KRAS Mutations and Resistance to Cetuximab
KRAS MUTATIONS AND CETUXIMAB MONOTHERAPY. In 2006, Lièvre and colleagues reported results from a retrospective study of frozen tumor tissue from 30 patients with metastatic CRC considered to be EGFR positive by immunohistochemistry (IHC) and treated with cetuximab alone (1 patient), cetuximab combined with irinotecan (25 patients), or cetuximab combined with irinotecan plus 5-FU and leucovorin (FOLFIRI regimen, 4 patients).14 In most cases, cetuximab was given as a third-line or later therapy following disease progression with irinotecan-based chemotherapy. KRAS mutations were present in tumors from 13 patients (43%). While none of the 11 patients who responded to cetuximab had a KRAS mutation, 13 of 19 non-responders harbored tumors with mutated KRAS. Patients with no evidence of KRAS mutations had a significantly longer overall survival (OS) compared with those who had tumors with the mutated gene (median 16.3 versus 6.9 months, respectively; P = .016). To explain these results these authors hypothesized that regardless of the expression level of EGFR, mutations in KRAS are able to activate the downstream RAS/MAPK pathway inducing cellular proliferation through signaling pathways that are not inhibited by cetuximab, which acts upstream from the KRAS protein.
More recently, the same group of French investigators reported on a retrospective study of a larger independent series of frozen or paraffin-embedded samples from 89 patients with metastatic CRC treated with cetuximab after disease progression with irinotecan-based chemotherapy.15 As seen in the previous study, the presence of KRAS mutations, which occurred in 27% of patients, was associated with resistance to cetuximab therapy, with 0% of responders having KRAS mutations compared with 40% of non-responders (P < .001). Patients with KRAS mutations also had shorter median progression-free survival (PFS) and median OS. Not surprisingly, these results were confirmed when the authors performed a multivariate analysis of pooled data from both studies.
In 2007, Khambata-Ford et al published results from an international, prospective exploratory clinical trial aimed at systematically identifying biomarkers associated with disease control to cetuximab therapy.16 This study enrolled 110 patients with metastatic CRC and at least 1 prior chemotherapy regimen for advanced disease. All patients had a pretreatment biopsy. Gene expression profiling was used to identify candidate biomarkers associated with disease control (to be discussed in a later section) and mutational analysis of EGFR, KRAS, and BRAF were also performed. DNA analysis of tumor biopsies from 80 patients showed that KRAS mutations were present in 3 (11%) of 27 patients with disease control (ie, who had an objective response or stable disease) and 27 (51%) of 53 non-responding patients, suggesting a correlation between the presence of KRAS mutations and lack of response to cetuximab therapy (P = .0003). These authors did not find a significant difference in PFS depending on KRAS status, which they thought might be due to insufficient power to detect differences given the relatively small number of patients with KRAS mutations and the early events experienced by patients with wild-type KRAS.
In a French retrospective study, Di Fiore et al found that the presence of KRAS mutations was highly predictive of non-response to cetuximab plus chemotherapy in a series of 59 patients with previously treated metastatic CRC.17 Patients in this study were treated with a cetuximab plus irinotecan or cetuximab plus oxaliplatin chemotherapy regimen. DNA samples for polymerase-chain reaction (PCR) amplified KRAS sequencing analysis were obtained from paraffin-embedded tissue samples from primary tumors (n = 53) and from metastases (n = 6). The investigators detected KRAS mutations in samples from 16 (27%) of 59 patients. No KRAS mutations were found in the 12 patients who had a complete response (CR) or partial response (PR). In contrast among the 16 patients who had a mutation, 13 had progressive disease (PD) and 3 had SD. Di Fiore and coworkers noted that an important criticism of this type of predictive biomarker study is that the clinical assessment of response to EGFR therapy is made in metastatic disease whereas the presence of the KRAS mutation is assessed primarily from the primary tumor. To alleviate this concern within this small study, for the 5 patients for whom samples were available the KRAS mutation status was identical between the primary tumor and the metastasis.
De Roock and colleagues18 in Belgium reported on a retrospective study of KRAS mutational status in 113 patients with irinotecan-refractory metastatic CRC treated with cetuximab monotherapy or cetuximab plus irinotecan. They detected KRAS mutations in 40.7% of patients. KRAS mutations were found in none of the patients with a CR or PR and in 42 (51.9%) of 81 nonresponders (PD or SD). In the univariate analysis the difference in median PFS between patients with wild type KRAS and mutant KRAS did not reach statistical significance. However, among patients treated with combination therapy, those with wild type KRAS had a significantly longer median PFS compared with those with mutant KRAS (P = .016) The authors noted no statistical difference in median PFS when comparing the monotherapy groups. De Roock et al also observed that there may be a subgroup of patients with mutant KRAS who might benefit from long-term disease stabilization with cetuximab treatment. Others have reported individual cases of patients with mutated KRAS responding to cetuximab therapy.13,22
KRAS Mutations and Cetuximab Combination Therapy. At the ASCO 2008 GI Cancers Symposium, Tejpar et al reported their analysis of KRAS mutation state and EGFR ligand mRNA expression (discussed below) in paraffin-embedded archival samples of tumor primaries from 95 patients treated with cetuximab plus irinotecan for metastatic CRC.21 KRAS mutations, which were present in 35% of patients, were significantly associated with objective responses (P < .0001). The estimated median OS was 16.7 weeks for patients with KRAS mutations and 45.9 weeks for patients with wild-type KRAS (P = .0001).
