Expression of epidermal growth factor receptor and human epidermal growth factor receptor 2 in periocular squamous cell carcinomas of horses Diane V. H. Hendrix, DVM, and Kim M. Newkirk, DVM, PhD
Objective—To determine whether epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2) are expressed in periocular squamous cell carcinomas (SCCs) of horses. Sample—Biopsy specimens of SCCs from 46 horses. Procedures—Pathology records were searched retrospectively for biopsy specimens of periocular SCCs obtained from horses. Slides of the specimens were reviewed histologically to confirm the SCC diagnosis and stained for EGFR and HER2 by immunohistochemical methods. For both EGFR and HER2, the immunohistochemical staining intensity and percentage of stain-positive cells on the slides were determined. Results—43 of 46 (93%) SCCs were immunoreactive for EGFR. The median score for EGFR staining intensity was 4 (range, 2 to 12), and the median number of mitotic figures was 8 mitotic figures/10 hpfs (range, 0 to 34 mitotic figures/10 hpfs). Mitotic index was not correlated with the percentage of EGFR stain–positive cells or staining intensity. Of the 43 EGFR-immunoreactive SCCs, 38 had stain present primarily in the cytoplasm and 5 had stain equally distributed between the cytoplasm and cell membranes. Thirty-five of 46 (76%) SCCs were immunoreactive for HER2. Mitotic index was not correlated with the percentage of HER2 stain–positive cells or staining intensity. Of the 35 HER2-immunoreactive SCCs, the stain was present primarily in the cytoplasm and 7 had stain equally distributed between the cytoplasm and cell membranes. Conclusions and Clinical Relevance—Results indicated that most periocular SCCs of horses expressed EGFR and HER2, which suggested that equine periocular SCCs might respond to treatment with EGFR inhibitors. (Am J Vet Res 2014;75:912–917)
S
quamous cell carcinomas are the most common tumors of the eyes and adnexa of horses and commonly affect the eyelids, third eyelid (nictitating membrane), conjunctiva, and limbus.1,2 Breeds most frequently affected by SCC include Appaloosa, American Paint Horse, and draft breeds, and the prevalence of SCC is positively associated with age.1 In horses, development of SCC is thought to be related to an increased susceptibility to ultraviolet radiation carcinogenesis because all periocular SCCs express p53, lightly pigmented horses are predisposed to the development of periocular SCC, and the prevalence of SCCs is positively associated with altitude, longitude, and mean annual solar radiation.1,3 When left untreated, periocular SCC can invade the local soft tissues, bony orbit, sinuses, and brain and Received April 23, 2014. Accepted June 23, 2014. From the Departments of Small Animal Clinical Sciences (Hendrix) and Biomedical and Diagnostic Sciences (Newkirk), College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996. Supported by the Morris Animal Foundation. Presented in abstract form at the 44th Annual Conference of the American College of Veterinary Ophthalmologists, Rio Grande, Puerto Rico, November 2013. The authors thank Misty Bailey for technical editing. Address correspondence to Dr. Hendrix ([email protected]). 912
14-04-0115r.indd 912
EGFR HER SCC TBST
ABBREVIATIONS
Epidermal growth factor receptor Human epidermal growth factor receptor Squamous cell carcinoma Tris-buffered saline solution with Tween 20
metastasize to regional lymph nodes, salivary glands, and thorax. Following treatment, SCC recurrence most commonly involves the eyelid or orbit.4 Treatment of periocular SCC in horses is dependent on tumor location, tumor size, extent of tissue invasion, vision status, intended purpose of the affected horse, equipment availability, and financial constraints. Generally, the treatment of choice is surgical excision of the tumor followed by adjunctive treatment such as cryotherapy, iridium-192, strontium-90, intralesional cisplatin, or 5-fluorouracil.2 Although many treatment modalities are used for the treatment of periocular SCCs in horses, few of them are available to general practitioners, and most are associated with some risk for the personnel administer the treatment. In human medicine, there has been a tremendous amount of preclinical and clinical research to evaluate the role of EGFRs in tumor growth and targeted cancer treatment. Four types of EGFRs (EGFR, HER2, HER3, AJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:53 AM
and HER4) have been identified, of which EGFR and HER2 have been studied most frequently. Most healthy epithelial tissues express EGFR and HER2, which are instrumental for proliferation, regeneration, differentiation, and development of epithelial cells. Increased expression of EGFR and HER2 from basal levels has been associated with carcinogenesis and tumor progression.5 Consequently, a new wave of pharmaceutical development has focused on the creation of treatments that inhibit EGFR. In human patients with certain tumors, the type of EGFR expressed by the tumor has prognostic value and is used to tailor treatment because some drugs selectively inhibit some receptors but not others.6 In a study7 of 19 cats with cutaneous SCCs, 14 had tumors that tested positive for EGFR expression and SCC expression of EGFR was predictive of a poor prognosis. Expression of EGFR has also been identified in oral SCCs of cats,8 malignant epithelial nasal tumors of dogs,9 and SCCs and other epithelial tumors of humans.10,11 Human epidermal growth factor receptor 2 has been most commonly studied in human patients with breast or colorectal cancer, and testing for HER2 immunoreactivity has been used extensively for targeted treatment with EGFR-inhibiting drugs.12–15 Because healthy epithelial cells also have EGFRs, use of EGFR inhibitors is commonly associated with the development of rashes and occasional adverse effects associated with ocular tissues such as dysfunctional tear syndrome, blepharitis, and eyelash changes.16,17 The detection of EGFR or HER2 expression in periocular SCCs of horses would indicate that preclinical trials to investigate the effectiveness of EGFR inhibitors for targeted treatment of such tumors are warranted. Although EGFR inhibitors are generally administered IV in human patients because of concerns about actual or potential metastatic disease, it might be possible to administer EGFR inhibitors intralesionally in periocular SCCs of horses because of their low rate of metastasis. Epidermal growth factor receptor inhibitors are expensive. Theoretically, for horses with periocular SCCs, intralesional injection of EGFR inhibitors would greatly decrease the cost of treatment and prevent the adverse systemic effects associated with those drugs. Of course, investigation of EGFR inhibitors in cell cultures would be necessary to determine the optimal drug concentration for cytotoxicity. However, because of the targeted nature of EGFR inhibitors, it is first necessary to determine whether periocular SCCs of horses express EGFR or HER2 before preclinical or clinical trials are initiated. Also, elucidation of whether the target EGFR proteins are located on the cell membrane or in the cytoplasm will assist investigators in choosing the drug most likely to be effective for those trials. The purpose of the study reported here was to determine whether EGFR and HER2 are expressed by periocular SCCs of horses. Materials and Methods Specimens—University of Tennessee Veterinary Medical Center pathology records from 2001 through 2011 were searched for biopsy specimens of periocular SCCs obtained from horses. Specimens were selected for the study on the basis of the reported histologic AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 913
diagnosis, tumor location, and availability of archived tumor specimens for further evaluation. The search resulted in the identification of 46 periocular SCCS that were eligible for the study, of which 14 involved the eyelid, 12 involved the nictitating membrane, 11 involved the cornea and limbus, and 9 involved the conjunctiva. Histologic slides of the SCCs were reviewed by a board-certified anatomic pathologist (KMN) to confirm the diagnosis. For horses that were treated at the University of Tennessee Veterinary Medical Center, medical records were reviewed to obtain information about the patient’s response to treatment, and referring veterinarians or owners were contacted by telephone to obtain follow-up information for horses that were not treated at the university. Additionally, an eyelid, conjunctiva, nictitating membrane, and cornea were harvested from 1 eye of each of 3 adult horses with clinically normal eyes that were euthanized for reasons unrelated to the study for use as control specimens. This study did not require oversight or approval from an institutional animal care and use committee because all SCC specimens were obtained from the pathology laboratory’s tissue archives and the control specimens were obtained after the horses were euthanized. Immunohistochemical staining—All biopsy specimens were stored in blocks of paraffin. Each specimen was sectioned into slices that were 5 µm thick, and each slice was placed on a charged