Annals of Oncology Advance Access originally published online on September 17, 2007
Annals of Oncology 2008 19(1):185-190; doi:10.1093/annonc/mdm444
clinical trials |
Markers in the epidermal growth factor receptor pathway and skin toxicity during erlotinib treatment
1 Medical Oncology Branch, Center for Cancer Research
2 Biostatistics and Data Management Section, Center for Cancer Research
3 National Clinical Target Validation Laboratory, Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, MD, USA
* Correspondence to: Dr S. X. Yang, National Clinical Target Validation Laboratory, Division of Cancer Treatment and Diagnosis, National Cancer Institute, 37 Convent Drive, Building 37, Room 1048A, Bethesda, MD, USA. Fax: +1-301-480-4679; E-mail: xy32m{at}nih.gov
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Background: Skin toxicity is a common adverse effect of erlotinib and other anti-epidermal growth factor receptor (EGFR) agents. The aim of the study was to explore the relationship between markers in the EGFR pathway and skin rash.
Patients and methods: Eighteen patients with metastatic breast cancer were treated with daily oral erlotinib at 150 mg. Skin biopsies were obtained at baseline and after 1 month of treatment in 15 patients. EGFR, phosphorylated EGFR (pEGFR), phosphorylated mitogen-activated protein kinase (pMAPK), and phosphorylated Akt (pAkt) or Ki67 were examined quantitatively by immunohistochemistry.
Results: 11 of 18 (61%, 95% confidence interval 35.7% to 82.7%) patients developed skin rash. pAkt at baseline was significantly higher in patients with no rash than those with a grade 1 or 2 rash (18.8 ± 8.3 versus 2.4 ± 1.2 versus 3.3 ± 3.3; P = 0.0017 for trend). There was a trend towards a significant increase of pMAPK in skin posttreatment with increasing grade of rash (no rash versus grade 1 versus grade 2 rash: 4.5 ± 2.3 versus 8.4 ± 4.2 versus 19.4 ± 4.6; P = 0.036). Other markers were not associated with rash.
Conclusions: pAkt was significantly associated with not developing a rash and may have a predictive utility for skin toxicity in patients treated with erlotinib and possibly with other anti-EGFR agents.
Key words: EGFR pathway markers, erlotinib, skin toxicity
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Erlotinib is an oral small molecule inhibitor of epidermal growth factor receptor (EGFR) tyrosine kinase (TK) [1] with activity in non-small-cell lung cancer [2]. Skin toxicity is a frequently seen side-effect of anti-EGFR agents with an incidence of 47% to 100% [3]. The dermatologic reactions from anti-EGFR agents, which include antibodies against the extracellular ligand-binding domain of the receptor and small molecules that inhibit activation of the EGFR-TK [2,4,5,6], are commonly found in sites where EGFR is expressed, such as the basal epidermal keratinocytes of epidermis, sebaceous glands, and hair follicles [7,8,9]. Histopathological findings of the skin lesion reveal folliculitis and perifolliculitis with a diffuse neutrophilic infiltrate in the dermis [10]. It has been speculated that cutaneous toxicity from anti-EGFR therapy may be a result of an inflammatory response secondary to EGFR inhibition and/or decreased keratinocyte proliferation/maturation. However, the underlying mechanisms of rash development are still largely unknown, and there are no predictive markers available.
In our trial, 18 patients with metastatic breast cancer were treated with an oral daily dose of erlotinib at 150 mg until disease progression or withdrawal from toxicity; grade 1 and 2 rash were observed in 61% (11 of 18) of patients [8]. To test the hypothesis that rash was due to changes in the EGFR signaling pathway following treatment with erlotinib, EGFR, phosphorylated EGFR (pEGFR), phosphorylated mitogen-activated protein kinase (pMAPK), phosphorylated Akt (pAkt) or Ki67 in skin pre- and posttreatment and the changes from pre- to posttreatment were examined and their relationship to skin toxicity were analyzed. In addition, gene expression profiling study was carried out to explore the potential mechanisms of skin toxicity.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
patients, skin biopsies, and pharmacokenetics
Eligibility criteria, treatment plan, clinical, and toxicity evaluation have been described previously [8]. The study was approved by the Institutional Review Board of the National Cancer Institute (NCI). All patients gave written informed consent before treatment and biopsy procedure. From January to December 2002, 18 patients with metastatic breast cancer were treated with daily oral erlotinib at 150 mg. Fifteen patients were assessable for tissue correlative studies. Toxic effects for all patients were graded by the NCI Common Toxicity Criteria, version 2.0 [11]. As for skin rash/desquamation, lesions without symptoms were classified as grade 1, lesions with symptoms (<50% body surface) as grade 2, lesions with symptoms (>50% body surface) as grade 3, and exfoliative, ulcerative or bullous erythroderma as grade 4.