KRAS Mutations and Resistance to Panitumumab
KRAS MUTATIONS AND PANITUMUMAB MONOTHERAPY. At the 2007 14th European Cancer Conference (ECCO 14) held in Barcelona, Spain, Freeman and colleagues reported on a retrospective KRAS mutation analysis of biopsy samples from patients with metastatic CRC enrolled in 4 panitumumab monotherapy safety and efficacy studies.19 Among the 709 patients treated in these studies, they identified 59 patients who gave consent for the study and for whom response data and pretreatment tumor samples were available for analysis. PCR and subsequent genomic sequencing detected KRAS mutations in tumors from 21 patients. As had been seen with cetuximab, none of the responders to panitumumab had KRAS mutations. Among the patients with wild type KRAS, 16% had a PR, 45% SD, and 39% PD. In contrast, among the patients with mutated KRAS, 24% had SD and 76% had PD. The authors reported a statistically significant association between KRAS mutation status and response to panitumumab (P = .013).
In early 2008, Amado and colleagues published final results of the largest study to date on the predictive value of KRAS in terms of response to an EGFR inhibitor. They successfully assessed mutant KRAS status from formalin-fixed, paraffin-embedded tumor sections from 427 of 463 patients with chemotherapy-refractory metastatic CRC enrolled in a phase III, randomized study that compared panitumumab plus best supportive care (BSC) with BSC alone.20 All patients had to have tumors with EGFR expression in 1% or more of tumors cells as assessed by IHC. The primary objective of the biomarker analysis was to determine if tumor wild-type versus mutant KRAS status had an impact on the relative effect of panitumumab compared with BSC on PFS.
KRAS mutations were found in samples from 40% of patients in the panitumumab treatment arm and 46% of patients in the BSC arm. The authors reported that the positive effect of panitumumab versus BSC alone on PFS was significantly greater in patients with wild-type KRAS compared with patients with mutant KRAS. Among the patients who had wild-type KRAS tumors, the median PFS was 12.3 weeks for panitumumab compared with 7.3 weeks for BSC. In contrast within the mutant KRAS group, the median PFS was 7.4 weeks for panitumumab compared with 7.3 weeks for BSC. The link between wild-type KRAS status and longer PFS compared with mutant KRAS status persisted when considering the 168 patients assigned to the BSC group who received panitumumab upon disease progression. Among these crossover patients, the PFS was 16.4 weeks for those with wild-type KRAS versus 7.9 weeks for those with mutant KRAS (Hazard ratio [HR], 0.32; 95% CI, 0.2 to 0.45). Response rates to panitumumab were 17% in the wild-type KRAS group and 0% in the mutant KRAS group.
Amado et al20 noted that the 100% predictive value for lack of objective response to panitumumab monotherapy associated with the presence of mutant KRAS suggests that the inhibition of the RAS/RAF/MAPK signaling pathway is the primary mechanism of action of panitumumab in metastatic CRC. They also pointed out that the presence of wild-type KRAS status was not sufficient for response to panitumumab monotherapy, indicating the presence of yet to be identified mechanisms for primary and treatment-emergent resistance to this anti-EGFR monoclonal antibody. Possible reasons given for primary resistance include
- ▪ The presence of dominant oncogenic pathways independent of EGFR
- ▪ The possible acquisition of KRAS mutant status later in carcinogenesis, leading to tumor heterogeneity
- ▪ The possibility that the KRAS assay used (which had been shown to detect > 90% of known activating mutations in CRC) missed some mutations
- ▪ Activation of additional tyrosine receptors such as VEGF, platelet-derived growth factor receptor, and insulin-like growth factor 1 receptor
- ▪ Activating mutations of other signaling proteins (eg, PI3K, Scr, RAF) acting downstream of EGFR
- ▪ Loss of function mutations of tumor-suppressor genes
The results of this trial were the basis for the September 2007 decision by the European Medicines Agency (EMEA; www.emea.europa.eu) to grant conditional approval, valid throughout the European Union, for panitumumab in the treatment of patients with EGFR expressing metastatic CRC with wild-type KRAS after failure of 5-FU plus irinotecan- or oxaliplatin-based chemotherapy.
KRAS and Panitumumab Combination Therapy. At the 2008 GI Cancers Symposium, Hecht et al presented an interim analysis of the irinotecan cohort enrolled in the US community-based PACCE phase III, randomized study.23 In this study, patients were assigned to irinotecan- or oxaliplatin-based first-line chemotherapy, depending on investigator's preference, and then they were randomized to receive bevacizumab with or without panitumumab. The primary endpoint, PFS, was powered only in the oxaliplatin cohort (N = 800). Analyses of the irinotecan cohort (N = 200) were designed as descriptive and included an exploratory biomarker analysis. Results of the analysis for KRAS status indicate that the addition of panitumumab to irinotecan-based chemotherapy plus bevacizumab was associated with a slight but significantly higher overall response rate in patients with wild-type versus mutant KRAS (Figure 2). Further analyses of biomarkers from PACCE were expected to be presented at the ASCO 2008 annual meeting.