microscope slide, air-dried, and then heated at 60°C for 15 minutes. The tissue specimens on each slide were deparaffinized with xylene and rehydrated through graded ethanols to deionized water in a routine manner. Two slides (1 for EGFR and 1 for HER2) from each biopsy specimen were prepared for immunohistochemical staining. For the immunohistochemical staining procedure for EGFR, the slides were rinsed with deionized water and soaked in TBST for 10 minutes. Slides were then loaded on an autostainer,a where all procedures were performed at room temperature (22°C), and slides were rinsed with TBST between each step. The slides were then exposed to a 3% hydrogen peroxide solution to block endogenous peroxidase for 5 minutes, a serumfree protein blocking solutionb for 5 minutes, mouse anti-EGFr clone:31G7c (primary antibody) at a dilution of 1:40 for 30 minutes, the secondary reagents in a horseradish peroxidase–labeled polymer systemd for 30 minutes, and chromagene for 10 minutes. The slides were removed from the autostainer, rinsed in deionized water, and stained for 5 seconds with hematoxylin.f Then, the slides were blued in ammonia water, dehydrated through ethanol, and cleared with xylene, and a coverslip was placed over the tissue specimen. For the immunohistochemical staining procedure for HER2, slides were heated with EDTA buffer (pH, 9) in a steamer at 95°C for 20 minutes, then cooled for 20 minutes for heat-induced epitope retrieval. After epitope retrieval, all slides were rinsed in deionized water and soaked in TBST for 10 minutes. Slides were then loaded on the autostainera and subsequently processed in the same manner as those that were stained with EGFR, except that the primary antibody used was mouse anti-HER2 clone:CB11g at a dilution of 1:40 for 30 minutes. 913
9/19/2014 10:55:53 AM
percentage of stain-positive cells was classified into 1 of 3 categories (1 = ≤ 40%, 2 = > 40% to 70%, or 3 = > 70%) and no reactivity score was calculated.18
Slide evaluation—For each slide, the staining intensity, location of the stain within the cells, and percentage of stain-positive cells were scored as described for EGFR7–9 and HER2.18 Briefly, for both EGFR and HER2, the staining intensity was scored on a 4-point scale (0 = no staining, 1 = low intensity, 2 = moderate intensity, or 3 = high intensity). The location of the stain within the cells was categorized as cytoplasm, cell membrane, or both.7–9,18 For EGFR, the percentage of stain-positive cells was classified into 1 of 4 categories (1 = ≤ 10%, 2 = > 10% to 30%, 3 = > 30% to 60%, or 4 = > 60%). The EGFR reactivity score was defined as the product of the intensity score and the category for the percentage of stain-positive cells, and an EGFR reactivity score ≥ 2 was considered positive.7–9 For HER2, the
Statistical analysis—The respective correlations between mitotic indices and staining intensity and percentage of stain-positive cells for both EGFR and HER2 were determined with the Pearson product moment correlation. Statistical softwareh was used for all analyses, and values of P ≤ 0.05 were considered significant. Results Specimens—Periocular SCC tissue specimens from 46 horses were evaluated. Breeds represented included American Paint Horse (n = 13), Appaloosa (9),
Table 1—Median (range) values for various measures of expression of EGFR and HER2 in periocular SCC (n = 46) and normal (control; 3) tissue specimens obtained from horses. Type of EGFR
Location of lesion
EGFR
Cornea
HER2
Specimen type
Percentage of stainpositive cells*
SCC (n = 11) Control (n = 3) Conjunctiva SCC (n = 9) Control (n = 3) Nictitating membrane SCC (n = 10) Control (n = 3) Eyelid SCC (n = 13) Control (n = 3) Cornea SCC (n = 6) Control (n = 3) Conjunctiva SCC (n = 7) Control (n = 3) Nictitating membrane SCC (n = 8) Control (n = 3) Eyelid SCC (n = 14) Control (n = 3)
4 (2–4) 1 4 (2–4) 3 3 (1–4) 2 4 (2–4) 3 3 1 3 (2–3) 3 3 (2–3) 3 3 (2–3) 3
Staining intensity†
Reactivity‡
1 (1–3) 2 1 (1–3) 2 1.5 (1–2) 2 2 (1–2) 2 2 2 2 (1–2) 2 1.5 (1–3) 2 2 (1–2) 2
4 (2–12) 2 4 (2–12) 6 3 (2–8) 4 6 (2–8) 6 — — — — — — — —
Of the 46 SCC specimens, 43 stained positive for EGFR and 35 stained positive for HER2. For all values, the lack of a range in parenthesis indicates that all specimens had the same value for that variable. *For EGFR, the percentage of stain-positive cells was classified into 1 of 4 categories (1 = ≤ 10%, 2 = > 10% to 30%, 3 = > 30% to 60%, or 4 = > 60%), whereas for HER2, the percentage of stain-positive cells was classified into 1 of 3 categories (1 = ≤ 40%, 2 = > 40% to 70%, 3 = > 70%). †For both EGFR and HER2, the staining intensity was scored on a 4-point scale (0 = no staining, 1 = low intensity, 2 = moderate intensity, or 3 = high intensity). ‡For EGFR only, the reactivity score was defined as the product of the staining intensity score and the category for the percentage of stain-positive cells, and an EGFR reactivity score ≥ 2 was considered positive. — = Not calculated. Table 2—Distribution of immunohistochemical stains for EGFR and HER2 in periocular SCC (n = 46) and clinically normal (control; 3) tissue specimens obtained from horses. Location of immunohistochemical stain in stain-positive cells (No. of samples) Type of EGFR EGFR
Location of lesion Cornea Conjunctiva Nictitating membrane Eyelid
HER2
Cornea Conjunctiva Nictitating membrane Eyelid
Specimen type
Cytoplasm only
Cell membrane only
SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control
7 0 2 0 6 0 7 0 2 3 0 3 4 3 6 3
0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Equally between cytoplasm and cell membrane 1 0 2 0 0 0 2 0 2 0 4 0 1 0 0 0
Primarily cytoplasm
Primarily cell membrane
3 0 5 0 4 3 4 0 2 0 3 0 3 0 8 0
0 0 0 3 0 0 0 3 0 0 0 0 0 0 0 0
See Table 1 for key. 914
14-04-0115r.indd 914
AJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:53 AM
Quarter Horse (7), Tennessee Walking Horse (7), and 7 other breeds (10). Of the 46 affected horses, 27 were geldings, 15 were mares, and 4 were stallions. The mean ± SD age of horses at the time the biopsy specimens were obtained was 13 ± 5.2 years. Thirty-seven of the SCC specimens were obtained from patients examined at the University of Tennessee Veterinary Medical Center; the remaining 9 specimens were obtained from outside submissions to the pathology laboratory. The clinically normal (control) tissue specimens were obtained from 2 Appaloosas and 1 Mangalarga Marchador. Immunohistochemical staining— The EGFR staining intensity and percentage of stain-positive cells were consistent among control specimens (Table 1). For the control eyelid skin specimens, the stain was located in the cell membrane to a slightly greater extent than in the cytoplasm, and the stain was most intense in the basal region and decreased in intensity through the superficial layers. For the conjunctival specimens, the stain was located primarily in the cell membranes, and its intensity varied throughout the specimens. For the nictitating membrane specimens, the stain was located in both the cell membranes and cytoplasm of cells in the basal layer, and its intensity was varied throughout the specimens. For corneal specimens, staining was limited to the cell membranes of the basal cells at the limbus (Table 2). Forty-three of 46 (93%) SCCs had an EGFR reactivity score ≥ 2 (Figure 1). Of the 3 SCCs that had a reactivity score of 0 or 1, 2 were obtained from the nictitating membrane and 1 was obtained from the eyelid. For the 43 EGFR stain–positive SCCs with a reactivity score ≥ 2, the median reactivity score was 4 (range, 2 to 12; Table 1) and the median number of mitotic figures was 8 mitotic figures/10 hpfs (range, 0 to 34 mitotic figures/10 hpfs). The mitotic index was not significantly correlated with the percentage of stain-positive cells (r = 0.03; P = 0.403) or staining intensity (r = 0.04; P = 0.523). The stain was located in the cytoplasm of all 43 EGFR stain–positive SCCs; it was located only in the cytoplasm of 22 tumors, was located primarily in the cytoplasm of 16 tumors, and equally distributed between the cytoplasm and cell membrane of 5 tumors (Table 2). For the control specimens stained for detection of HER2, the stain was located only in the cytoplasm (Table 2). In AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 915
the corneal specimens, the percentage of stain-positive cells was greatest in the basal layers, whereas in the eye-
Figure 1—Representative photomicrographs of sections of periocular SCCs of horses that were immunohistochemically stained for EGFR and depict the 4 possible scores for staining intensity. A—Photomicrograph of a section of an SCC removed from an eyelid; staining intensity, 0 (no staining). B—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 1 (low intensity). Notice that the stain is located primarily in the cytoplasm. C—Photomicrograph of a section of an SCC removed from a conjunctiva; staining intensity, 2 (moderate intensity). Notice that the stain is located primarily in the cytoplasm. D—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 3 (high intensity). Notice that the stain is again located primarily in the cytoplasm. Immunohistochemical stain; bar = 10 µm.
