Prognostic Significance of HER2 and MLH1 Protein Expression in Colorectal Cancer

Nair CP , Gajjar KK , Vora HH and Ghosh NR

Published Date: 2018-10-25

Nair CP1, Gajjar KK2, Vora HH3 and Ghosh NR4*

1Department of Cancer Biology, The Gujarat Cancer & Research Institute (GCRI), Asarwa, Ahmedabad, Gujarat, India

2Tumor Biology Lab, The Gujarat Cancer & Research Institute (GCRI), Asarwa, Ahmedabad, Gujarat, India

3Immuno-hematology Lab, The Gujarat Cancer & Research Institute (GCRI), Asarwa, Ahmedabad, Gujarat, India

4Department of Cancer Biology, The Gujarat Cancer & Research Institute (GCRI), Asarwa, Ahmedabad, Gujarat, India

*Corresponding Author:
Nandita Ghosh
Department of Cancer Biology
The Gujarat Cancer & Research Institute
NCH Compound, Asarwa, Ahmedabad-380016, Gujarat, India
Tel: +919879391224
E-mail: nandita.ghosh@gcriindia.org

Received date: October 15, 2018; Accepted date: October 23, 2018; Published date: October 25, 2018

Citation: Nair CP, Gajjar KK, Vora HH, Ghosh NR (2018) Prognostic Significance of HER2 and MLH1 Protein Expression in Colorectal Cancer. J Neoplasm. Vol.3 No.2:34. doi:10.21767/2576-3903.100034

Visit for more related articles at Journal of Neoplasm

Abstract

In colorectal cancer (CRC), numerous studies showed HER2 overexpression and amplification as a therapeutic and prognostic target. Further, it was suggested that genetic instability of mismatch repair gene MLH1 might increase negative predictive value of HER2 overexpression. Genetic or epigenetic inactivation of MLH1 gene is generally associated with loss of immunohistochemical expression of the corresponding protein. Hence, present study aimed to investigate the prevalence of HER2 and MLH1 protein expression by immunohistochemistry in untreated CRC patients (N=82) and further to examine their association with various clinicopathological variables and prognosis. Cytoplasmic HER2 and MLH1 protein expression was found in 31% and 29% of patients, respectively. With regard to clinicopathological parameters, incidence of MLH1 positivity was significantly higher in patients with adenocarcinomas as compared to mucinous adenocarcinomas (p=0.021). In relation to survival analysis, HER2 negative expression was associated with unfavorable RFS (p=0.074) and OS (p=0.027) as compared to HER2 positive expression in total patients. Further a trend of reduced OS was observed with HER2 negative expression in subgroups of patients with early stage (p=0.082), colon cancer (p=0.099) and rectal cancer (p=0.098). Similar trend of reduced RFS (p=0.072) and OS (p=0.083) was noted with HER2 negative expression in patients treated with adjuvant therapy. However, MLH1 protein expression failed to show any prognostic or predictive value. Further, significant positive correlation was observed between HER2 and MLH1 protein expression (p<0.001). In conclusion, absence of cytoplasmic HER2 protein expression could be a useful biomarker to identify high risk group of CRC patients with poor clinical outcome.

Keywords

HER2 protein expression; MLH1 protein expression; Colorectal cancer; Immunohistochemistry; Prognosis

Introduction

Colorectal cancer (CRC) ranks the third most common cancer and the fourth leading cause of cancer death worldwide, accounting for about 1.4 million new cases and nearly 700 000 deaths in 2012 [1]. Despite advances in chemotherapeutic treatment in CRC, it is still not capable of preventing recurrence in all patients [2]. Therefore, identification of new therapeutic targets is essential which can improve efficacy of treatment in CRC patients.

Compounds targeted against specific tumor proteins are under trial for many cancers. One of these targets is Human Epidermal Growth Factor Receptor 2 (HER2 ), which is primarily associated with breast cancer [2]. In CRC, recent research efforts have focused on the HER2 expression due to its proven importance in breast cancer. The HER2 receptor is a 1255 amino acid, 185 kD transmembrane glycoprotein located at the long arm of human chromosome 17 (17q12). The HER receptors exist as monomers on cell surface and may be activated upon dimerization with HER2 or other family members. Homo or heterodimerization results in the autophosphorylation of tyrosine residues within the cytoplasmic domain of the receptors and initiates a variety of signaling pathways, resulting in cell proliferation, survival, differentiation, angiogenesis, and invasion [3]. Further, in patients with breast and gastric cancer having HER2 positive tumors, HER2 -targeting agents such as trastuzumab, pertuzumab, and lapatinib are part of the standard therapy. Besides, in many epithelial malignancies such as cancer of the lung, prostate, bladder, pancreas, esophagus and stomach, over expression of HER2 has been reported. However, overexpression of HER2 in CRC shows a wide range of variability between 0-84% in different studies [4]. Further, conflicting reports exist about the prevalence of HER2 over expression in CRC as well as the relationship between HER2 over expression and the clinicopathological features. Moreover, a very scarce data is available for HER2 testing particularly in colorectal adenocarcinomas and use of HER2 therapy in the treatment of CRC patients has been less extensively investigated. Therefore, correlations of HER2 expression with clinicopathological features and its therapeutic and prognostic implications in CRC urgently require clarification.

