CN Mobile Logo

Search form


Identifying Predictive and Surrogate Markers of Erlotinib Antitumor Activity Other Than Rash

Identifying Predictive and Surrogate Markers of Erlotinib Antitumor Activity Other Than Rash

ABSTRACT: The identification of predictive or surrogate markers of response to HER1/epidermal growth factor receptor (EGFR) inhibitor treatment would permit selection of patients most likely to respond to such treatment. Markers could consist of tumor characteristics (eg, characteristics of the receptor or downstream signaling molecules and determinants of resistance) or host characteristics (eg, pharmacokinetic parameters and toxicities). The occurrence of rash may constitute a surrogate marker of response to erlotinib (Tarceva) treatment in patients with non–small-cell lung cancer and other cancers. The erlotinib marker identification program has been designed to identify and investigate other candidate markers by analysis of a large number of clinical samples from patients enrolled in erlotinib trials in non–small-cell lung cancer, including the phase III TALENT and TRIBUTE trials of erlotinib combined with chemotherapy and the phase III BR.21 trial of erlotinib monotherapy in advanced non–small-cell lung cancer. This program should both contribute to understanding of the molecular biology of HER1/EGFR inhibition and result in identification of potential markers that can be evaluated in the clinical setting.

Erlotinib (Tarceva) belongs to a novel class of anticancer agents that inhibit the HER1/epidermal growth factor receptor (EGFR) tyrosine kinase. This agent has shown promising activity in early-phase evaluation in patients with non-smallcell lung cancer and is now being evaluated in phase III trials. Identification of a marker that could readily be used to reliably predict response to erlotinib in non-small-cell lung cancer would permit effective targeting of treatment to those patients most likely to respond. Identification of a surrogate marker for antitumor activity would also help to optimize clinical benefit of treatment. There are numerous challenges inherent to identification of such predictive and surrogate markers. Studies of Prognostic Factors in Non-Small-Cell Lung Cancer Prognostic factors in non-smallcell lung cancer may offer insights into identification of predictors of response, particularly if the prognostic factors can serve as targets of therapy. However, prognostic factors in a disease are not necessarily mark- ers that predict response to treatment. Predictive markers can be specific to particular therapies, or can be nonspecific for different types of therapy. An obvious choice for a marker that could predict response to HER1/EGFR inhibitors is overexpression of the receptor. However, analysis of receptor characteristics by immunohistochemistry in patients with non-small-cell lung cancer has generally indicated that HER1/EGFR overexpression alone is not a strong predictor of prognosis.[1] One study in patients undergoing surgery showed that whereas high HER1/EGFR expression alone was not a strong prognostic indicator, the combination of high HER1/EGFR and high HER2 expression was predictive of poorer survival and the combination of low HER1/EGFR and low HER2 expression was predictive of improved survival; these results need to be confirmed in larger groups of patients.[2] Other studies using immunohistochemistry have suggested that HER1/EGFR expression is a strong predictor of prognosis in head and neck, ovarian, cervical, and bladder cancers, and a moderate predictor in gastric, breast, colorectal, and endometrial cancers.[1] It should be noted, however, that the effort to evaluate the prognostic utility of measuring receptor expression is confounded by problems with accurately assessing receptor expression and the absence of a standardized scoring system for levels of expression. More important from the vantage point of desiring a predictive marker for response to treatment, to date there is no evidence of a correlation between level of receptor expression and clinical response with any receptor inhibitor. The complexity of identifying genetic prognostic features that might serve as predictive markers is highlighted by two recent studies. Wigle et al analyzed mRNA of 3,000 genes using microarray techniques to identify markers of non-small-cell lung cancer recurrence after surgery, and found 16 genes that were prognostic indicators[3]; 15 of these, including both genes that were overexpressed and those that were underexpressed, demonstrated statistically significant differences in expression level between good and bad prognosis groups. The molecular prognostic factors did not correlate with histologic subtype, and prognosis was independent HER1/EGFR expression. In another study by Sugita et al utilizing tissue from metastatic non-small-cell lung cancer, microarray analysis of 3,000 genes yielded 20 genes (all testis-like or melanoma-like) that discriminated between good and bad prognosis groups[4]; none of these genes was identical to the 15 identified by Wigle et al. Considerations in Clinical Evaluation of Targeted Therapies There is no universally recognized measurable biologic predictor or marker of response to conventional chemotherapy in non-small-cell lung cancer. Selection of regimens has been based on potency as indicated by objective response rates. It has been demonstrated that combination therapy is superior to monotherapy in terms of survival advantage, but no combination has been shown to be superior to another among