Skin biopsies were obtained at baseline and a month after erlotinib treatment. The samplings were taken from the same sites as the baseline biopsies to facilitate the pharmacodynamic (marker) studies. All patients underwent 4 mm punch biopsies of unaffected areas of skin from the upper back under local anesthesia. For each of these sites, a sample was immediately fixed in formalin and subsequently embedded in paraffin.
Pharmacokinetic sampling and analysis were described previously [8]. Briefly, heparinized blood samples were obtained before erlotinib and then approximately on day 28 at 0, 1, 3, 5, 8, 12, and 24 h after erlotinib administration. Plasma erlotinib concentrations were measured using a validated liquid chromatographic assay with ultraviolet detection [12].
antibody validation, immunohistochemistry, and quantitative analyses
Validation of antibodies to pEGFR, pAkt, and pMAPK by western blotting and immunocytochemistry on the EGF-responsive MCF10A human breast epithelial cells fixed with formalin was described previously [8]. Immunohistochemical staining for EGFR, pEGFR, pMAPK, pAkt, and Ki67 in skin sections on formalin-fixed paraffin-embedded paired biopsies before and during erlotinib treatment from 15 patients was also previously described [8, 13]. These biomarkers were scored quantitatively with the assistance of the Automated Cellular Imaging System (ChromaVision, San Juan Capistrano, CA) [8].
statistical analysis
The association between EGFR, pEGFR, pMAPK, pAkt, or Ki67 in skin (pretreatment, posttreatment, and the difference from pre- to posttreatment), and increasing grade of rash (0 to 2) was determined using a Jonckheere–Terpstra test [14]. The association between the degree of skin rash and pharmacokinetic parameters, such as total area under the curve (AUC), steady state drug concentration, peak plasma concentration, or trough concentration was also evaluated using the Jonckheere–Terpstra test. All P values were two-tailed, and presented without formal adjustment for multiple comparisons. Because of the many tests carried out in this exploratory analysis, all P values <0.01 were considered statistically significant and 0.01 < P < 0.05 as trends.
gene expression profiling and analysis
Tumor biopsies from 10 patients were obtained using an 18-gauge needle (six biopsies) or punch biopsy for skin lesion (four biopsies) 1 month after erlotinib treatment. Among the 10 patients, five did not experience rash, three developed grade 1 rash, and two had grade 2 rash. Procedures for snap frozen biopsy, RNA extraction, and array hybridization have been previously described [13]. All analyses were carried out on log transformed median-normalized expression as previously described [13]. Genes that were called absent in 50% of the array samples were excluded from analysis with 8220 genes retained. Among these, the top 90% varied genes (6502 genes) were used for statistical data analysis.
The evaluation of which gene ontology categories were differentially expressed between the rash and nonrash group patients was carried out using a functional class scoring analysis as previously described [13, 15]. Functional class scoring is a more powerful method of identifying differentially expressed gene classes than the more common overrepresentation analysis or annotation of gene lists based on individually analyzed genes. The functional class scoring analysis for gene ontology classes was done using BRB-ArrayTools (available at http://linus.nci.nih.gov).
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
acneiform rash
The acneiform rash (most often acne-like) occurred a few days after administration of erlotinib and reached a maximum 2 to 3 weeks following erlotinib treatment. The characteristic rash was frequently found on the face, neck and retroauricular area, scalp, and upper trunk. Grade 1 and 2 rash were observed in 11 of 18 (61%, 95% confidence interval 35.7% to 82.7%) patients. Among the 15 patients who had skin biopsies, six did not experience a rash, five developed grade 1 rash, and four had grade 2 rash. Histopathological evaluation revealed perivascular and diffuse infiltration of neutrophilic infiltrate in the dermis (Figure 1).