Furthermore, recently, Sanguedolce et al. suggested that genetic instability of mismatch repair gene MLH1 (MutL homolog 1) and MSH2 (MutS homolog 2) might increase negative predictive value of HER2 overexpression [5]. Moreover, genetic or epigenetic inactivation of MLH1 gene is generally associated with loss of immunohistochemical expression of the corresponding protein. MLH1 is a protein in humans that is encoded by MLH1 gene located on short arm of chromosome 3 (3p22.2). One report suggested that immunohistochemical analysis of MLH1 and MSH2 expression is a rapid and accurate method for identifying large bowel tumors of MSI-H phenotype [6]. They observed loss of MLH1 in 90.9% of MSI-H carcinomas whereas all MSI-L and MSS (microsatellite stable) showed normal expression of MLH1. Recent studies suggest that immunohistochemical analysis of MLH1 gene product expression may identify MSI-H colorectal adenocarcinomas [7]. Moreover, several studies have demonstrated a better prognosis for patients with colorectal MSI-H carcinomas with respect to patients with MSI-L/MSS tumors [7,8]. Therefore, studying MLH1 protein expression along with HER2 expression may give additional information with regard to prediction of treatment response and prognosis in CRC patients. Hence, present study aimed to study the prevalence of HER2 and MLH1 protein expression and their association with clinicopathological parameters in CRC patients. Further, prognostic and predictive value of HER2 and MLH1 was evaluated.

Materials and Methods

Patients

A total 82 untreated histologically confirmed CRC patients at The Gujarat Cancer and Research Institute, Ahmedabad, India, between 2015 and 2017, were enrolled in this study. A detailed clinical history including tumor size, lymph node status, histological grade, disease stage, treatment given, etc. during follow up was obtained from case files maintained at the Medical Record Department of the institute. TNM classification with World Health Organization (WHO) Grading System was taken into consideration for pathological staging. Primary treatment offered to all patients was surgery or surgery followed by adjuvant chemotherapy and/or radiotherapy. The main chemotherapeutic treatment included were 5-FU and leucovorin, oral Capecitabine, or in combination with Oxaliplatin. Out of 82 patients enrolled, 78 patients who followed up for the period of 24 months or until death within that period, were included for Over-all survival (OS) analysis. For OS, 3 patients with stage IV cancer and one patient who lost to follow-up were excluded. Out of 78 patients, 15 patients with persistent disease were excluded from relapse–free survival (RFS) analysis. Thus, 63 patients were included for RFS analysis. Survival analysis was also performed in the subgroups of patients with early stage and advanced stage disease; as well as in colon cancer and rectal cancer patients after sub-grouping them according to tumor site. Further, to investigate the predictive value of HER2 and MLH1, survival analysis was also performed in patients treated with adjuvant therapy. The patient and tumor characteristics are shown in Table 1.

Characteristics N (%)
Age (years) (Range: 20-80 years) Median age: 56 years
<56 41 (50)
≥56 41 (50)
Gender
Female 37 (45)
Male 45 (55)
Tumor site
Colon 48 (59)
Rectum 34 (41)
Tumor size
T2 14 (17)
T3 62 (76)
T4 06 (07)
TNM stage
I 08 (10)
II 39 (48)
III 32 (39)
IV 03 (03)
Tumor differentiation
Well 13 (16)
Moderate 59 (72)
Poor 10 (12)
Histological differentiation
Adenocarcinoma 61 (74)
Mucinous adenocarcinoma 21 (26)
Necrosis
Absent 75 (92)
Present 07 (08)
Lymphatic permeation
Absent 66 (80)
Present 16 (20)
Vascular permeation
Absent 78 (95)
Present 04 (05)
Recurrence/Metastasis (N=63)
Absent 57 (90)
Present 06 (10)
Disease outcome (N=78)
Alive 62 (79)
Dead 16 (21)

Table 1:Patient and tumor characteristics.