accepted therapies. Against this background, a number of methodologic and statistical issues plague evaluation of targeted therapies in the clinical trial setting. Are targeted therapies applicable to a whole population of patients with the same histologically heterogeneous histology? Is it possible to demonstrate efficacy in the conventional clinical trial setting without selecting patients based on the presence of a specific therapeutic target? Is expression of a specific target sufficient to warrant specific targeted therapy? The potential impact of applying a targeted therapy to a population with histologically defined disease can be appreciated by a hypothetical example involving the HER2 inhibitor trastuzumab (Herceptin), which is active in HER2-positive breast cancer. Based on findings in registration trials in metastatic breast cancer, it can be assumed that there is no disadvantage in using trastuzumab in HER2- negative patients and that survival in HER2-negative patients is 20 months. In a hypothetical clinical trial evaluating standard chemotherapy with or without trastuzumab in 750 patients of whom 30% have HER2-positive disease, overall survival would be 21.6 months in trastuzumab-treated patients and 20 months in standard therapy patients, based on existing data on survival rates. This difference would not achieve statistical significance, leading to the conclusion that trastuzumab has no or little effectiveness in this setting. Demonstration of a statistically significant difference in survival would require a study population of approximately 2,500 patients on the assumption that 30% had HER2-positive disease. Additional difficulties in evaluating targeted therapies are evident from a real-world study of trastuzumab in patients with HER2-positive non- small-cell lung cancer.[5] In this study, cisplatin/gemcitabine (Gemzar) vs cisplatin/gemcitabine/trastuzumab produced response rates of 41% vs 36%, median times to progression of 7.2 vs 6.3 months, and median progression- free survival of 7.0 vs 6.1 months. HER2 overexpression was found in only 20% of patients screened using immunohistochemistry. HER2 gene amplification was detected in only 2% of patients screened using fluorescence in situ hybridization (FISH); tumor regression greater than 50% occurred in 5 of 6 patients with HER2 gene amplification on FISH who received trastuzumab. It can thus be appreciated that precisely what constitutes a marker predictive of response may depend both on determining what level of the marker is associated with specific risk and on the availability of techniques to accurately detect and measure that marker. Candidate Markers for HER1/EGFR Inhibitors HER1/EGFR signaling pathways constitute a highly complex system. HER1 is responsible for activation of a number of signaling pathways that lead to cell proliferation. However, there are a number of other molecules and receptors that could also be responsible for cell proliferation. Thus, HER1/EGFR tyrosine kinase activity could be part of a system in which a large number of different receptors/ molecules are responsible for exponential amplification of signaling pathways for proliferation and other cell functions, or part of a system in which activation of a small number of kinases downstream from the receptor is responsible for proliferation and other functions, perhaps through a clonal event involving the receptor molecule. The latter situation is now known to obtain in the case of HER2 gene amplification in metastatic breast cancer, increased BCR/ABL expression resulting from a translocation in chronic myelogenous leukemia, and in a small number of other settings. The essential steps in identifying potential predictive/surrogate markers for response to HER1/EGFR inhibitors are outlined in Figure 1. The initial step is to characterize the receptor and the downstream molecules activated by activation of the receptor. The ultimate test of the marker is that it accurately predicts clinical response to treatment and provides a ready means for preselecting patients and optimizing treatment benefit. Markers may consist of (1) tumor characteristics- eg, characteristics of the target receptor or downstream signaling molecules or determinants of resistance to HER1/ EGFR inhibition; or (2) patient characteristics- eg, pharmacokinetics and toxicities. Tumor Characteristics
Target characteristics that could constitute potential markers include concentration or level of expression of the inactive receptor, or characteristics of the activated receptor, including binding ligands, dimerization partners, mutations in receptor or downstream molecules, and phosphorylation status of downstream kinases. With regard to potential gene alterations, the EGFRvIII variant is associated with a constitutively active (ligand-independent) tyrosine kinase, and is known to be common in some central nervous system tumors; it has been reported to be present in 15% to 30% of non- small-cell lung cancer tumors, but its frequency using high-precision methods of detection and its significance in non-small-cell lung cancer remain largely undefined. Little currently is known about other HER1/EGFR mutations or polymorphisms that might affect response to inhibitors. Erlotinib inhibits phosphorylation of HER1/EGFR in a dose-dependent manner.[6] Recent investigations have provided evidence that erlotinib treatment results in dephosphorylation of important downstream kinases ERK and p70s6K (Figure 2), indicating that HER1/EGFR tyrosine kinase inhibition is accompanied by reduced activation of signaling pathways.