Figure 2—Representative photomicrographs of sections of periocular SCCs of horses that were immunohistochemically stained for HER2 and depict the 4 possible scores for staining intensity. A—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 0 (no staining). B—Photomicrograph of a section of an SCC removed from a nictitating membrane; staining intensity, 1 (low intensity). Notice that the stain is located primarily in the cytoplasm. C—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 2 (moderate intensity). Notice that the stain is equally distributed between the cytoplasm and cell membranes. D—Photomicrograph of a section of an SCC removed from a nictitating membrane; staining intensity, 3 (high intensity). Notice that the stain is again located primarily in the cytoplasm. Immunohistochemical stain; bar = 10 µm. 915
9/19/2014 10:55:55 AM
lid, conjunctiva, and nictitating membrane specimens, the percentage of stain-positive cells was greatest in the superficial layers of the epithelium. Thirty-five of 46 (76%) SCCs stained positive for HER2 (Figure 2). Of the 11 HER2 stain–negative SCCs, 2 were obtained from the conjunctiva, 5 were obtained from the cornea, and 4 were obtained from the nictitating membrane. The mitotic index was not significantly correlated with the percentage of stain-positive cells (r = –0.14; P = 0.884) or staining intensity (r = –0.16; P = 0.770). The stain was located in the cytoplasm of all 35 HER2 stain–positive SCCs; it was located only in the cytoplasm of 12 tumors, was located primarily in the cytoplasm of 16 tumors, and was equally distributed between the cytoplasm and cell membrane of 7 tumors (Table 2). Discussion Results of the present study indicated that most periocular SCCs of horses expressed EGFR and HER2, and the percentage of cells that stained positive for EGFR and HER2 in those tumors was generally greater than that in control specimens. These results, along with the fact that abnormalities in EGFR status such as overexpression, mutation, signaling errors, and dysregulated receptor trafficking have been associated with the development of SCC and other tumors,16 suggest that administration of anti-EGFR drugs might be beneficial for the treatment of periocular SCCs in horses. For the equine periocular SCCs of the present study that expressed EGFR and HER2, the immunohistochemical stain was generally located primarily in the cytoplasm, whereas for feline cutaneous SCC that expressed EGFR, the stain was generally located in the cell membrane,7,8 with only 5 of 197 and 4 of 138 tumors having faint to moderate cytoplasmic staining. Similar to cutaneous SCCs of cats, malignant epithelial nasal tumors of dogs generally express EGFR in the cell membrane.9 Immunohistochemical staining of dermal SCCs of human patients is generally heterogeneous, with the stain accumulating primarily in the cytoplasm in the less differentiated areas of the tumor,10,11 which is believed to occur because of accumulation of EGFRs in the cytoplasm owing to excessive stimulation of the cell by epidermal growth factor or epidermal growth factor α or secondary to changes in the EGFR molecule.10 The control specimens of the present study expressed HER2 in the cytoplasm but not in the cell membrane. Conversely, of the 35 SCCs that expressed HER2, 23 had expression of HER2 in the cell membrane as well as in the cytoplasm. In a study15 in which fluorescence in situ hybridization was used to determine HER2 expression by human colorectal tumors, membrane staining was associated with gene amplification of the HER2 molecule but the mechanism of cytoplasmic staining was not identified. In the present study, the percentage of HER2 stain–positive cells in SCCs that involved the cornea was greater than that in the control specimens; however, for SCCs that involved the other ocular tissues, the percentage of HER2 stain–positive cells was similar to that for the corresponding control specimens. In humans, the basal HER2 expression is low in healthy epithelial cells and is increased from bas916
14-04-0115r.indd 916
al levels in fetal tissues and neoplasms.19 Expression of HER2 is localized primarily in the superficial layers of the ocular surface epithelium of humans,20 which was also observed in the control specimens of all equine tissues evaluated in the present study except for the cornea. The finding that HER2 is expressed primarily on the surfaces of clinically normal tissues suggests that HER2 is preferentially expressed by differentiated postmitotic epithelial cells.20 Interpretation of HER2 reactivity in tissues has not been completely elucidated. Most research into HER2 reactivity has involved breast cancer in human patients. In human breast tumors, HER2 expression is increased 20% to 30% from that in healthy breast tissue,12–14 and tumors that have an overexpression of HER2 tend to be more aggressive and are associated with a worse prognosis, compared with HER2-negative tumors.12–14 Human patients treated with trastuzumab, a monoclonal antibody against the extracellular domain of HER2, after breast cancer surgery had an increased survival rate and decreased tumor recurrence rate, compared with similar patients who were not treated with trastuzumab.13,14 However, because trastuzumab affects only the extracellular domain of HER2, it has no effect on HER2 localized in the cytoplasm, a fact that has directed interpretation of HER2 immunoreactivity in breast cancer.21 Investigators of another study15 report that human patients with colorectal tumors in which HER2 is localized in the cytoplasm might have a poorer prognosis than do patients with HER2-negative tumors. Nonetheless, the relevance and prognostic value of HER2 immunoreactivity patterns in various tumors and species require further investigation. Knowledge of the staining pattern and location of the different types of EGFRs is important when administration of anti-EGFR drugs is considered. The 3 types of anti-EGFR drugs include monoclonal antibodies such as cetuximab, which targets the extracellular portion of the EGFR molecule; small-molecule tyrosine kinase inhibitors such as erlotinib and gefitinib, which target the intracellular protein kinase domain; and dual tyrosine kinase inhibitors such as lapatinib, which targets both EGFR and HER2.16 Promising results have been achieved in human patients with cutaneous SCCs treated with cetuximab as monotherapy or as a radiation-sensitizing drug.22–24 In a phase II clinical trial,25 positive results were achieved for most human patients with aggressive cutaneous SCC that were administered gefitinib prior to radiation therapy or surgery. Currently, most of the available literature on drug treatment for SCC involves human patients with aggressive or metastatic SCC. Limitations of the present study include the inability to assess the respective correlations between standard treatments and EGFR immunoreactivity and treatment response. Although treatment and followup data were obtained for each horse, the variation in SCC size, numerous treatment modalities used, lack of a standardized follow-up protocol for many horses, and client financial constraints prevented us from evaluating those correlations. Additionally, antibodies against EGFR and HER2, which have not been evaluated by Western blot analysis, were used as the primary antiAJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:55 AM
bodies in the immunohistochemical staining protocols of the present study because equine-specific antibodies against EGFR and HER2 have not been developed. Murine and equine EGFR and HER2 proteins are 91% and 88% homologous, respectively.i Given the specificity of the staining distributions in the control eyelid, conjunctiva, and cornea specimens, it appears that the murine antibodies used in the immunohistochemical staining protocols of this study selectively reacted with equine EGFR and HER2. It was difficult to compare the results of the present study with other studies because there is great variation in the expression of EGFR and HER2 among species and studies. Some of this variation is undoubtedly caused by differences among species and between clinically normal and neoplastic tissues, but it could also be caused by tumor stage and variations in tissue fixation and storage (including length of storage), primary antibodies used to identify EGFR expression, antigen retrieval methods, and staining techniques.15 In the present study, most of the periocular SCCs obtained from horses expressed EGFR and HER2, which suggested that administration of EGFR inhibitors might be useful for the treatment of those tumors, assuming that extracellular or intracellular EGFR and HER2 have a pathogenic role in the development of SCC. Variation in immunoreactivity to both EGFR and HER2 among the SCCs of the present study might indicate that each tumor should be immunohistochemically evaluated prior to treatment to determine which EGFR inhibitor is likely to be most effective, as is currently done for human patients with breast cancer. However, some human patients with colorectal tumors that do not have immunoreactivity to EGFR respond favorably to treatment with EGFR inhibitors,26 and if periocular SCCs of horses respond similarly, immunohistochemical evaluation of each tumor may not be necessary. Further research is necessary to determine the efficacy and feasibility of EGFR inhibitors for the treatment of periocular SCC in horses. a. b. c. d. e. f. g. h. i.