KRAS as a Marker of Response to VEGF Inhibitors
Given that various data suggested the potential clinical relevance of the RAS/RAF/MEK/ERK pathway and inactivation of p53 to the efficacy of anti-VEGF therapy, Ince et al performed a retrospective analysis of data from participants in the AVF 2107 phase III randomized trial. 24 AVF 2107 was the pivotal trial to demonstrate that the addition of bevacizumab to first-line irinotecan plus bolus 5-FU/LV (IFL) significantly improved survival in patients with metastatic CRC. 25 The primary objective of the retrospective analysis was to determine if mutations in KRAS, BRAF, or p53, or P53 overexpression would help identify subgroups of patients more likely to respond to bevacizumab as determined by OS. 24 Another objective was to determine if these biomarkers might have prognostic value for OS, independent of treatment regimen. The investigators were able to obtain tumor samples and outcome data for 267 patients (33% of the overall AVF 2107 trial participants), who had demographic and clinical characteristics similar to the overall patient population. For the KRAS mutant versus wild-type patient subgroups, the hazard ratios for death were less than 1 when comparing OS in the IFL plus bevacizumab group with IFL plus placebo (Figure 3). Thus, according to these data KRAS status has no impact on panitumumab's positive effect as monotherapy in patients with metastatic CRC. The authors noted similar results for the other 3 biomarkers studied. However, patients with wild-type KRAS and BRAF genes had significantly improved OS irrespective of treatment, compared with patients who had one or both genes mutated. These data suggest that KRAS and BRAF mutational analysis might have a general prognostic value. Independent and preferably prospective studies are necessary to confirm these results.
Spotlight Commentary on KRAS Mutations and Resistance to Chemotherapy
According to Dr Heinz-Josef Lenz, the data described above clearly show that single-agent anti-EGFR therapy will not be of benefit for patients with CRC marked by mutant KRAS. However, the only 10% to 40% realistic chance that currently available EGFR inhibitor monotherapy will work in patients with wild KRAS clearly indicates the need for more markers (such as the EGFR ligands discussed below) to identify potential responders. Dr Lenz also said that data presented at ASCO 2008 from the CRYSTAL, EPIC, and OPUS studies (Table 3) would show whether KRAS will be a useful predictive biomarker for EGFR inhibitors given in combination with chemotherapy.
With regards to the type of KRAS testing currently available, Dr Lenz said that generally 2 types of tests are used: (1) sequencing, which requires tumor micro-dissection and might miss collecting a sample with mutated KRAS and (2) PCR with allele specific primers [such as the DxS test used in the Amado et al 20 panitumumab study and recently registered for marketing in the European Union (EU)]. Dr Lenz said that there are issues with the accuracy of these tests. He noted that the EU approval for panitumumab specified that KRAS be tested with a “standardized test” to avoid the use of potentially non-standardized tests done in research laboratories. [Note that currently there is no US FDA approved KRAS test and the US approved indications for panitumumab and cetuximab do not yet require it.]
Dr Sabine Tejpar divides the issue of KRAS into the following 3 parts: (1) KRAS as a predictor of response to EGFR inhibitors, (2) KRAS as a predictor of response to chemotherapy, and (3) KRAS as a general prognostic biomarker.
1. With regard to KRAS as an EGFR-inhibitor response predictor, Dr Tejpar said that the data do indeed show a predictive effect connecting the wild type KRAS genotype with response. According to her, these very good data have led to a paradigm shift so that as a group patients with mutated KRAS will not be treated with EGFR-targeted therapies. Dr Tejpar observed that there might be some mutant KRAS subtypes that might benefit from anti-EGFR therapy, but there will probably be very little research interest in investigating these potential subgroups. On the other hand there might be quite a bit of interest to find new therapies for the approximately 40% of patients with metastatic CRC that has mutant KRAS and whose disease has progressed despite first-line treatment.
2. Dr Tejpar said that data from the CRYSTAL study at ASCO 2008 might help evaluate the question of whether KRAS has value as a general predictor for response to chemotherapy-alone (without an EGFR inhibitor).
3. In terms of KRAS as a general prognostic marker, Dr Tejpar said that the data from Ince et al 24 (discussed above) do indeed show a small but concrete effect, in that patients with KRAS mutations did worse than those with wild type KRAS. As yet to be published results from KRAS biomarker analysis of the PETCC 3-EORTC study indicate that KRAS has no impact in predicting relapse in the adjuvant setting.
The weight of the current data on KRAS mutational status and its impact on response to EGFR inhibitors is impressive; however, as pointed out by Dr Bertagnolli all the data are from retrospective and correlative work, not from prospective studies.
EGFR Ligands and Response to EGFR inhibitors
EGFR Ligands and Response to Cetuximab Therapy
In addition to the KRAS mutation analysis discussed above, Khambata-Ford et al also performed transcriptional profiling on RNA samples obtained from fresh-frozen samples of mandatory pretreatment metastatic biopsies in patients with CRC enrolled in a cetuximab monotherapy trial.16 The purpose of this expression profile analysis was to identify genes whose expression correlated with improved clinical response. The authors found 129 candidate RNAs markers differentially expressed between 25 patients with disease control (CR, PR, or SD) and 55 nonresponders included in this analysis. They narrowed down the field to the following top 3 candidates with the lowest P values: ecto-5'-nucleotidase (CD73; an enzyme involved in purine metabolism that may have prognostic value in CRC and pancreatic cancer) and the EGFR ligands epiregulin, and amphiregulin. Metastases from patients with disease control had higher levels of CD73, epiregulin, and amphiregulin RNA expression compared with nonresponders to cetuximab therapy. When the authors compared median PFS in patients with high and low epiregulin and amphiregulin expression (using the median signal intensity as the cutoff), they found a significant association between high expression and longer PFS (Figure 4). This correlation persisted when patient age and ECOG performance status were included in the analysis as covariates (epiregulin P = .007, HR = 0.51; amphiregulin P = .0009, HR = 0.43). Similar statistically significant results were observed when the authors examined the relationship between ligand RNA expression levels in the disease control group compared with the group that did not respond to cetuximab treatment (epiregulin, P = .000015; amphiregulin P = .000025).