[7] It remains, unclear, however, to what degree this inhibition is associated with clinical outcome in non-small-cell lung cancer. Determinants of acquired resistance to HER1/EGFR inhibitors could be used to predict response to treatment or guide dose modifications. In an ongoing study, HN5 head and neck cancer cells were successively passaged in increasing concentrations of erlotinib, resulting in increased erlotinib 50% inhibitory concentrations.[ 8] This decrease in sus- ceptibility was accompanied by downregulation of HER1, HER2, and HER3 receptors and by increased levels of phosphorylated Akt kinase, indicating the activity of other potentially important signaling pathways distinct from HER1/EGFR. These findings thus far indicate that acquired resistance to erlotinib in vitro is a progressive phenomenon associated with downregulation of the HER family of receptors (possibly including selection of rare cells with no/low HER expression) and upregulation of other genes involved in signaling. Translation of such findings into implications for the in vivo setting requires further study. Other recent data indicate the possibility of innate resistance to HER1/ EGFR inhibitors. Screening non- small-cell lung cancer patients for treatment with the HER1/EGFR inhibitor gefitinib (Iressa) revealed a number of genetic alterations and resulted in identification of GRG- (gefitinib resistance gene-1) as a candidate gene associated with gefitinib resistance[9]; the gene encodes a cellsurface protein closely associated with HER2. Initial analysis indicated that GRG- expression was inversely associated with clinical response to treatment (P = .04). Clinical testing of this and other candidate resistance genes is ongoing. Patient Characteristics
As noted, patient characteristics that may serve as predictive or surrogate markers include toxicities and host pharmacokinetic characteristics. As reviewed by Prez-Soler elsewhere in this supplement, there are accumulating data that rash may serve as a surrogate marker for effectiveness of erlotinib and other HER1/EGFR inhibitors and could prove to be useful in guiding drug dosage. With regard to pharmacokinetic parameters, we have recently performed a study assessing the relationship between plasma erlotinib steadystate concentrations and survival in head and neck cancer patients from a phase II trial of erlotinib.[10] When hazard ratios of survival are analyzed according to pharmacokinetic results, the median plasma concentrations of both erlotinib and OSI-420 at the 5- 10 hour postdose window (hazard ratio 1.054, P = .021 and hazard ratio 1.422, P = .005, respectively), as well as those of OSI-420 at trough window (20-25 hours) (hazard ratio 1.387, P = .0014), predict for improved overall survival. Analysis of rash in this study population also showed a significant predictive effect of rash for prolonged survival. We did not find any correlation between erlotinib plasma concentrations at steady state and presence of rash, although this latter finding may be due to the relatively small number of patients included in the analysis. Other potential pharmacokinetic variables that require investigation to ascertain potential effects on response to HER1/EGFR inhibitor treatment include cytochrome P450 polymorphisms, particularly involving the 3A4 and 4A5 isoenzymes, and potential pharmacokinetic interactions with chemotherapy or other concurrently administered medications. Overall, more specific and powerful correlative analyses for pharmacokinetic factors and adverse effects are needed to identify potential predictive/ surrogate markers. Erlotinib Marker Identification Program The erlotinib marker identification program has been designed to investigate the potential utility of a large number of candidate markers in non- small-cell lung cancer (Figure 3). To this end, large numbers of clinical samples are being obtained from patients enrolled in the phase III TALENT and TRIBUTE trials of erlotinib combined with chemotherapy as firstline treatment of advanced non-smallcell lung cancer, the phase III BR.21 trial of erlotinib monotherapy in advanced, refractory non-small-cell lung cancer, and a phase II pharmacodynamics study in which biopsy samples are being obtained prior to and during erlotinib therapy with the primary intent of assessing effects of treatment on signaling pathways. Candidate marker analysis is to be performed using standard and novel techniques, including reverse transcriptase polymerase chain reaction and tissue microarray construction for highthroughput screening. It is hoped that this program will result in discovery and validation of predictive/surrogate markers that will permit identification of patients most likely to benefit from erlotinib treatment. Conclusion Numerous opportunities exist to identify candidate predictive and surrogate markers for response to erlotinib and other HER1/EGFR inhibitors, and development of novel methods for marker analysis is ongoing. Further research is necessary to more fully understand the components and interactions of the HER signaling pathways. Analysis of a large number of clinical samples from non-smallcell lung cancer patients in erlotinib trials in the erlotinib marker identifi cation program will permit advances in understanding the molecular biology of response and nonresponse to HER1/EGFR inhibition and may result in identification of markers that will allow optimal benefit to be gained from treatment with erlotinib and other targeted agents.