HER2 and MLH1 protein expression by Immunohistochemistry (IHC)

For the study of HER2 and MLH1 protein expression, paraffin embedded tumor tissue blocks were retrieved from the histopathology department of the institute. Briefly, 4 μm thick sections were cut from the formalin fixed paraffin embedded tumor tissue blocks and were mounted on 3-aminopropyletriethoxy silane coated glass slides. The immunohistochemical staining was carried out using Anti-c-erbB-2 (HER2 /neu) mouse monoclonal antibody IgG1 (1:20 dilution, clone-CB11, Biogenex, Fremont, CA) and MLH1 mouse monoclonal antibody IgG1 (1:20 dilution, clone- G168-15, BD Biosciences Pharmingen) along with Mouse and Rabbit specific HRP/DAB (ABC) Detection IHC kit from Abcam, as per manufacturer’s protocol recommendations. Antigenicity was retrieved by heating the tissue sections in 10 mM tri-sodium citrate buffer (pH-6.0) solution for 20 minutes in a pressure cooker prior to application of the primary antibody. For antigen retrieval, the pressure cooker was filled with 1500 ml of distilled water and the cooker was placed on the conventional electric stove, covered with an unlocked lid, and heated to a boil. Then the IHC jar, having tissue sections in 10mM tri-sodium citrate buffer (pH-6.0) solution, was placed in the pressure cooker and tissues were boiled for 20 minutes at temperature ranging from 95 to 100°C. All sections were scored independently by two independent researchers in a blinded fashion. A semi-quantitative scoring method ranging from negative (no staining or, 10% of cells stained) to 3+ (1+staining in 11-30% of the cells: weak, 2+ staining in 31-50% of cells: moderate and 3+ staining in >50% of cells: intense) was used.

Ethics statement

Written consent of the patients who underwent surgery at the Department of Surgical Oncology was obtained, prior to primary tumor tissue collection. The study was approved by the Institutional Scientific and Ethical Review Committees.

Statistical analysis

The statistical data analysis was performed with the help of SPSS (Stastistical Package for the Social Sciences) software. Two tailed chi square (χ2 )test was used to determine the association of HER2 and MLH1 protein expression with clinicopathological variables. RFS and OS were calculated using Kaplan-Meier method and Log rank test. Correlation between two parameters was calculated using Spearman’s correlation coefficient (r) method. P value ≤ 0.05 was considered significant.

Results

Incidence of HER2 and MLH1 protein expression

The expression of HER2 and MLH1 protein was localized in cytoplasm of the epithelial cells of colon and rectal tumors.

HER2 positive protein expression was observed in 31% (25/82) of patients whereas 69% (57/82) of patients showed HER2 negative protein expression. Further, there was a preponderance of negative expression of MLH1 in 71% (58/82) of the patients and 29% (24/82) showed positive MLH1 protein expression. Representative pattern of HER2 and MLH1 protein expression in CRC patients is shown in Figure 1.

neoplasm-histochemical

Figure 1:Immuno-histochemical staining of HER2 and MLH1 in tumor tissues (40x).

Correlation of HER2 and MLH1 protein expression with clinicopathological parameters

A trend of higher HER2 positive expression was observed in ≥ 56 years age group patients (39%) as compared to <56 years age group patients (22%,χ2 = 2.820, r = +0.185, p = 0.095 ) Further, when the patients were sub grouped in four different age groups with a 15 years gap period (20-35, 36-50, 51-65 and 66-80), no such trend was noted. Moreover, a trend of higher HER2 positive expression was observed and in patients having absence of necrosis (33%) as compared to those having presence of necrosis (0%,χ2 = 3.357, r = −0.202, p = 0.068) . Further, a significant higher MLH1 positivity was bserved in patients with adenocarcinoma (36%) as compared to those with mucinous adenocarcinoma (9%;χ2= 5.316, r = −0.255, p = 0.021) and a trend of higher incidence of MLH1 positivity was observed in males (37%) as compared to females (χ2 = 5.223, r = +0.198, p = 0.074) . In addition, when MLH1 protein expression was correlated with four different age groups, an increasing trend of MLH1 positivity was noted with the increase in the age (χ2 = 5.223, r = +0.198, p = 0.074) . The observed MLH1 positivity was 0% (20-35 years), 30% (36-50 years), 32% (51-65 years), and 42% (66-80 years).

Intercorrelation between HER2 and MLH1 protein expression

A significant positive correlation was observed between HER2 and MLH1 protein expression (RFS : 0%,0 / 21;OS : 4%,1/ 23) .

