This article is mentioned in:

Introduction

Though considered rare, anal cancer incidence has increased in the last 20 years in the USA (1), with an estimated 8,080 new cases and 1,080 associated mortalities in 2016 (2). The most frequent histological type is squamous cell carcinoma (SCC), responsible for ~95% of anal tumors (3). The risk factors for anal cancer include human papillomavirus (HPV) infection, immunodeficiency, immunosuppression and tobacco smoking; however being HPV+ has been demonstrated to be the most important factor (4,5). Among the distinct HPV types, HPV16 is the most frequently identified in anal SCC (6) and may cause infected cells to progress from intraepithelial neoplasia to high-grade dysplasia, and finally to invasive cancer (5,7). Additionally, an increased prevalence of infection with high-risk HPV (HR-HPV) has been identified in the anal canals of women with HPV-associated pathologies, including cancer of the vulva, vagina and cervix, compared with in women without these HPV-associated pathologies (8).

It has been revealed that HR-HPV infections, particularly with concomitantly elevated expression of the viral gene products E6 and E7, are involved in cell cycle entry and cell proliferation (9,10). E7 protein binds retinoblastoma protein, inducing its degradation and aberrant cell cycle progression by upregulating p21 and p16 (11). E6 protein is involved in several oncogenic events, leading to p53 degradation, nuclear factor (NF)-κB pathway activation under hypoxic conditions, and the upregulation of human telomerase reverse transcriptase, which promotes cell immortalization (11). A previous study identified NF-κB as an independent predictor of disease-free survival (DFS) in patients treated with chemotherapy (12).

Raf kinase inhibitor protein (RKIP) is described as an NF-κB suppressor (13), an inhibitor of the Raf/mitogen-activated protein kinase/extracellular signal-regulated kinase (ERK) kinase (MEK)/ERK signaling pathway (14), and a regulator of G protein-coupled receptors (15). RKIP is expressed in several normal human tissues (16), and previous studies have demonstrated the prognostic value of the loss of RKIP expression in various gastrointestinal tumors, including colorectal cancer, gastrointestinal stromal tumors and gastric cancer (17–19). To date, only a limited number of molecular biomarkers have been identified that are able to predict treatment outcomes or response in anal cancer, including HPV infection, p16 protein expression (20,21), and the overexpression of multidrug resistance-associated protein 1, excision repair cross-complementation group 1 and thymidylate synthase (22). KRAS proto-oncogene (KRAS) mutations, which are predictors of patient outcomes (23) in colorectal tumors, were identified at a low frequency in tumors of the anal canal (24). Therefore, the aim of the present study was to evaluate the clinical implications of RKIP expression using immunohistochemistry in a series of invasive lesions (SCCs and adenocarcinomas) and in situ lesions [high-grade squamous intra-epithelial lesions (HSILs)] of the anal canal.

Materials and methods

Patients and samples

The resected tumors of 48 patients with anal cancer (27 female, 21 male), diagnosed at Barretos Cancer Hospital (Barretos, SP, Brazil) between June 2000 and August 2010, were evaluated and the clinical data are summarized in Table I. The mean age of the patients was 55.9 years (range, 27–86 years). KRAS and BRAF mutation status data were retrieved from our previous study (24), while HPV16 and HPV18 status, and β-globin, p16, Ki-67, minichromosome maintenance protein complex (MCM) and topoisomerase II α (TOP2A) expression immunohistochemical analyses were retrieved from another previous study (25) and are summarized in Table II. All patients included in the present study provided written informed consent, and the study was approved by the Barretos Cancer Hospital Ethical Committee (Barretos, SP, Brazil) approval under the protocol number 310/2010. The data of all patients alive by December 2012 were censored.

Table I. Clinicopathological characteristics of the patients and association with RKIP expression.

Table I.

Clinicopathological characteristics of the patients and association with RKIP expression.