As mentioned above, Tejpar et al reported their analysis of KRAS mutation state as well as amphiregulin and epiregulin mRNA expression in patients treated with cetuximab plus irinotecan for metastatic CRC.21 These investigators isolated RNA and DNA for analysis from biopsies of the primary tumors preserved in standard archival formalin-fixed paraffin-embedded blocks. Amphiregulin and epiregulin mRNA expression was assessed by quantitative reverse-transcription (RT)-PCR methods. The authors observed a significant correlation between wild-type KRAS genotype and high ligand expression (P < .0001 for amphiregulin and P = .035 for epiregulin). There was a positive association between high amphiregulin expression and longer median OS in all patients and in those with wild-type KRAS tumor genotype (Figure 5). Similar results were also seen with epiregulin.
Spotlight Commentary on EGFR Ligands and Response to EGFR Inhibitors
Dr Tejpar said that the EGFR ligand epiregulin and amphiregulin are useful surrogate biomarkers for tumors that depend on EGFR for daily survival. She pointed out that they are only helpful with KRAS wild-type because these ligands act directly on EGFR. According to Dr Tejpar, within the KRAS wild-type the presence of these ligands is indicative of approximately a 60% chance of response to EGFR inhibitors. Not enough is known about the dynamics of these intracellular signal transduction systems to determine how the non-responding tumors achieve independence from the KRAS pathway.
Dr Tejpar said that these studies of fresh frozen tissue sections from liver metastases and archival paraffin-embedded samples from primary tumors found a high correlation between ligand expression in the primary tumor and the liver metastases. She noted that the sensitivity and reliability of the assay was a little better in the metastatic tissue than the primary, probably because the primary tissue is more easily contaminated with surrounding normal tissue. These studies showed that it is possible to do reliable RNA expression profiling with mRNA purified from formalin-fixed, paraffin-embedded tissue.
In Dr Tejpar's opinion, the results need to be validated with independent series to establish prospectively their predictive power before they can be applied to the clinic.
EGFR Expression as a Predictor of Response to EGFR Inhibitors
EGFR Expression Measured by IHC
While both the cetuximab and panitumumab FDA approved indications in colorectal cancers are for the treatment of “EGFR-expressing” metastatic CRC,26,27 mounting evidence indicates that EGFR expression as determined by IHC is not a useful predictor of response to EGFR inhibitors. In various studies, the degree of EGFR expression by IHC had no significant correlation with clinical response in patients treated with cetuximab15,28–30 and cetuximab has activity in patients with tumors that do not express EGFR by IHC.31,32 Similarly, panitumumab has been shown to also have anti-tumor activity in patients with low or undetectable EGFR expression as measured by IHC.33 Given these data it is not surprising that the National Comprehensive Cancer Network (NCCN) 2008 guidelines for colon cancer do not recommend routine EGFR testing to help determine whether or not to use cetuximab or panitumumab therapy.2
EGFR Gene Copy Number Measured by FISH
Investigators determined to pursue the seemingly intuitive connection between EGFR expression and response to EGFR inhibitors are using fluorescence in situ hybridization (FISH) to assess EGFR gene copy number from paraffin-embedded samples. Three groups of investigators have uncovered tantalizing evidence in retrospective studies suggesting a link between EGFR copy number and response to anti-EGFR treatment. The different groups are exploring various methodologies to quantify EGFR copy number and determine cutoff values for comparisons.
Moroni et al screened tumors from 31 patients with metastatic CRC who have either an objective response (n = 10) or SD or PD (n = 21) after treatment with either cetuximab or panitumumab.22 Eight of 9 assessable responding patients had an increased EGFR copy number by FISH compared with only 1 of 21 non-responders (P < .0001). In supportive in vitro experiments using CRC cell lines, the concentration of cetuximab that completely inhibited proliferation of cells with amplified EGFR copy number had no effect on proliferation of cells with non-amplified EGFR copy number. More recently, the same group of investigators reported the results of an exploratory analysis in patients with metastatic CRC refractory to standard therapy treated with panitumumab plus BSC (n = 58) versus BSC alone (n = 34).34 Patients were selected on the basis of tumor sample availability and adequacy for FISH testing from a larger pool of patients enrolled in a phase III trial. These authors found that among patients treated with panitumumab, those with EGFR gene copy number below a certain value (ie, 2.5/nucleus or < 40% of tumor cells expressing chromosome 7 polysomy) predicted for a shorter PFS and OS, and for a lack of response compared with patients with values above these cutoff limits. These cutoff parameters showed no effect on survival or response in patients who were treated with BSC alone.
Cappuzzo et al retrospectively analyzed EGFR copy number by FISH in samples from 85 patients with chemotherapy-refractory CRC treated with cetuximab, with the primary objective of identifying the EGFR FISH score that best associated with response rate (RR).35 These authors reported that a mean 2.92 EGFR gene copies/cell was the cutoff value that best discriminated between the responders and nonresponders to cetuximab. With this cutoff they obtained a sensitivity of 58.6% (95% CI, 47.1–70.1) and a specificity of 93.3% (95% CI, 80.6–100). With these parameters, the 43 patients with EGFR FISH-positive tumors had a significantly higher RR (P = .0001) and longer time to disease progression (P = .02) compared with the patients with EGFR FISH-negative tumors. As noted by Cappuzzo et al,35 there is an urgent need to pursue investigation of these findings in prospective clinical trials that employ carefully standardized assays.