Model S3400, Dako Corp, Carpinteria, Calif. Protein block, serum-free, Dako Corp, Carpinteria, Calif. EGFr Kit (Clone 31G7), Invitrogen, Frederick, Md. EnVision+System HRP Anti-Mouse, Dako Corp, Carpinteria, Calif. 3,3′-diaminobenzidine tetrahydrochloride DAB+, Dako Corp, Carpinteria, Calif. Hematoxylin 2, Richard Allen Scientific, Kalamazoo, Mich. Mouse anti-HER2 (c-erbB-2), Clone CB11, Invitrogen, Frederick, Md. SigmaStat 3.0, Systat Software Inc, San Jose, Calif. BLAST, National Center for Biotechnology Information, National Institutes of Health, Bethesda, Md. Available at: blast.ncbi. nlm.nih.gov/. Accessed May 13, 2014.
References 1. 2. 3.
Dugan SJ, Curtis CR, Roberts SM, et al. Epidemiologic study of ocular/adnexal squamous cell carcinoma in horses. J Am Vet Med Assoc 1991;198:251–256. Hendrix DVH. Equine ocular squamous cell carcinoma. Clin Tech Equine Pract 2005;4:87–94. Teifke JP, Lohr CV. Immunohistochemical detection of P53 overexpression in paraffin wax-embedded squamous cell carcinomas of cattle, horses, cats and dogs. J Comp Pathol 1996;114:205–210.
AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 917
4.
5. 6. 7.
8. 9.
10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20.
21.
22. 23. 24.
25. 26.
Dugan SJ, Roberts SM, Curtis CR, et al. Prognostic factors and survival of horses with ocular/adnexal squamous cell carcinoma: 147 cases (1978–1988). J Am Vet Med Assoc 1991;198:298– 303. Zwick E, Bange J, Ullrich A. Receptor tyrosine kinases as targets for anticancer drugs. Trends Mol Med 2002;8:17–23. Saxena R, Dwivedi A. ErbB family receptor inhibitors as therapeutic agents in breast cancer: current status and future clinical perspective. Med Res Rev 2012;32:166–215. Sabattini S, Marconato L, Zoff A, et al. Epidermal growth factor receptor expression is predictive of poor prognosis in feline cutaneous squamous cell carcinoma. J Feline Med Surg 2010;12:760–768. Looper JS, Malarkey DE, Ruslander D, et al. Epidermal growth factor receptor expression in feline oral squamous cell carcinomas. Vet Comp Oncol 2006;4:33–40. Shiomitsu K, Johnson CL, Malarkey DE, et al. Expression of epidermal growth factor receptor and vascular endothelial growth factor in malignant canine epithelial nasal tumours. Vet Comp Oncol 2009;7:106–114. Groves RW, Allen MH, MacDonald DM. Abnormal expression of epidermal growth factor receptor in cutaneous epithelial tumours. J Cutan Pathol 1992;19:66–72. Nazmi MN, Dykes PJ, Marks R. Epidermal growth factor receptors in human epidermal tumours. Br J Dermatol 1990;123:153–161. Gutierrez C, Schiff R. HER2: biology, detection, and clinical implications. Arch Pathol Lab Med 2011;135:55–62. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER2/neu proto-oncogene in human breast and ovarian cancer. Science 1989;244:707–712. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177–182. Blok EJ, Kuppen PJ, van Leeuwen JE, et al. Cytoplasmic overexpression of HER2: a key factor in colorectal cancer. Clin Med Insights Oncol 2013;7:41–51. Gaffney DC, Soyer HP, Simpson F. The epidermal growth factor receptor in squamous cell carcinoma: an emerging drug target. Australas J Dermatol 2014;55:24–34. Borkar DS, Lacouture ME, Basti S. Spectrum of ocular toxicities from epidermal growth factor receptor inhibitors and their intermediate-term follow-up: a five-year review. Support Care Cancer 2013;21:1167–1174. Gill M, Jain K, Manjari M, et al. Expression of Her-2/neu in colon carcinoma and its correlation with the histologic grades and the lymph nodes status. J Clin Diagn Res 2011;5:1564–1568. Press MF, Cordon-Cardo C, Slamon DJ. Expression of the HER2/neu proto-oncogene in normal human adult and fetal tissues. Oncogene 1990;5:953–962. Liu Z, Carvajal M, Carraway CA, et al. Expression of the receptor tyrosine kinases, epidermal growth factor receptor, ErbB2, and ErbB3, in human ocular surface epithelia. Cornea 2001;20:81–85. Wolff AC, Hammond ME, Schwartz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 2007;25:118–145. Kim S, Eleff M, Nicolaou N. Cetuximab as primary treatment for cutaneous squamous cell carcinoma to the neck. Head Neck 2011;33:286–288. Miller K, Sherman W, Ratner D. Complete clinical response to cetuximab in a patient with metastatic cutaneous squamous cell carcinoma. Dermatol Surg 2010;36:2069–2074. Göppner D, Nekwasil S, Franke I, et al. Successful combination therapy of a locally advanced squamous cell carcinoma of the skin with cetuximab and gamma-irradiation. J Dtsch Dermatol Ges 2010;8:826–828. Lewis CM, Glisson BS, Feng L, et al. A phase II study of gefitinib for aggressive cutaneous squamous cell carcinoma of the head and neck. Clin Cancer Res 2012;18:1435–1446. 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 2005;23:1803–1810. 917
9/19/2014 10:55:55 AM
Objective—To determine whether epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2) are expressed in periocular squamous cell carcinomas (SCCs) of horses. Sample—Biopsy specimens of SCCs from 46 horses. Procedures—Pathology records were searched retrospectively for biopsy specimens of periocular SCCs obtained from horses. Slides of the specimens were reviewed histologically to confirm the SCC diagnosis and stained for EGFR and HER2 by immunohistochemical methods. For both EGFR and HER2, the immunohistochemical staining intensity and percentage of stain-positive cells on the slides were determined. Results—43 of 46 (93%) SCCs were immunoreactive for EGFR. The median score for EGFR staining intensity was 4 (range, 2 to 12), and the median number of mitotic figures was 8 mitotic figures/10 hpfs (range, 0 to 34 mitotic figures/10 hpfs). Mitotic index was not correlated with the percentage of EGFR stain–positive cells or staining intensity. Of the 43 EGFR-immunoreactive SCCs, 38 had stain present primarily in the cytoplasm and 5 had stain equally distributed between the cytoplasm and cell membranes. Thirty-five of 46 (76%) SCCs were immunoreactive for HER2. Mitotic index was not correlated with the percentage of HER2 stain–positive cells or staining intensity. Of the 35 HER2-immunoreactive SCCs, the stain was present primarily in the cytoplasm and 7 had stain equally distributed between the cytoplasm and cell membranes. Conclusions and Clinical Relevance—Results indicated that most periocular SCCs of horses expressed EGFR and HER2, which suggested that equine periocular SCCs might respond to treatment with EGFR inhibitors. (Am J Vet Res 2014;75:912–917)
S
quamous cell carcinomas are the most common tumors of the eyes and adnexa of horses and commonly affect the eyelids, third eyelid (nictitating membrane), conjunctiva, and limbus.1,2 Breeds most frequently affected by SCC include Appaloosa, American Paint Horse, and draft breeds, and the prevalence of SCC is positively associated with age.1 In horses, development of SCC is thought to be related to an increased susceptibility to ultraviolet radiation carcinogenesis because all periocular SCCs express p53, lightly pigmented horses are predisposed to the development of periocular SCC, and the prevalence of SCCs is positively associated with altitude, longitude, and mean annual solar radiation.1,3 When left untreated, periocular SCC can invade the local soft tissues, bony orbit, sinuses, and brain and Received April 23, 2014. Accepted June 23, 2014. From the Departments of Small Animal Clinical Sciences (Hendrix) and Biomedical and Diagnostic Sciences (Newkirk), College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996. Supported by the Morris Animal Foundation. Presented in abstract form at the 44th Annual Conference of the American College of Veterinary Ophthalmologists, Rio Grande, Puerto Rico, November 2013. The authors thank Misty Bailey for technical editing. Address correspondence to Dr. Hendrix ([email protected]). 912