Parameter	n	Low RKIP, n (%)	High RKIP, n (%)	P-value
Sex				0.999
Female	27	24 (88.9)	3 (11.1)
Male	21	18 (85.7)	3 (14.3)
Ethnicity				0.019a
Caucasian	40	37 (92.5)	3 (7.5)
Non-caucasian	8	5 (62.5)	3 (37.5)
Age, years				0.047a
≤46	12	8 (66.7)	4 (33.3)
47–66	24	22 (91.7)	2 (8.3)
≥67	12	12 (100)	0 (0.0)
Tobacco consumption				0.665
No	23	19 (82.6)	4 (17.4)
Yes (current or past)	22	20 (90.9)	2 (9.1)
Clinical grade				0.038a
Stage 0	6	3 (50)	3 (50)
Stages I and II	24	22 (91.7)	2 (8.3)
Stages III and IV	15	14 (93.3)	1 (6.7)
Histological type				0.037a
SCC	26	23 (88.5)	3 (11.5)
Adenocarcinoma	14	14 (100)	0
HSIL	8	5 (62.5)	3 (37.5)
Clinical response				0.847
No evidence	24	20 (83.3)	4 (16.7)
Complete	12	11 (91.7)	1 (8.3)
Progression	7	6 (85.7)	1 (14.3)
Status				0.743
Deceased (by cancer)	13	12 (92.3)	1 (7.7)
Deceased (not by cancer)	4	4 (100.0)	0 (0.0)
Alive without disease	26	21 (80.8)	5 (19.2)
Alive with disease	5	5 (100.0)	0 (0.0)

{ label (or @symbol) needed for fn[@id='tfn1-ol-0-0-8828'] } Where data is missing regarding certain patients, it will not available from the patient medical records.

a P<0.05. SCC, squamous cell carcinoma; HSIL, high-grade squamous intra-epithelial lesions; RKIP, Raf kinase inhibitor protein.

Table II. Pathological characteristics of the patients and association with RKIP expression.

Table II.

Pathological characteristics of the patients and association with RKIP expression.

Parameter	n	Low RKIP labeling, n (%)	High RKIP labeling, n (%)	P-value
HPV16				0.013b
Negative	10	6 (60.0)	4 (40.0)
Positive	38	36 (94.7)	2 (5.3)
β-globina				0.999
Weak	3	3 (100.0)	0 (0.0)
Strong	45	39 (86.7)	6 (13.3)
p16a				0.999
Negative	13	11 (84.6)	2 (15.4)
Positive	34	30 (88.2)	4 (11.8)
NA	1	NA	NA
Ki-67a				0.637
1+2+3+	12	10 (83.3)	2 (16.7)
4+	35	31 (88.6)	4 (11.4)
NA	1	NA	NA
MCMa				0.999
Negative	9	8 (88.9)	1 (11.1)
Positive	39	34 (87.2)	5 (12.8)
TOP2Aa				0.329
1+2+	32	27 (90.0)	3 (10.0)
3+	12	9 (75.0)	3 (25.0)
NA	4	NA	NA

a Immunohistochemical values acquired from reference (25).

b P<0.05. NA, not available; HPV, human papillomavirus; MCM, minichromosome maintenance protein complex; TOP2A, topoisomerase II α; RKIP, Raf kinase inhibitor protein.

Samples, fixed for 24 h at room temperature in 10% buffered formalin and embedded in paraffin blocks, were retrieved from the Department of Pathology of Barretos Cancer Hospital. Subsequently, 5-µm-thick sections were placed on histological slides, and hematoxylin and eosin staining (1 min in hematoxylin and 30 sec in eosin at room temperature) was performed to confirm the initial diagnosis using a light microscope (magnification, ×400).

Immunohistochemistry

Sections (5-µm thick) were subjected to immunohistochemical staining according to the streptavidin-biotin peroxidase method using an UltraVision Large Volume Detection system Anti-Polyvalent, HRP kit (cat. no. TP-125-HL; LabVision; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Briefly, slides were xylene-deparaffinized and rehydrated in a descending alcohol series (100, 90, 70 and 50%, followed by water), prior to antigen-retrieval for 20 min at 98°C in 10 mM citrate buffer (pH=6.0; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). Endogenous peroxidase blocking was performed by incubation with hydrogen peroxide (3% in methanol v/v; Sigma-Aldrich; Merck KGaA) for 10 min at room temperature. Protein blocking was performed for 5 min with UV Block Plus (LabVision; Thermo Fisher Scientific, Inc.). Subsequently, the slides were incubated with a polyclonal primary anti-RKIP antibody (dilution, 1:1,000; cat. no. 07–137; Upstate Biotechnology, Inc., Lake Placid, NY, USA) for 60 min at room temperature, followed by a secondary biotinylated goat anti-polyvalent antibody (part of the aforementioned kit) for 5 min at room temperature, after which they were incubated with the streptavidin-peroxidase complex for 5 min (LabVision; Thermo Fisher Scientific, Inc.) at room temperature. Chromogen color development was accomplished with 3,3′-diaminobenzidine (5 min at room temperature) and a Gill-2 hematoxylin counterstain (30 sec at room temperature).