Personeni et al reported preliminary results of an analysis of EGFR and HER2 gene copy number by FISH in samples from primary tumors (n =55) or metastases (n = 15) of patients with metastatic CRC treated with cetuximab alone or in combination with irinotecan.36 The authors chose to examine 2 parameters, the absolute gene copy number and chromosome centromeres, and their frequencies in 100 tumor cells. After testing multiple cutoff values, they found no link between EGFR and HER2 copy numbers with objective response, time to progression, and OS. They also reported no correlation between EGFR copy number and protein expression (determined by IHC, with the positivity cutoff defined as 10% of tumor cells stained).
Spotlight Commentary on EGFR Expression and Response to EGFR Inhibitors
Dr Tejpar said that these data clearly show how important it is to keep IHC data separate from FISH results. IHC is not quantitative and not sufficiently specific because of the large number of monoclonal antibodies currently in use and the variable results they generate. Dr Tejpar pointed out that there is a need to define biological expression parameters (eg, type, pattern, cutoff values) for quality assurance and result reproducibility. Currently, regulatory approved indications for EGFR inhibitors do not specify any of these parameters.
Dr Tejpar said that while for breast cancer FISH studies of HER2/neu have demonstrated that amplified target genes are translated into proteins that can be detected by IHC, in CRC EGFR gene amplification has not been shown to correlate with increased EGFR expression. According to Dr Tejpar, the available data in CRC suggest that there is something there indicating that patients with more copies of EGFR tend to respond better with EGFR-targeted therapy. However, the predictive power of this correlation is very weak and thus the research effort is low.
According to Dr Tejpar, before FISH can be applied to clinical practice technical standards (eg, cutoff values for positivity and negativity) need to be developed. Cutoff values have been difficult to develop in CRC because of great challenges in scoring results from FISH analyses. There is very low scoring reproducibility between readers and within different areas of the same tumor. Thus, the FISH EGFR methodology in CRC is not yet ready for the clinic because of low reproducibility and reliability, and weak impact on prognosis.
Skin Toxicity as a Predictor of Response to EGFR Inhibitors
Not surprisingly given that EGFR is strongly expressed in skin and has a role in maintaining skin integrity,28 most patients treated with EGFR inhibitors experience skin reactions with various degrees of severity. Subgroup analyses of phase II and III studies with cetuximab28–30,37–40 and panitumumab41,42 have demonstrated a correlation between severity of skin toxicity and response to EGFR inhibitors in patients with metastatic CRC. A more recent preliminary report by Humblet et al of a phase III BSC ± panitumumab study noted that patients with grade 2 to 4 skin toxicity had longer median PFS and OS compared with patients with grade 1 skin toxicity.43 These authors also observed that patients who were more bothered by skin toxicity reported better overall health-related quality of life (HRQoL) and reduced CRC symptoms.
In addition to also finding an association between higher grade skin toxicity and significantly improved ORR and OS (but not PFS), the 2008 pooled analysis by Lièvre et al (discussed above) also observed no correlation between skin toxicity and KRAS mutations.15 These authors pointed out that compared with skin toxicity, KRAS mutational status appears to be a more powerful predictor of resistance to cetuximab therapy. In their analysis, objective responses were seen in no patients with KRAS mutations compared with 23 (23%) of 56 patients with grade 0 to 1 skin toxicity. These investigators noted that unlike skin toxicity to EGFR inhibitors which cannot yet be predicted, KRAS mutation status can be determined before the start of therapy to help decide on an appropriate course of treatment.
Tejpar and co-investigators recently reported interesting preliminary results from the EVEREST study, a phase I/II dose cetuximab escalation study in patients with metastatic CRC who had no or slight skin reactions to standard dose cetuximab.44 They found improved ORR in these patients following cetuximab dose escalations of up to 500 mg/m2.
According to the Cox regression model built by De Roock et al to predict survival based on their previously discussed retrospective analysis of KRAS mutational status, skin toxicity had an independent impact on ORR, PFS, and OS of patients with metastatic CRC treated with cetuximab.18 These authors noted that skin toxicity is a difficult marker to use to predict response because of the absence of toxicity criteria designed to systematically assess the effects of EGFR inhibitors and of the need to determine at which time point skin toxicity should be graded to have the highest predictive value.
Spotlight Commentary on Skin Toxicity
Reflecting what is evident in the literature, the general expert consensus is that skin toxicity to EGFR inhibitors is a prognostic factors independent of KRAS mutational status. Dr Tejpar added that skin rash is not a perfect predictor because some patients respond to EGFR inhibitors even if they do not develop a rash. A lot of ongoing research is focusing on identifying the molecular basis for skin toxicity to develop clinically useful biomarkers.
Microsattelite Instability and Response to Chemotherapy
Results from several primarily retrospective studies suggest that patients with CRC marked by high microsatellite instability (MSI-H) or defective mismatch repair (MMR-D) have improved survival compared with those with low or stable microsatellite instability (MSI-L/S).45–48 In a preliminary subgroup analysis of CALGB 89803, Bertagnolli et al observed that the addition of irinotecan to adjuvant 5-FU/LV may improve disease-free survival (DFS) in patients with stage III colon cancers that have MSI-H.49 Investigators analyzed 482 tumors from the 1,264 patients randomized to this study and found 75 (16%) with MSI-H. Among the patients with MSI-H tumors, those treated with adjuvant 5-FU/LV plus irinotecan had a significantly improved DFS compared with patients treated only with 5-FU/LV (P = .18). A final report of this analysis is forthcoming. An important contribution of this study is that the investigators were able to use an IHC method commonly available in clinical laboratory settings to obtain results that correlated nicely with those obtained with complex genotyping techniques. This suggests that this type of MSI analysis could be performed in more common clinical settings and not just in specialized research laboratories.