14-04-0115r.indd 912
EGFR HER SCC TBST
ABBREVIATIONS
Epidermal growth factor receptor Human epidermal growth factor receptor Squamous cell carcinoma Tris-buffered saline solution with Tween 20
metastasize to regional lymph nodes, salivary glands, and thorax. Following treatment, SCC recurrence most commonly involves the eyelid or orbit.4 Treatment of periocular SCC in horses is dependent on tumor location, tumor size, extent of tissue invasion, vision status, intended purpose of the affected horse, equipment availability, and financial constraints. Generally, the treatment of choice is surgical excision of the tumor followed by adjunctive treatment such as cryotherapy, iridium-192, strontium-90, intralesional cisplatin, or 5-fluorouracil.2 Although many treatment modalities are used for the treatment of periocular SCCs in horses, few of them are available to general practitioners, and most are associated with some risk for the personnel administer the treatment. In human medicine, there has been a tremendous amount of preclinical and clinical research to evaluate the role of EGFRs in tumor growth and targeted cancer treatment. Four types of EGFRs (EGFR, HER2, HER3, AJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:53 AM
and HER4) have been identified, of which EGFR and HER2 have been studied most frequently. Most healthy epithelial tissues express EGFR and HER2, which are instrumental for proliferation, regeneration, differentiation, and development of epithelial cells. Increased expression of EGFR and HER2 from basal levels has been associated with carcinogenesis and tumor progression.5 Consequently, a new wave of pharmaceutical development has focused on the creation of treatments that inhibit EGFR. In human patients with certain tumors, the type of EGFR expressed by the tumor has prognostic value and is used to tailor treatment because some drugs selectively inhibit some receptors but not others.6 In a study7 of 19 cats with cutaneous SCCs, 14 had tumors that tested positive for EGFR expression and SCC expression of EGFR was predictive of a poor prognosis. Expression of EGFR has also been identified in oral SCCs of cats,8 malignant epithelial nasal tumors of dogs,9 and SCCs and other epithelial tumors of humans.10,11 Human epidermal growth factor receptor 2 has been most commonly studied in human patients with breast or colorectal cancer, and testing for HER2 immunoreactivity has been used extensively for targeted treatment with EGFR-inhibiting drugs.12–15 Because healthy epithelial cells also have EGFRs, use of EGFR inhibitors is commonly associated with the development of rashes and occasional adverse effects associated with ocular tissues such as dysfunctional tear syndrome, blepharitis, and eyelash changes.16,17 The detection of EGFR or HER2 expression in periocular SCCs of horses would indicate that preclinical trials to investigate the effectiveness of EGFR inhibitors for targeted treatment of such tumors are warranted. Although EGFR inhibitors are generally administered IV in human patients because of concerns about actual or potential metastatic disease, it might be possible to administer EGFR inhibitors intralesionally in periocular SCCs of horses because of their low rate of metastasis. Epidermal growth factor receptor inhibitors are expensive. Theoretically, for horses with periocular SCCs, intralesional injection of EGFR inhibitors would greatly decrease the cost of treatment and prevent the adverse systemic effects associated with those drugs. Of course, investigation of EGFR inhibitors in cell cultures would be necessary to determine the optimal drug concentration for cytotoxicity. However, because of the targeted nature of EGFR inhibitors, it is first necessary to determine whether periocular SCCs of horses express EGFR or HER2 before preclinical or clinical trials are initiated. Also, elucidation of whether the target EGFR proteins are located on the cell membrane or in the cytoplasm will assist investigators in choosing the drug most likely to be effective for those trials. The purpose of the study reported here was to determine whether EGFR and HER2 are expressed by periocular SCCs of horses. Materials and Methods Specimens—University of Tennessee Veterinary Medical Center pathology records from 2001 through 2011 were searched for biopsy specimens of periocular SCCs obtained from horses. Specimens were selected for the study on the basis of the reported histologic AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 913
diagnosis, tumor location, and availability of archived tumor specimens for further evaluation. The search resulted in the identification of 46 periocular SCCS that were eligible for the study, of which 14 involved the eyelid, 12 involved the nictitating membrane, 11 involved the cornea and limbus, and 9 involved the conjunctiva. Histologic slides of the SCCs were reviewed by a board-certified anatomic pathologist (KMN) to confirm the diagnosis. For horses that were treated at the University of Tennessee Veterinary Medical Center, medical records were reviewed to obtain information about the patient’s response to treatment, and referring veterinarians or owners were contacted by telephone to obtain follow-up information for horses that were not treated at the university. Additionally, an eyelid, conjunctiva, nictitating membrane, and cornea were harvested from 1 eye of each of 3 adult horses with clinically normal eyes that were euthanized for reasons unrelated to the study for use as control specimens. This study did not require oversight or approval from an institutional animal care and use committee because all SCC specimens were obtained from the pathology laboratory’s tissue archives and the control specimens were obtained after the horses were euthanized. Immunohistochemical staining—All biopsy specimens were stored in blocks of paraffin. Each specimen was sectioned into slices that were 5 µm thick, and each slice was placed on a charged microscope slide, air-dried, and then heated at 60°C for 15 minutes. The tissue specimens on each slide were deparaffinized with xylene and rehydrated through graded ethanols to deionized water in a routine manner. Two slides (1 for EGFR and 1 for HER2) from each biopsy specimen were prepared for immunohistochemical staining. For the immunohistochemical staining procedure for EGFR, the slides were rinsed with deionized water and soaked in TBST for 10 minutes. Slides were then loaded on an autostainer,a where all procedures were performed at room temperature (22°C), and slides were rinsed with TBST between each step. The slides were then exposed to a 3% hydrogen peroxide solution to block endogenous peroxidase for 5 minutes, a serumfree protein blocking solutionb for 5 minutes, mouse anti-EGFr clone:31G7c (primary antibody) at a dilution of 1:40 for 30 minutes, the secondary reagents in a horseradish peroxidase–labeled polymer systemd for 30 minutes, and chromagene for 10 minutes. The slides were removed from the autostainer, rinsed in deionized water, and stained for 5 seconds with hematoxylin.f Then, the slides were blued in ammonia water, dehydrated through ethanol, and cleared with xylene, and a coverslip was placed over the tissue specimen. For the immunohistochemical staining procedure for HER2, slides were heated with EDTA buffer (pH, 9) in a steamer at 95°C for 20 minutes, then cooled for 20 minutes for heat-induced epitope retrieval. After epitope retrieval, all slides were rinsed in deionized water and soaked in TBST for 10 minutes. Slides were then loaded on the autostainera and subsequently processed in the same manner as those that were stained with EGFR, except that the primary antibody used was mouse anti-HER2 clone:CB11g at a dilution of 1:40 for 30 minutes. 913