The author(s) have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.


1. Nicholson RI, Gee JM, Harper ME: EGFR and cancer prognosis. Eur J Cancer 37(suppl 4):S9-S15, 2001.
2. Sridhar SS, Seymour L, Shepherd FA: Inhibitors of epidermal-growth-factor receptors: A review of clinical research with a focus on non-small-cell lung cancer. Lancet Oncol 4:397-406, 2003.
4:397-406, 2003. 3. Wigle DA, Jurisica I, Radulovich N, et al: Molecular profiling of non-small cell lung cancer and correlation with disease-free survival. Cancer Res 62:3005-3008, 2002.
4. Sugita M, Geraci M, Gao B, et al: Combined use of oligonucleotide and tissue microarrays identifies cancer/testis antigens as biomarkers in lung carcinoma. Cancer Res 62:3971-3979, 2002.
5. Gatzemeier U, Groth G, Hirsh V, et al: Gemcitabine/cisplatin alone and with trastuzumab (Herceptin) in patients with nonsmall cell lung cancer overexpressing HER2: Results of a randomized phase II study (abstract 1185). Proc Am Soc Clin Oncol 21:297A, 2002.
6. Moyer JD, Barbacci ED, Iwata KK, et al: Induction of apoptosis and cell cycle arrest by CP-358,774, an inhibitor of epidermal growth factor receptor tyrosine kinase. Cancer Res 57:4838-4848, 1997.
7. Hidalgo M, Malik S, Rowinsky E, et al: Inhibition of the epidermal growth factor receptor (EGFR) by OSI-774, a specific EGFR inhibitor in malignant and normal tissues of cancer patients (abstract 281). Proc Am Soc Clin Oncol 20:71a, 2001.
8. Pérez-Soler R, Ling Y-H, Lia M, et al: Molecular mechanisms of resistance to the HER1/EGFR tyrosine kinase inhibitor erlotinib HCI in human cell lines (abstract 762). Proc Am Soc Clin Oncol 22:190, 2003.
9. Natale RB, Shak S, Aronson N, et al: Quantitative gene expression in non-small cell lung cancer from paraffin-embedded tissue specimens: Predicting response to gefitinib, an EGFR kinase inhibitor (abstract 763). Proc Am Soc Clin Oncol 22:190, 2003.
10. Soulières D, et al: J Clin Oncol (in press).

By clicking Accept, you agree to become a member of the UBM Medica Community.