The ongoing ECOG 5202 randomized phase III trial is using MSI as a factor to stratify patients who have undergone surgery for stage II colon cancer to adjuvant treated with FOLFOX ± bevacizumab or observation alone (www.clinicaltrials.org). Investigators are stratifying patients based on disease stage (IIA or IIB) and MSI status. Patients with MSI-L/S are considered high risk and those with MSI-H are considered low risk. High risk patients are being randomized to one of the 2 chemotherapy arms; low risk patients undergo observation alone. This trial is expected to enroll over 3,500 patients.
Spotlight Commentary on Microsattelite Instability
Dr Bertagnolli said that their results with MSI-H do not have a currently relevant clinical application because 5-FU/LV monotherapy is no longer the standard for chemotherapy for patients with metastatic CRC. An important unresolved question, raised by important data reported by Ribic et al, 46 is whether 5-FU based chemotherapy is ineffective or even harmful for patients with MSI-H tumors. There are currently no data to show how the MSI-H results would apply to oxaliplatin-based chemotherapy or to chemotherapy with other agents currently used. Continued interest in pursuing this type of MSI biomarker research in patients with CRC might become overshadowed by the developing KRAS story.
Conclusions and Future Prospects
KRAS mutational status has made significant inroads into becoming a clinically useful prognostic biomarker for the management of patients with metastatic CRC. A large body of primarily retrospective data indicates a strong correlation between KRAS wild-type tumor status and clinical benefit from monotherapy with EGFR monoclonal antibodies. Data from forthcoming studies should provide insights in the potential value of this biomarker as a predictor of response to EGFR inhibitors given in combination therapy. More work is needed to standardize the methodology necessary for improving the reliability, predictability, and reproducibility of test results for this and other emerging biomarkers. The clinical usefulness of EGFR expression status determined by IHC is not supported by a large number of studies. High EGFR gene copy number by FISH appears to be weakly associated with response to EGFR inhibitors, but at a level too low to be clinically useful. Skin toxicity to EGFR inhibitors is a valid predictor of response that has unclear applications. Other areas of ongoing clinical discovery in the field of biomarkers for CRC include gene expression and predictors of response to 5-FU-based chemotherapy,50,51 circulating tumor cells,52 and UGT1A1*28 homozygosity as a potential predictor for severe toxicity.53
1. American Cancer Society: Cancer Facts & Figures 2007
2. National Comprehensive Cancer Network: Clinical Practice Guidelines in Oncology: Colon Cancer. V1.2008. National Comprehensive Cancer Network. 2008. Accessed: 11/13/2007.
3. American Cancer Society: Cancer Facts & Figures 2008
4. Ferlay J, Bray, F., Pisani, P., Parkin, D. M.: GLOBOCAN 2002: Cancer incidence, mortality and prevalence worldwide. IARC CancerBase No.5 version 2.0. 2004. Accessed: 3/28/2008.
5. Jemal A, Siegel R, Ward E, et al: Cancer statistics, 2008. CA Cancer J Clin 58:71–96, 2008
6. Yoo PS, Lopez-Soler RI, Longo WE, et al: Liver resection for metastatic colorectal cancer in the age of neoadjuvant chemotherapy and bevacizumab. Clin Colorectal Cancer 6:202–207, 2006
7. Goldberg RM, Rothenberg ML, Van CE, et al: The continuum of care: a paradigm for the management of metastatic colorectal cancer. Oncologist 12:38–50, 2007
8. Meyerhardt JA, Mayer RJ: Systemic therapy for colorectal cancer. N Engl J Med 352:476–487, 2005
9. Bandres E, Zarate R, Ramirez N, et al: Pharmacogenomics in colorectal cancer: The first step for individualized-therapy. World J Gastroenterol 13:5888–5901, 2007
10. Tejpar S: The multidisciplinary management of gastrointestinal cancer. The use of molecular markers in the diagnosis and treatment of colorectal cancer. Best Pract Res Clin Gastroenterol 21:1071–1087, 2007
11. Baselga J, Rosen N: Determinants of RASistance to Anti—Epidermal Growth Factor Receptor Agents. J Clin Oncol 2008
12. Locker GY, Hamilton S, Harris J, et al: ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 24:5313–5327, 2006
13. Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F, et al: Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res 67:2643–2648, 2007
14. Lievre A, Bachet JB, Le Corre D, et al: KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 66:3992–3995, 2006
15. Lievre A, Bachet JB, Boige V, et al: KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab. J Clin Oncol 26:374–379, 2008
16. Khambata-Ford S, Garrett CR, Meropol NJ, et al: Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol 25:3230–3237, 2007
17. Di Fiore F, Blanchard F, Charbonnier F, et al: Clinical relevance of KRAS mutation detection in metastatic colorectal cancer treated by Cetuximab plus chemotherapy. Br J Cancer 96:1166–1169, 2007