9/19/2014 10:55:53 AM
percentage of stain-positive cells was classified into 1 of 3 categories (1 = ≤ 40%, 2 = > 40% to 70%, or 3 = > 70%) and no reactivity score was calculated.18
Slide evaluation—For each slide, the staining intensity, location of the stain within the cells, and percentage of stain-positive cells were scored as described for EGFR7–9 and HER2.18 Briefly, for both EGFR and HER2, the staining intensity was scored on a 4-point scale (0 = no staining, 1 = low intensity, 2 = moderate intensity, or 3 = high intensity). The location of the stain within the cells was categorized as cytoplasm, cell membrane, or both.7–9,18 For EGFR, the percentage of stain-positive cells was classified into 1 of 4 categories (1 = ≤ 10%, 2 = > 10% to 30%, 3 = > 30% to 60%, or 4 = > 60%). The EGFR reactivity score was defined as the product of the intensity score and the category for the percentage of stain-positive cells, and an EGFR reactivity score ≥ 2 was considered positive.7–9 For HER2, the
Statistical analysis—The respective correlations between mitotic indices and staining intensity and percentage of stain-positive cells for both EGFR and HER2 were determined with the Pearson product moment correlation. Statistical softwareh was used for all analyses, and values of P ≤ 0.05 were considered significant. Results Specimens—Periocular SCC tissue specimens from 46 horses were evaluated. Breeds represented included American Paint Horse (n = 13), Appaloosa (9),
Table 1—Median (range) values for various measures of expression of EGFR and HER2 in periocular SCC (n = 46) and normal (control; 3) tissue specimens obtained from horses. Type of EGFR
Location of lesion
EGFR
Cornea
HER2
Specimen type
Percentage of stainpositive cells*
SCC (n = 11) Control (n = 3) Conjunctiva SCC (n = 9) Control (n = 3) Nictitating membrane SCC (n = 10) Control (n = 3) Eyelid SCC (n = 13) Control (n = 3) Cornea SCC (n = 6) Control (n = 3) Conjunctiva SCC (n = 7) Control (n = 3) Nictitating membrane SCC (n = 8) Control (n = 3) Eyelid SCC (n = 14) Control (n = 3)
4 (2–4) 1 4 (2–4) 3 3 (1–4) 2 4 (2–4) 3 3 1 3 (2–3) 3 3 (2–3) 3 3 (2–3) 3
Staining intensity†
Reactivity‡
1 (1–3) 2 1 (1–3) 2 1.5 (1–2) 2 2 (1–2) 2 2 2 2 (1–2) 2 1.5 (1–3) 2 2 (1–2) 2
4 (2–12) 2 4 (2–12) 6 3 (2–8) 4 6 (2–8) 6 — — — — — — — —
Of the 46 SCC specimens, 43 stained positive for EGFR and 35 stained positive for HER2. For all values, the lack of a range in parenthesis indicates that all specimens had the same value for that variable. *For EGFR, the percentage of stain-positive cells was classified into 1 of 4 categories (1 = ≤ 10%, 2 = > 10% to 30%, 3 = > 30% to 60%, or 4 = > 60%), whereas for HER2, the percentage of stain-positive cells was classified into 1 of 3 categories (1 = ≤ 40%, 2 = > 40% to 70%, 3 = > 70%). †For both EGFR and HER2, the staining intensity was scored on a 4-point scale (0 = no staining, 1 = low intensity, 2 = moderate intensity, or 3 = high intensity). ‡For EGFR only, the reactivity score was defined as the product of the staining intensity score and the category for the percentage of stain-positive cells, and an EGFR reactivity score ≥ 2 was considered positive. — = Not calculated. Table 2—Distribution of immunohistochemical stains for EGFR and HER2 in periocular SCC (n = 46) and clinically normal (control; 3) tissue specimens obtained from horses. Location of immunohistochemical stain in stain-positive cells (No. of samples) Type of EGFR EGFR
Location of lesion Cornea Conjunctiva Nictitating membrane Eyelid
HER2
Cornea Conjunctiva Nictitating membrane Eyelid
Specimen type
Cytoplasm only
Cell membrane only
SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control SCC Control
7 0 2 0 6 0 7 0 2 3 0 3 4 3 6 3
0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Equally between cytoplasm and cell membrane 1 0 2 0 0 0 2 0 2 0 4 0 1 0 0 0
Primarily cytoplasm
Primarily cell membrane
3 0 5 0 4 3 4 0 2 0 3 0 3 0 8 0
0 0 0 3 0 0 0 3 0 0 0 0 0 0 0 0
See Table 1 for key. 914
14-04-0115r.indd 914
AJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:53 AM
Quarter Horse (7), Tennessee Walking Horse (7), and 7 other breeds (10). Of the 46 affected horses, 27 were geldings, 15 were mares, and 4 were stallions. The mean ± SD age of horses at the time the biopsy specimens were obtained was 13 ± 5.2 years. Thirty-seven of the SCC specimens were obtained from patients examined at the University of Tennessee Veterinary Medical Center; the remaining 9 specimens were obtained from outside submissions to the pathology laboratory. The clinically normal (control) tissue specimens were obtained from 2 Appaloosas and 1 Mangalarga Marchador. Immunohistochemical staining— The EGFR staining intensity and percentage of stain-positive cells were consistent among control specimens (Table 1). For the control eyelid skin specimens, the stain was located in the cell membrane to a slightly greater extent than in the cytoplasm, and the stain was most intense in the basal region and decreased in intensity through the superficial layers. For the conjunctival specimens, the stain was located primarily in the cell membranes, and its intensity varied throughout the specimens. For the nictitating membrane specimens, the stain was located in both the cell membranes and cytoplasm of cells in the basal layer, and its intensity was varied throughout the specimens. For corneal specimens, staining was limited to the cell membranes of the basal cells at the limbus (Table 2). Forty-three of 46 (93%) SCCs had an EGFR reactivity score ≥ 2 (Figure 1). Of the 3 SCCs that had a reactivity score of 0 or 1, 2 were obtained from the nictitating membrane and 1 was obtained from the eyelid. For the 43 EGFR stain–positive SCCs with a reactivity score ≥ 2, the median reactivity score was 4 (range, 2 to 12; Table 1) and the median number of mitotic figures was 8 mitotic figures/10 hpfs (range, 0 to 34 mitotic figures/10 hpfs). The mitotic index was not significantly correlated with the percentage of stain-positive cells (r = 0.03; P = 0.403) or staining intensity (r = 0.04; P = 0.523). The stain was located in the cytoplasm of all 43 EGFR stain–positive SCCs; it was located only in the cytoplasm of 22 tumors, was located primarily in the cytoplasm of 16 tumors, and equally distributed between the cytoplasm and cell membrane of 5 tumors (Table 2). For the control specimens stained for detection of HER2, the stain was located only in the cytoplasm (Table 2). In AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 915
the corneal specimens, the percentage of stain-positive cells was greatest in the basal layers, whereas in the eye-
Figure 1—Representative photomicrographs of sections of periocular SCCs of horses that were immunohistochemically stained for EGFR and depict the 4 possible scores for staining intensity. A—Photomicrograph of a section of an SCC removed from an eyelid; staining intensity, 0 (no staining). B—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 1 (low intensity). Notice that the stain is located primarily in the cytoplasm. C—Photomicrograph of a section of an SCC removed from a conjunctiva; staining intensity, 2 (moderate intensity). Notice that the stain is located primarily in the cytoplasm. D—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 3 (high intensity). Notice that the stain is again located primarily in the cytoplasm. Immunohistochemical stain; bar = 10 µm.