The slides were scored in a blinded manner by one pathologist and one histologist using semi-quantitative criteria, as previously described (18,26) using a light microscope (magnification, ×400). The labeling score was determined as the sum of the percentage of positive cells (0, negative; 1, <5% immunoreactive cells; 2, 5–50% immunoreactive cells; 3, >51% immunoreactive cells) and the labeling intensity (0, negative; 1, weak; 2, moderate; 3, strong) in the tumor tissue. RKIP expression was classified as low (scores between 0 and 4) or high (scores 5 and 6). This scoring system was used to compare strongly positive RKIP samples vs. samples with weak/moderate expression. Normal stomach sections expressing RKIP were used as a positive control (19).

Statistical analysis

Analysis of descriptive characteristics was performed to characterize the study population using SPSS statistical software (version 21; IBM Corp., Armonk, NY, USA). Comparisons between groups were performed using Fisher's exact tests and the Kaplan-Meier method was applied to assess survival rates, using a log-rank test to compare the curves. P<0.05 was considered to indicate a statistically significant difference.

Results

Patient characteristics and survival analysis

Follow-up was performed every 6 months or as necessary (until the point of mortality), with a mean follow-up time of 57.2 months (range, 2.1–158.4 months). From the 48 cases, 8 were diagnosed with HSIL, 14 were diagnosed with adenocarcinomas and 26 were diagnosed with SCC. Log-rank analysis revealed a lower 5-year survival rate for patients with adenocarcinoma (40.8%), compared with for those with SCC and HSIL (76.7 and 80%, respectively; P=0.0479; Fig. 1A). No significant difference was observed in the disease-free survival rates between the three tumor types (P=0.201; Fig. 1B). There was no significant difference in the overall survival of patients based on clinical stage (Fig. 1C), but clinical stages III/IV were associated with a poorer disease-free survival rate (a 5-year disease-free survival rate of 37.3% vs. 62.5 and 82.6% for clinical stages 0 and I/II, respectively; P<0.01; Fig. 1D). Additionally, older patients (≥67 years old) were revealed to experience a poorer 5-year survival rate [41.7 vs. 80.2 and 74.5% for young-(≤46 years old) and medium-age patients (47–66 years old), respectively; P=0.085, Fig. 1E], although there was no significant difference in disease-free survival (Fig. 1F).

Figure 1. Survival curves based on the various clinicopathological features of patients with anal cancer. (A) Overall survival and (B) disease-free survival of patients with HSIL, adenoca and SCC. (C) Overall survival and (D) disease-free survival based on clinical stage. (E) Overall survival and (F) disease-free survival based on patient age at diagnosis. HSIL, high-grade squamous intra-epithelial lesion; SCC, squamous-cell carcinoma; adenoca, adenocarcinoma; y.o., years old.

Immunohistochemistry findings

Table I presents the clinicopathological features of the patients as associated with RKIP expression. High RKIP expression was associated with non-Caucasian patients (37.5 vs. 7.5% in Caucasians; P=0.019) and younger (≤46 years old) patients (33.3 vs. 0.0% of older patients; P=0.047). Additionally, high RKIP expression was associated with lower tumor clinical stages (50% in stage 0 vs. 6.7% in stages III/IV; P=0.038). Finally, anal HSILs (37.5%) exhibited the greatest percentage of high RKIP expression, followed by anal SCCs (11.5%) and adenocarcinomas (0.0%) (P=0.037; Table I, Fig. 2). There was no significant difference in overall- or disease-free survival between RKIP-high and RKIP-low expressing patients (Fig. 3A and B). Similarly, following the stratification of patients by tumor type (HSIL and SCC), there was no significant difference in survival based on RKIP expression (Fig. 3C and D). All the patients with adenocarcinoma demonstrated low tumor RKIP expression.

Figure 2. (A-C) Immunohistochemical analysis of RKIP in anal cancer samples. (A) High expression in high-grade squamous intra-epithelial lesions (magnification, ×400). (B) High expression in squamous-cell carcinomas (magnification, ×400). (C) Low expression in adenocarcinomas (magnification, ×400). (D) High expression in normal stomach tissues, which were used as positive controls (magnification, ×400). RKIP, Raf kinase inhibitor protein.

Figure 3. Survival curves based on RKIP expression groups in anal cancer samples. (A) Overall survival and (B) disease-free survival based on the RKIP expression in all samples. (C) Overall survival of patients with HSIL based on RKIP expression. (D) Overall survival of patients with SCC based on RKIP expression. RKIP, Raf kinase inhibitor protein.