18. De Roock W, Piessevaux H, De SJ, et al: KRAS wild-type state predicts survival and is associated to early radiological response in metastatic colorectal cancer treated with cetuximab. Ann Oncol 19:508–515, 2008
19. Freeman D, Juan T, Meropol NJ et al: Association of somatic KRAS gene mutations and clinical outcome in patients (pts) with metastatic colorectal cancer (mCRC) receiving panitumumab monotherapy. Eur J Cancer Suppl 5: 239, 2007 (abstr O3014)
20. Amado RG, Wolf M, Peeters M, et al: Wild-Type KRAS Is Required for Panitumumab Efficacy in Patients With Metastatic Colorectal Cancer. J Clin Oncol 2008
21. Tejpar S, De Roock W, Biesmans B et al: High amphiregulin and epiregulin expression in KRAS wild type colorectal primaries predicts response and survival benefit after treatment with cetuximab and irinotecan for metastatic disease. ASCO Gastrointestinal Cancers Symposium 2008 (abstr 411)
22. Moroni M, Veronese S, Benvenuti S, et al: Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in colorectal cancer: a cohort study. Lancet Oncol 6:279–286, 2005
23. Hecht JR, Mitchell E, Chidiac T et al: Interim results from PACCE: Irinotecan (Iri)/bevacizumab (bev) ± panitumumab (pmab) as first-line treatment (tx) for metastatic colorectal cancer (mCRC). ASCO Gastrointestinal Cancers Symposium 2008 (abstr 279)
24. Ince WL, Jubb AM, Holden SN, et al: Association of k-ras, b-raf, and p53 status with the treatment effect of bevacizumab. J Natl Cancer Inst 97:981–989, 2005
25. Hurwitz H, Fehrenbacher L, Novotny W, et al: Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335–2342, 2004
26. Erbitux (cetuximab) product information.2006
27. Vectibix (panitumumab) product information.2006
28. Cunningham D, Humblet Y, Siena S, et al: Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 351:337–345, 2004
29. Saltz LB, Meropol NJ, Loehrer PJ, Sr., et al: Phase II trial of cetuximab in patients with refractory colorectal cancer that expresses the epidermal growth factor receptor. J Clin Oncol 22:1201–1208, 2004
30. Vallbohmer D, Zhang W, Gordon M, et al: Molecular determinants of cetuximab efficacy. J Clin Oncol 23:3536–3544, 2005
31. Chung KY, Shia J, Kemeny NE, et al: Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 23:1803–1810, 2005
32. Hebbar M, Wacrenier A, Desauw C, et al: Lack of usefulness of epidermal growth factor receptor expression determination for cetuximab therapy in patients with colorectal cancer. Anticancer Drugs 17:855–857, 2006
33. Mitchell EP, Hecht JR, Baranda J, et al: Panitumumab activity in metastatic colorectal cancer (mCRC) patients (pts) with low or negative tumor epidermal growth factor receptor (EGFr) levels: An updated analysis. J Clin Oncol (Meeting Abstracts) 25:4082–2007
34. Sartore-Bianchi A, Moroni M, Veronese S, et al: Epidermal growth factor receptor gene copy number and clinical outcome of metastatic colorectal cancer treated with panitumumab. J Clin Oncol 25:3238–3245, 2007
35. Cappuzzo F, Finocchiaro G, Rossi E, et al: EGFR FISH assay predicts for response to cetuximab in chemotherapy refractory colorectal cancer patients. Ann Oncol 2007
36. Personeni N, De Hertogh G, Storkel S, et al: Outcome prediction to cetuximab in advanced colorectal cancer: Analysis of EGFR and HER2 gene copy number by fluorescence in situ hybridization. J Clin Oncol (Meeting Abstracts) 25:10569–2007
37. Zhang W, Gordon M, Press OA, et al: Cyclin D1 and epidermal growth factor polymorphisms associated with survival in patients with advanced colorectal cancer treated with Cetuximab. Pharmacogenet Genomics 16:475–483, 2006
38. Lenz HJ, Van CE, Khambata-Ford S, et al: Multicenter phase II and translational study of cetuximab in metastatic colorectal carcinoma refractory to irinotecan, oxaliplatin, and fluoropyrimidines. J Clin Oncol 24:4914–4921, 2006
39. Bokemeyer C, Bondarenko I, Makhson A, et al: Cetuximab plus 5-FU/FA/oxaliplatin (FOLFOX-4) versus FOLFOX-4 in the first-line treatment of metastatic colorectal cancer (mCRC): OPUS, a randomized phase II study. J Clin Oncol (Meeting Abstracts) 25:4035–2007
40. Van Cutsem E, Nowacki M, Lang I, et al: Randomized phase III study of irinotecan and 5-FU/FA with or without cetuximab in the first-line treatment of patients with metastatic colorectal cancer (mCRC): The CRYSTAL trial. J Clin Oncol (Meeting Abstracts) 25:4000–2007
41. Gibson TB, Ranganathan A, Grothey A: Randomized phase III trial results of panitumumab, a fully human anti-epidermal growth factor receptor monoclonal antibody, in metastatic colorectal cancer. Clin Colorectal Cancer 6:29–31, 2006
42. Berlin J, Van Cutsem E, Peeters M, et al: Predictive value of skin toxicity severity for response to panitumumab in patients with metastatic colorectal cancer (mCRC): A pooled analysis of five clinical trials. J Clin Oncol (Meeting Abstracts) 25:4134–2007
43. Humblet Y, Peeters M, Siena S, et al: Association of skin toxicity (ST) severity with clinical outcomes and health-related quality of life (HRQoL) with panitunumab (Pmab). J Clin Oncol (Meeting Abstracts) 25:4038–2007