Figure 2—Representative photomicrographs of sections of periocular SCCs of horses that were immunohistochemically stained for HER2 and depict the 4 possible scores for staining intensity. A—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 0 (no staining). B—Photomicrograph of a section of an SCC removed from a nictitating membrane; staining intensity, 1 (low intensity). Notice that the stain is located primarily in the cytoplasm. C—Photomicrograph of a section of an SCC removed from a cornea; staining intensity, 2 (moderate intensity). Notice that the stain is equally distributed between the cytoplasm and cell membranes. D—Photomicrograph of a section of an SCC removed from a nictitating membrane; staining intensity, 3 (high intensity). Notice that the stain is again located primarily in the cytoplasm. Immunohistochemical stain; bar = 10 µm. 915
9/19/2014 10:55:55 AM
lid, conjunctiva, and nictitating membrane specimens, the percentage of stain-positive cells was greatest in the superficial layers of the epithelium. Thirty-five of 46 (76%) SCCs stained positive for HER2 (Figure 2). Of the 11 HER2 stain–negative SCCs, 2 were obtained from the conjunctiva, 5 were obtained from the cornea, and 4 were obtained from the nictitating membrane. The mitotic index was not significantly correlated with the percentage of stain-positive cells (r = –0.14; P = 0.884) or staining intensity (r = –0.16; P = 0.770). The stain was located in the cytoplasm of all 35 HER2 stain–positive SCCs; it was located only in the cytoplasm of 12 tumors, was located primarily in the cytoplasm of 16 tumors, and was equally distributed between the cytoplasm and cell membrane of 7 tumors (Table 2). Discussion Results of the present study indicated that most periocular SCCs of horses expressed EGFR and HER2, and the percentage of cells that stained positive for EGFR and HER2 in those tumors was generally greater than that in control specimens. These results, along with the fact that abnormalities in EGFR status such as overexpression, mutation, signaling errors, and dysregulated receptor trafficking have been associated with the development of SCC and other tumors,16 suggest that administration of anti-EGFR drugs might be beneficial for the treatment of periocular SCCs in horses. For the equine periocular SCCs of the present study that expressed EGFR and HER2, the immunohistochemical stain was generally located primarily in the cytoplasm, whereas for feline cutaneous SCC that expressed EGFR, the stain was generally located in the cell membrane,7,8 with only 5 of 197 and 4 of 138 tumors having faint to moderate cytoplasmic staining. Similar to cutaneous SCCs of cats, malignant epithelial nasal tumors of dogs generally express EGFR in the cell membrane.9 Immunohistochemical staining of dermal SCCs of human patients is generally heterogeneous, with the stain accumulating primarily in the cytoplasm in the less differentiated areas of the tumor,10,11 which is believed to occur because of accumulation of EGFRs in the cytoplasm owing to excessive stimulation of the cell by epidermal growth factor or epidermal growth factor α or secondary to changes in the EGFR molecule.10 The control specimens of the present study expressed HER2 in the cytoplasm but not in the cell membrane. Conversely, of the 35 SCCs that expressed HER2, 23 had expression of HER2 in the cell membrane as well as in the cytoplasm. In a study15 in which fluorescence in situ hybridization was used to determine HER2 expression by human colorectal tumors, membrane staining was associated with gene amplification of the HER2 molecule but the mechanism of cytoplasmic staining was not identified. In the present study, the percentage of HER2 stain–positive cells in SCCs that involved the cornea was greater than that in the control specimens; however, for SCCs that involved the other ocular tissues, the percentage of HER2 stain–positive cells was similar to that for the corresponding control specimens. In humans, the basal HER2 expression is low in healthy epithelial cells and is increased from bas916
14-04-0115r.indd 916
al levels in fetal tissues and neoplasms.19 Expression of HER2 is localized primarily in the superficial layers of the ocular surface epithelium of humans,20 which was also observed in the control specimens of all equine tissues evaluated in the present study except for the cornea. The finding that HER2 is expressed primarily on the surfaces of clinically normal tissues suggests that HER2 is preferentially expressed by differentiated postmitotic epithelial cells.20 Interpretation of HER2 reactivity in tissues has not been completely elucidated. Most research into HER2 reactivity has involved breast cancer in human patients. In human breast tumors, HER2 expression is increased 20% to 30% from that in healthy breast tissue,12–14 and tumors that have an overexpression of HER2 tend to be more aggressive and are associated with a worse prognosis, compared with HER2-negative tumors.12–14 Human patients treated with trastuzumab, a monoclonal antibody against the extracellular domain of HER2, after breast cancer surgery had an increased survival rate and decreased tumor recurrence rate, compared with similar patients who were not treated with trastuzumab.13,14 However, because trastuzumab affects only the extracellular domain of HER2, it has no effect on HER2 localized in the cytoplasm, a fact that has directed interpretation of HER2 immunoreactivity in breast cancer.21 Investigators of another study15 report that human patients with colorectal tumors in which HER2 is localized in the cytoplasm might have a poorer prognosis than do patients with HER2-negative tumors. Nonetheless, the relevance and prognostic value of HER2 immunoreactivity patterns in various tumors and species require further investigation. Knowledge of the staining pattern and location of the different types of EGFRs is important when administration of anti-EGFR drugs is considered. The 3 types of anti-EGFR drugs include monoclonal antibodies such as cetuximab, which targets the extracellular portion of the EGFR molecule; small-molecule tyrosine kinase inhibitors such as erlotinib and gefitinib, which target the intracellular protein kinase domain; and dual tyrosine kinase inhibitors such as lapatinib, which targets both EGFR and HER2.16 Promising results have been achieved in human patients with cutaneous SCCs treated with cetuximab as monotherapy or as a radiation-sensitizing drug.22–24 In a phase II clinical trial,25 positive results were achieved for most human patients with aggressive cutaneous SCC that were administered gefitinib prior to radiation therapy or surgery. Currently, most of the available literature on drug treatment for SCC involves human patients with aggressive or metastatic SCC. Limitations of the present study include the inability to assess the respective correlations between standard treatments and EGFR immunoreactivity and treatment response. Although treatment and followup data were obtained for each horse, the variation in SCC size, numerous treatment modalities used, lack of a standardized follow-up protocol for many horses, and client financial constraints prevented us from evaluating those correlations. Additionally, antibodies against EGFR and HER2, which have not been evaluated by Western blot analysis, were used as the primary antiAJVR, Vol 75, No. 10, October 2014
9/19/2014 10:55:55 AM
bodies in the immunohistochemical staining protocols of the present study because equine-specific antibodies against EGFR and HER2 have not been developed. Murine and equine EGFR and HER2 proteins are 91% and 88% homologous, respectively.i Given the specificity of the staining distributions in the control eyelid, conjunctiva, and cornea specimens, it appears that the murine antibodies used in the immunohistochemical staining protocols of this study selectively reacted with equine EGFR and HER2. It was difficult to compare the results of the present study with other studies because there is great variation in the expression of EGFR and HER2 among species and studies. Some of this variation is undoubtedly caused by differences among species and between clinically normal and neoplastic tissues, but it could also be caused by tumor stage and variations in tissue fixation and storage (including length of storage), primary antibodies used to identify EGFR expression, antigen retrieval methods, and staining techniques.15 In the present study, most of the periocular SCCs obtained from horses expressed EGFR and HER2, which suggested that administration of EGFR inhibitors might be useful for the treatment of those tumors, assuming that extracellular or intracellular EGFR and HER2 have a pathogenic role in the development of SCC. Variation in immunoreactivity to both EGFR and HER2 among the SCCs of the present study might indicate that each tumor should be immunohistochemically evaluated prior to treatment to determine which EGFR inhibitor is likely to be most effective, as is currently done for human patients with breast cancer. However, some human patients with colorectal tumors that do not have immunoreactivity to EGFR respond favorably to treatment with EGFR inhibitors,26 and if periocular SCCs of horses respond similarly, immunohistochemical evaluation of each tumor may not be necessary. Further research is necessary to determine the efficacy and feasibility of EGFR inhibitors for the treatment of periocular SCC in horses. a. b. c. d. e. f. g. h. i.