RKIP expression was correlated with HPV16 infection, KRAS and BRAF mutation status, and the expression of β-globin, p16, Ki-67, MCM and TOP2A, as detected by immunohistochemistry. There were no positive cases of HPV18. All 30 cases that were analyzed for KRAS mutation were wild type, and only 1 case of the 35 analyzed was positive for a BRAF V600E mutation. A decreased percentage of HPV16+ cases were identified with high expression of RKIP, compared with the HPV16− cases (5.3 vs. 40%; P=0.013; Table II), indicating a possible correlation between HPV16 infection and RKIP expression.

Discussion

A number of previous studies have demonstrated that RKIP expression has prognostic value in several tumor types, including gastric, colorectal and gastrointestinal stromal tumors (17–19). However, other studies have identified a number of molecular distinctions between anal canal tumors and their colorectal counterparts (24,27). Therefore, the present study aimed to evaluate RKIP expression in invasive and in situ tumors of the anal canal, and to correlate findings with the clinicopathological data.

The results of the present study revealed a poorer overall survival rate in patients diagnosed with adenocarcinoma, compared with in those diagnosed with SCC or HSIL. An improved prognosis was expected in patients with HSIL, considering that these in situ lesions are associated with <30% recurrence following treatment (5,28). Additionally, the results of the present study demonstrated an improved survival rate in patients with SCC, compared with in patients with adenocarcinoma. This supports the findings of a previous study that compared the overall survival and prognosis of patients with anal adenocarcinoma, SCC and rectal adenocarcinoma, in which the authors demonstrated a poorer overall survival rate and prognosis for patients with anal adenocarcinoma (28). This same trend was described by our group in a previous study (24). Furthermore, when considering patient age, the present study identified poorer survival rates in patients ≥67 years old compared with in younger patients (≤47 years old), supporting the results presented by a prior study that revealed younger patients experienced improved overall- and disease-free survival rates compared with older patients with anal cancer (29). In addition, the present study did not identify high RKIP expression in the tumors of any older patients.

RKIP is considered a tumor suppressor gene, and its expression may prevent RAF/MEK/ERK signal transduction (30), thus serving a function in the inhibition of cancer development. The present study identified an increased percentage of high RKIP expression in patients with in situ lesions (HSIL) than in patients with invasive lesions (SCC and adenocarcinoma). Notably, an increased percentage of patients at clinical stage 0 exhibited high RKIP expression, compared with patients with higher clinical stages (I/II or III/IV). Considering the patient stratification was dependent on RKIP expression, the lack of receiver operating characteristic analysis may be a limitation of the present study; however, an RKIP scoring system was used according to our previously published studies (26), a method that is widely used for the assessment of immunohistochemistry staining (31), and allowed for inter-study comparisons.

Infection with high-risk HPV (HR-HPV16 and HR-HPV18) is considered the major risk factor for anal cancer development (4). Additionally, a previous study demonstrated that HPV−/p16− patients experienced poorer outcomes compared with HPV+/p16+ patients (20). Another study revealed the relevance of p16 expression, reporting an association with improved overall- and disease-free survival rates (6); however, the results of the present study did not identify any correlation between the expression of RKIP and of p16, which may explain the lack of correlation between RKIP expression and survival. In the cohort of the present study, a decreased number proportion of HPV16+ patients exhibited high RKIP protein expression compared with HPV16− patients. Hu et al (32) analyzed RKIP expression via immunohistochemical analysis in normal and cancerous cervical tissues and revealed high RKIP expression in normal tissue, low expression in primary cancer and the lowest (or absent) levels of expression in metastatic disease. Considering that HR-HPV infection is a fundamental step for cervical cancer development (9,10), there may be an inverse association between HR-HPV infection and RKIP expression, as was observed in the present study cohort. However, more studies are required in order to elucidate the molecular basis of this finding.

In conclusion, to the best of our knowledge, this is the first study to investigate RKIP expression levels in a set of anal tumors. The results of the present study demonstrated that high RKIP expression is present in lesions with clinical features that are generally associated with a good prognosis, including lower age at diagnosis, in situ lesions and lesions of a lower clinical grade, and that HPV16 infection may affect RKIP expression; a finding that requires further investigation.

Acknowledgements

The present study was supported by the São Paulo Research Foundation (grant nos. 2010/16795-4 and 2011/08523-7) and Ministry of Science, Technology, Innovation and Communication (grant no. MCT/FINEP/CT-INFRA-PROINFRA 01/2011).

References