|
EGFR pathway markers and skin rash
First, to examine the baseline marker activity in skin, immunohistochemical staining in normal skin sections was carried out. EGFR and pEGFR at baseline were not associated with rash; nor were those of posttreatment and the changes from baseline to posttreatment (Table 1). As shown in Table 1, the mean (±standard error of the mean) pAkt level at baseline was significantly higher in patients with no rash than those with grade 1 rash or grade 2 rash (18.8 ± 8.3 versus 2.4 ± 1.2 versus 3.3 ± 3.3, respectively; P = 0.0017 by Jonckheere–Terpstra test for overall trend). For example, Akt activity at baseline was high in a patient with no rash compared with being low in a patient with grade 2 rash (Figure 2A and B). However, there were no significant associations between the level of MAPK activity or keratinocyte proliferation at baseline in skin, and the increasing grade of rash (Table 1). These data indicate that high Akt activity at baseline was highly correlated with not developing a rash. However, after erlotinib treatment, Akt activity was no longer associated with rash (Table 1).
|
|
Second, among posttreatment markers and the marker changes from pre- to posttreatment, MAPK activity posttreatment in skin exhibited a trend towards an association with increasing grade of rash (P = 0.036). It appears that patients with grade 2 rash had much higher MAPK activity. For example, MAPK activity was low in a patient with no rash versus high in a patient with grade 2 rash (Figure 2C and D). However, EGFR, pEGFR, pEGFR/EGFR ratio (data not shown) or Ki67 at baseline, posttreatment, and the differences from baseline to posttreatment were not associated with skin toxicity.
differentially expressed GO categories between rash and nonrash groups
In order to explore the role of MAPK activity posttreatment on skin toxicity, gene expression profiling study was carried out using tumor skin and/or tumor biopsies which were available for the study. As shown in Table 2, the representative significant gene ontology (GO) classes that separated the two groups were those of oxygen and reactive oxygen species metabolism (P = 0.0004), Wnt receptor signaling pathway (P = 0.0007), MAPK phosphatase activity (P = 0.002), T-cell activation (P = 0.003) or response to oxidative stress (P = 0.004). Of 11 (out of 1217) GOs statistically significant by Kolmogorov-Smirnov (KS) statistic at P < 0.005 level, 6.1 false positive GO categories would be expected by chance corresponding to a false discovery rate of 55.4% (6.1 of 11).
|
assessment of skin rash and pharmacokinetic parameters
Pharmacokinetic parameters and skin toxicity were assessed to evaluate the relationship between skin rash and exposure to erlotinib. No statistically significant associations were found between erlotinib pharmacokinetic parameters including AUC, mean steady state concentration, peak plasma concentration, trough concentration, and the incidence of rash although patients with grade 2 rash tended to have high trough plasma drug and steady state drug concentrations (data not shown). Figure 1 showed a grade 2 rash from a patient with a high steady state of plasma erlotinib concentration (3.355 µ/ml). Histopathological analysis of the skin biopsy showed inflammatory cells infiltrating dermis.
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Active Akt, but not EGFR or pEGFR, was significantly associated with not developing a rash. This finding is consistent with the putative role of Akt on mediating survival effect. Akt has been known to provide an essential survival signal required for differentiation and stratification of primary human keratinocytes [16]. It is reported that the activation of the PI3K/Akt pathway in keratinocyte differentiation depends on the activity of the EGFR, Src families of tyrosine kinases and the engagement of E-cadherin-mediated adhesion [17]. Active Akt, which is not necessarily in agreement with EGFR or pEGFR status, was found to be negatively associated with rash. However, after erlotinib treatment, Akt activity was no longer associated with rash, indicating that erlotinib treatment may have altered EGFR/Akt signaling cascade and/or keratinocyte differentiation (Table 1).
Unexpectedly, levels of MAPK activity had a trend in association with skin toxicity after daily erlotinib administration for a month. It was unclear whether MAPK activity was not suppressed by erlotinib in grade 2 rash patients or whether MAPK activity was induced by the toxic metabolites of erlotinib. In this regard, gene expression profiling study was carried out using tumor skin and/or tumor biopsies to delineate the role of MAPK activity on skin toxicity. Of 11 GO categories that significantly differed from the rash to the nonrash patients, the oxygen and reactive oxygen species metabolism or oxidative stress GO class (KS permutation P value = 0.00043) was of interest. The oxygen and reactive oxygen species metabolism is known to mediate activation of MAPK and nuclear factor-kappaB (NF-kappaB) signaling pathways in normal human epidermal keratinocytes [18]. It was possible that daily erlotinib administration or its metabolites may have caused reactive oxygen production or ‘oxidative stress’ in patients, leading to the activation of MAPK and NF-kappaB signaling pathways. Both pathways may have contributed to the inflammatory reaction and rash development.