Model S3400, Dako Corp, Carpinteria, Calif. Protein block, serum-free, Dako Corp, Carpinteria, Calif. EGFr Kit (Clone 31G7), Invitrogen, Frederick, Md. EnVision+System HRP Anti-Mouse, Dako Corp, Carpinteria, Calif. 3,3′-diaminobenzidine tetrahydrochloride DAB+, Dako Corp, Carpinteria, Calif. Hematoxylin 2, Richard Allen Scientific, Kalamazoo, Mich. Mouse anti-HER2 (c-erbB-2), Clone CB11, Invitrogen, Frederick, Md. SigmaStat 3.0, Systat Software Inc, San Jose, Calif. BLAST, National Center for Biotechnology Information, National Institutes of Health, Bethesda, Md. Available at: blast.ncbi. nlm.nih.gov/. Accessed May 13, 2014.
References 1. 2. 3.
Dugan SJ, Curtis CR, Roberts SM, et al. Epidemiologic study of ocular/adnexal squamous cell carcinoma in horses. J Am Vet Med Assoc 1991;198:251–256. Hendrix DVH. Equine ocular squamous cell carcinoma. Clin Tech Equine Pract 2005;4:87–94. Teifke JP, Lohr CV. Immunohistochemical detection of P53 overexpression in paraffin wax-embedded squamous cell carcinomas of cattle, horses, cats and dogs. J Comp Pathol 1996;114:205–210.
AJVR, Vol 75, No. 10, October 2014
14-04-0115r.indd 917
4.
5. 6. 7.
8. 9.
10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20.
21.
22. 23. 24.
25. 26.
Dugan SJ, Roberts SM, Curtis CR, et al. Prognostic factors and survival of horses with ocular/adnexal squamous cell carcinoma: 147 cases (1978–1988). J Am Vet Med Assoc 1991;198:298– 303. Zwick E, Bange J, Ullrich A. Receptor tyrosine kinases as targets for anticancer drugs. Trends Mol Med 2002;8:17–23. Saxena R, Dwivedi A. ErbB family receptor inhibitors as therapeutic agents in breast cancer: current status and future clinical perspective. Med Res Rev 2012;32:166–215. Sabattini S, Marconato L, Zoff A, et al. Epidermal growth factor receptor expression is predictive of poor prognosis in feline cutaneous squamous cell carcinoma. J Feline Med Surg 2010;12:760–768. Looper JS, Malarkey DE, Ruslander D, et al. Epidermal growth factor receptor expression in feline oral squamous cell carcinomas. Vet Comp Oncol 2006;4:33–40. Shiomitsu K, Johnson CL, Malarkey DE, et al. Expression of epidermal growth factor receptor and vascular endothelial growth factor in malignant canine epithelial nasal tumours. Vet Comp Oncol 2009;7:106–114. Groves RW, Allen MH, MacDonald DM. Abnormal expression of epidermal growth factor receptor in cutaneous epithelial tumours. J Cutan Pathol 1992;19:66–72. Nazmi MN, Dykes PJ, Marks R. Epidermal growth factor receptors in human epidermal tumours. Br J Dermatol 1990;123:153–161. Gutierrez C, Schiff R. HER2: biology, detection, and clinical implications. Arch Pathol Lab Med 2011;135:55–62. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER2/neu proto-oncogene in human breast and ovarian cancer. Science 1989;244:707–712. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177–182. Blok EJ, Kuppen PJ, van Leeuwen JE, et al. Cytoplasmic overexpression of HER2: a key factor in colorectal cancer. Clin Med Insights Oncol 2013;7:41–51. Gaffney DC, Soyer HP, Simpson F. The epidermal growth factor receptor in squamous cell carcinoma: an emerging drug target. Australas J Dermatol 2014;55:24–34. Borkar DS, Lacouture ME, Basti S. Spectrum of ocular toxicities from epidermal growth factor receptor inhibitors and their intermediate-term follow-up: a five-year review. Support Care Cancer 2013;21:1167–1174. Gill M, Jain K, Manjari M, et al. Expression of Her-2/neu in colon carcinoma and its correlation with the histologic grades and the lymph nodes status. J Clin Diagn Res 2011;5:1564–1568. Press MF, Cordon-Cardo C, Slamon DJ. Expression of the HER2/neu proto-oncogene in normal human adult and fetal tissues. Oncogene 1990;5:953–962. Liu Z, Carvajal M, Carraway CA, et al. Expression of the receptor tyrosine kinases, epidermal growth factor receptor, ErbB2, and ErbB3, in human ocular surface epithelia. Cornea 2001;20:81–85. Wolff AC, Hammond ME, Schwartz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 2007;25:118–145. Kim S, Eleff M, Nicolaou N. Cetuximab as primary treatment for cutaneous squamous cell carcinoma to the neck. Head Neck 2011;33:286–288. Miller K, Sherman W, Ratner D. Complete clinical response to cetuximab in a patient with metastatic cutaneous squamous cell carcinoma. Dermatol Surg 2010;36:2069–2074. Göppner D, Nekwasil S, Franke I, et al. Successful combination therapy of a locally advanced squamous cell carcinoma of the skin with cetuximab and gamma-irradiation. J Dtsch Dermatol Ges 2010;8:826–828. Lewis CM, Glisson BS, Feng L, et al. A phase II study of gefitinib for aggressive cutaneous squamous cell carcinoma of the head and neck. Clin Cancer Res 2012;18:1435–1446. 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 2005;23:1803–1810. 917
9/19/2014 10:55:55 AM
