy versus chemotherapy alone was terminated early for futility. (132) Another, fully human anti-IGF-1R monoclonal antibody, cixutumumab, is currently being evaluated in phase II randomized trials.
The discovery of EGFR mutations revolutionized treatment of NSCLC and triggered the paradigm shift from a one-size-fits-all to more personalized approaches. Currently, there are myriad novel therapeutic targets currently being investigated in laboratory and clinical experiments. An analysis of the major therapeutic advances in cancer treatment demonstrated that most breakthroughs, whereby drugs produce very high response rates in a tumor type, have been observed in uncommon tumors. (3) Classic examples, among others, are the use of KIT kinase inhibitors in KIT mutation-positive GIST and the use of BCR-ABL inhibitors in chronic myelogenous leukemia. (4,5) These tumor types develop predominantly as a result of a single molecular aberration; this small repertoire of underlying single abnormalities may be responsible both for the rare occurrence of associated tumors and being amenable to treatment once an appropriate targeted agent is identified. By contrast, it seems plausible that common tumors, such as NSCLC, are composed of multiple subsets of disease, each with its own molecular abnormalities (examples include EGFR-mutation-positivity, EML4-ALK-positivity, HER2 mutations, and others). Identifying the relevant molecular subtypes of this heterogeneous disease, and matching patients with appropriate targeted agents rather than performing large trials in unselected patients, is crucial if we are to make headway.(133)
According to this paradigm, recognizing even small subsets of disease may be critical. For example, HER2 mutations are identified in 3-10 % of NSCLC patients, yet deregulation of this signal might underlie resistance to EGFR-directed therapy. 111-113) It is highly likely that many patients have several abnormalities driving tumorigenesis and that single targeted agents will be insufficient for providing meaningful therapeutic results. Indeed, MET amplification or mutation is detected in 3-17 % of NSCLC patients, and may augment resistance to anti-EGFR therapy.(30,62)
To address this issue, clinical trials that combine EGFR and MET kinase inhibitor therapy have been developed.
In conclusion, chemotherapy improves survival in patients with advanced NSCLC,although its ultimate success has been limited. New therapeutic combinations involving targeted therapies and strategies hold promise for improved treatment outcomes, even in resistant patients. To achieve this benefit current classification schemes must be revised to incorporate molecular features in order to better address the requirements of a targeted therapy approach within the context of personalized medicine, and enable researchers to add promising new drugs to their therapeutic
armamentarium while sparing patients without relevant molecular aberrations from unnecessary treatments.
We thank to Ms. Joann Aaron for scientific review and editing of this article.
Disclosure of Conflicts of Interest
None of the authors have any conflict of interest relevant to the subject of this manuscript.
Table 1: Molecular abnormalities in NSCLC
Abbreviations: NSCLC, non-small cell lung cancer; TKIs, tyrosine kinase inhibitors
* Frequency in Caucasian population
** Frequency in East Asian population
*** Combined high copy number and polysomy
Table 2: Major randomized trials with targeted therapies for advanced NSCLC
Table 3: EGFR mutations
Abbreviations: TKIs; tyrosine kinase inhibitors
Figure 1: Therapeutic targets and novel agents in NSCLC
Janku F.(1), Garrido-Laguna I.(1), Petruzelka L.(2), Stewart DJ(3), Kurzrock R.(1)
(1)Department of Investigational Cancer Therapeutics (Phase I Clinical Trials Program), The University of Texas MD Anderson Cancer Center, Houston, Texas, 1515 Holcombe Blvd., Box 455, Houston, TX 77030, USA, (2)Department of Oncology, First Faculty of Medicine, The Charles University Prague, U Nemonice 2, 128 08 Praha 2, Czech Republic, (3)Department of Thoracic/Head & Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA.
Datum přednesení příspěvku: 28. 1. 2011