In conclusion, we have identified that baseline Akt activity is significantly higher in normal keratinocytes in patients who did not versus those who developed rash. Low skin Akt activity may help to identify the patients who are more likely to develop skin toxicity from erlotinib and perhaps from other anti-EGFR agents. In addition, we have found that levels of MAPK activity are linked to skin toxicity following erlotinib treatment. Therefore, to add a MAPK inhibitor to patients undergoing treatment with EGFR-TK inhibitors may help to reduce or manage rash. This observation warrants further preclinical and clinical investigation.
|
funding |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Intramural Research Program of the National Cancer Institute, National Institutes of Health.
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
We thank Dr S. Lipkowitz for his critical review and helpful discussion of this manuscript.
|
Footnotes |
|---|
Present address: The Cancer Institute of New Jersey, New Brunswick, NJ, USA 
Received for publication June 15, 2007. Accepted for publication August 10, 2007.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
1. Moyer JD, Barbacci EG, 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 (1997) 57:4838–4848.
2. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med (2005) 353:123–132.
3. Perez-Soler R, Delord JP, Halpern A, et al. HER1/EGFR inhibitor-associated rash: future directions for management and investigation outcomes from the HER1/EGFR inhibitor rash management forum. Oncologist (2005) 10:345–356.
4. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med (2004) 351:337–345.
5. Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA (2003) 290:2149–2158.
6. 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 (2006) 6:29–31.[Medline]
7. Baselga J. Skin as a surrogate tissue for pharmacodynamic end points: is it deep enough? Clin Cancer Res (2003) 9:2389–2390.
8. Tan AR, Yang X, Hewitt SM, et al. Evaluation of biologic end points and pharmacokinetics in patients with metastatic breast cancer after treatment with erlotinib, an epidermal growth factor receptor tyrosine kinase inhibitor. J Clin Oncol (2004) 22:3080–3090.
9. Hidalgo M, Siu LL, Nemunaitis J, et al. Phase I and pharmacologic study of OSI-774, an epidermal growth factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies. J Clin Oncol (2001) 19:3267–3279.
10. Segaert S, Van Cutsem E. Clinical signs, pathophysiology and management of skin toxicity during therapy with epidermal growth factor receptor inhibitors. Ann Oncol (2005) 16:1425–1433.
11. Neskovic-Konstantinovic Z, Nikolic-Vukosavljevic D, Brankovic-Magic M, et al. Expression of epidermal growth factor receptor in breast cancer, from early stages to advanced disease. J Exp Clin Cancer Res (1999) 18:347–355.[Web of Science][Medline]
12. Lepper ER, Swain SM, Tan AR, et al. Liquid-chromatographic determination of erlotinib (OSI-774), an epidermal growth factor receptor tyrosine kinase inhibitor. J Chromatogr B Analyt Technol Biomed Life Sci (2003) 796:181–188.[Web of Science][Medline]
13. Yang SX, Simon RM, Tan AR, et al. Gene expression patterns and profile changes pre- and post-erlotinib treatment in patients with metastatic breast cancer. Clin Cancer Res (2005) 11:6226–6232.
14. Hollander M, Wolfe DA. Nonparametric Statistical Methods (1999) 2nd edition. New York, NY: John Wiley and Sons, Inc. 189–269.
15. Pavlidis P, Qin J, Arango V, et al. Using the gene ontology for microarray data mining: a comparison of methods and application to age effects in human prefrontal cortex. Neurochem Res (2004) 29:1213–1222.[CrossRef][Web of Science][Medline]
16. Thrash BR, Menges CW, Pierce RH, McCance DJ. AKT1 provides an essential survival signal required for differentiation and stratification of primary human keratinocytes. J Biol Chem (2006) 281:12155–12162.
17. Calautti E, Li J, Saoncella S, et al. Phosphoinositide 3-kinase signaling to Akt promotes keratinocyte differentiation versus death. J Biol Chem (2005) 280:32856–32865.
18. Cooper KL, Liu KJ, Hudson LG. Contributions of reactive oxygen species and mitogen-activated protein kinase signaling in arsenite-stimulated hemeoxygenase-1 production. Toxicol Appl Pharmacol (2007) 218:119–127.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||