44. Tejpar S, Peeters M, Humblet Y, et al: Phase I/II study of cetuximab dose-escalation in patients with metastatic colorectal cancer (mCRC) with no or slight skin reactions on cetuximab standard dose treatment (EVEREST): Pharmacokinetic (PK), Pharmacodynamic (PD) and efficacy data. J Clin Oncol (Meeting Abstracts) 25:4037–2007
45. Popat S, Hubner R, Houlston RS: Systematic review of microsatellite instability and colorectal cancer prognosis. J Clin Oncol 23:609–618, 2005
46. Ribic CM, Sargent DJ, Moore MJ, et al: Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer. N Engl J Med 349:247–257, 2003
47. Malesci A, Laghi L, Bianchi P, et al: Reduced likelihood of metastases in patients with microsatellite-unstable colorectal cancer. Clin Cancer Res 13:3831–3839, 2007
48. Gryfe R, Kim H, Hsieh ET, et al: Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 342:69–77, 2000
49. Bertagnolli MM, Compton CC, Niedzwiecki D et al: Microsatellite instability predicts improved response to adjuvant therapy with irinotecan, 5-fluorouracil and leucovorin in stage III colon cancer. J Clin Oncol (Meeting Abstracts) 24: 2006 (abstr 10003)
50. Del Rio M., Molina F, Bascoul-Mollevi C, et al: Gene expression signature in advanced colorectal cancer patients select drugs and response for the use of leucovorin, fluorouracil, and irinotecan. J Clin Oncol 25:773–780, 2007
51. Schwab M, Zanger UM, Marx C, et al: Role of Genetic and Nongenetic Factors for Fluorouracil Treatment-Related Severe Toxicity: A Prospective Clinical Trial by the German 5-FU Toxicity Study Group. J Clin Oncol 2008
52. Meropol NJ, Cohen SJ, Iannotti N, et al: Circulating tumor cells (CTC) predict progression free (PFS) and overall survival (OS) in patients with metastatic colorectal cancer. J Clin Oncol (Meeting Abstracts) 25:4010–2007
53. Roth AD, Yan P, Dietrich D et al: Does UGT1A1*28 homozygosity predict for severe toxicity in patients treated with 5-fluorouracil (5-FU)-irinotecan (IRI)? Results of the PETACC 3-EORTC 40993-SAKK 60/00 trial comparing IRI/5-FU/folinic acid (FA) to 5-FU/FA in stage II-III colon cancer. ASCO Gastrointestinal Cancers Symposium 2008 (abstr 277)
54. Sobrero AF, Maurel J, Fehrenbacher L, et al: EPIC: Phase III Trial of Cetuximab Plus Irinotecan After Fluoropyrimidine and Oxaliplatin Failure in Patients With Metastatic Colorectal Cancer. J Clin Oncol 2008
55. Tol J, Koopman M, Rodenburg CJ, et al: A randomised phase III study on capecitabine, oxaliplatin and bevacizumab with or without cetuximab in first-line advanced colorectal cancer, the CAIRO2 study of the Dutch Colorectal Cancer Group (DCCG). An interim analysis of toxicity. Ann Oncol 19:734–738, 2008
CME Quiz/Evaluation Assessment Questionnaire
To earn CME credit after reading the article, complete this quiz, answering at least four of the five questions correctly. Make a photocopy of the completed answer form and send the original to: Lippincott Continuing Medical Education, Inc., (LCMEI), 770 Township Line Road, Suite 300, Yardley, PA 19067. Only the first entry will be considered for credit and must be received by May 31, 2009. Acknowledgment will be sent within 8 weeks of participation.
Please choose the one best answer for each question.
1. New studies have reported that the presence of KRAS mutations in patients with metastatic CRC predicts
a. response to monotherapy with epidermal growth factor receptor (EGFR)-inhibitors
b. resistance to combination therapy with vascular endothelial growth factor (VEGF) inhibitors and chemotherapy
c. response to combination therapy with EGFR inhibitors
d. resistance to monotherapy with EGFR inhibitors
2. Retrospective studies of biopsy samples from patients with metastatic CRC treated with cetuximab have shown that high expression levels of the EGFR ligands, epiregulin and amphiregulin, correlate with
a. response to therapy regardless of KRAS gene status
b. longer progression-free survival (PFS) and overall survival (OS) in patients with wild-type KRAS
c. more injection site reactions in patients with mutant KRAS
d. shorter PFS and OS in patients with wild-type KRAS
3. The absence of a skin rash during therapy with an EGFR inhibitor in a patient with metastatic colorectal cancer indicates that the patient will definitely not respond to treatment with this inhibitor.
4. Data from studies of colorectal cancer biopsies indicate that there is a highly significant direct correlation between EGFR expression assessed by immunohistochemistry and amplified EGFR gene copy number measured by fluorescence in situ hybridization.
5. Studies in patients with stage III colon cancer that link high levels of microsatellite instability (MSI-H) with improved survival with the addition of irinotecan to 5-fluorouracil (5-FU)/leucovorin adjuvant therapy
a. are definitive studies that impact current clinical practice
b. have been replicated in the metastatic setting with similar results
c. have no currently relevant clinical application
d. indicate that 5-FU-based therapy is beneficial to patients with MSI-H tumors
© 2008 Lippincott Williams & Wilkins, Inc.
More on ONCOLOGY-TIMES.com...