Survival analysis for RFS (N=63) and OS (N=78) in relation to HER2 and MLH1 protein expression in CRC patients

In total patients, Kaplan-Meier univariate survival analysis showed that patients having negative HER2 expression was associated with a trend of reduced RFS (14%,6 / 42; Log rank = 3.195,df =1, p = 0.074) Figure 2a and a significant reduced OS Figure 2b as compared to those with positive HER2 expression (RFS : 0%,0 / 21;OS : 4%,1/ 23). Further, according to disease stage, in early stage patients, a trend of high incidence of death was noted in HER2 negative patients as compared to those with HER2 positive patients (0%,0 /13; Log rank = 3.021,df =1, p = 0.082) . Moreover, according to tumor site, a trend of unfavorable OS was observed in colon cancer patients with negative HER2 expression (21%,7 / 34) as compared to those with positive HER2 expression (0%,0 /12;Log rank = 2.726,df =1, p = 0.099) . Further, a trend of unfavorable OS was also noted in rectal cancer patients with negative HER2 expression (32%;8 / 21) as compared to those with positive HER2 expression (9%,8 / 21; Log rank = 2.724,df =1, p = 0.098). However, MLH1 protein expression showed no significant association with RFS and OS in total patients or in any of the subgroups.

neoplasm-survival

Figure 2:Survival analysis in total CRC patients (a) A trend of reduced RFS and (b) A significant reduced OS in patients with negative HER2 expression as compared to those with positive HER2 expression.

Survival analysis for RFS (N=54) and OS (N=64) in relation to HER2 and MLH1 protein expression in CRC patients treated with adjuvant therapy

In total CRC patients treated with adjuvant therapy, Kaplan- Meier univariate survival analysis demonstrated that patients with negative HER2 expression showed a trend of reduced RFS (17%,6 / 36;Log rank = 3.231,df =1, p = 0.072) and a trend of reduced OS as compared to those with positive HER2 expression (RFS :0%,0 /18;OS : 5%,1/19) . Further, MLH1 expression showed no significant association with RFS or OS in patients treated with adjuvant therapy.

Discussion

HER2 has been shown to be an effective target for adjuvant therapy for breast cancer and gastric cancer patients. Currently the value of HER2 status in colorectal cancer is also being explored and a great degree of variation has been reported on detection of HER2 over expression in CRC by IHC [9]. In present study, we investigated the prevalence of HER2 over expression using Biogenex CB11 monoclonal antibody in cohort of 82 CRC patients. Only cytoplasmic staining of HER2 protein was found in 31% of the studied patients. In contrast, Dursun et al, using Biogenex CB11 reported 14% membranous staining and no cytoplasmic staining in colorectal neoplasia [10]. Kavanagh et al. detected HER-2 overexpression in 11% of CRC patients and showed membranous staining [11]. Further, Kountoutakis et al. and Gill et al. observed cytoplasmic staining along with membranous staining using NCL-CB11 monoclonal antibody, where the former found 22.64% HER2 positivity in colorectal adenocarcinoma and the latter found 65% HER2 positivity in colon carcinoma [12,13]. Schuell et al. using different HER2 antibody also reported similar incidence with 29.9% of 77 CRC patients showing HER2 immunopositivity [14]. Kruszewski et al. observed 24% of HER2 positive protein expression in a cohort of 50 CRC patients [15]. In addition, some studies in CRC have also reported incidence of HER2 positivity as low as 0.5% [16] to 11% [11]. The debate about such a wide range of HER2 expression might be attributed to several causes, such as primary antibody difference, a difference in scoring systems, a difference of sample size, racial differences, and heterogeneity of study population. The unique finding of the present study is that it showed only cytoplasmic immunostaining for HER2 expression as compared to other studies showing HER2 membranous and/or cytoplasmic staining. This result of the current study might suggest that probably the membrane bound HER2 was internalized into the cytoplasm upon activation or HER2 was activated internally via homodimerization [17]. Further, half et al. found that cytoplasmic HER2 was localized near rough endoplasmic reticulum (rER) [18], which led them to hypothesize that HER2 is likely to be derived from original peptide synthesis and thus cytoplasmic HER2 expression would be of genomic origin.

In present study, a trend of higher incidence of HER2 positivity was observed in older age group as compared to younger age group. Gill et al. and Kountourakis et al. also observed significantly higher immunopositivity of HER2 in older age patients in colon and colorectal carcinoma, respectively [13,12]. Additionally, although not significant, present study showed association between HER2 positive expression and absence of necrosis. One study in CRC showed that HER2 positivity was correlated with perineural invasion, vascular invasion, lymph node metastases, and higher TNM stage [19]. Tavangar et al. and Antonacopoulou et al. observed higher incidence of HER2 positivity in advanced stage patients as compared to early stage CRC patients [20,21]. On the other side, Schuell et al. and Sayadnejad et al. showed no significant association of HER2 expression with clinicopathological parameters in CRC patients [14,22].

Assuming that cytoplasmic HER2 has a prognostic role in CRC, it might also be involved in tumor pathogenesis like membranous HER2 in breast cancer. Hence, the current study evaluated the prognostic significance and observed that HER2 positive patients had a significantly improved overall survival (p=0.027) and recurrence free survival (p=0.074) time than their respective counter parts. Additionally, the subgroups of early stage, colon cancer and rectal cancer patients with HER2 positive tumors, displayed a trend towards better disease outcome. In accordance, although not significant, Tu et al. also observed higher survival rates of HER2 positive patients as compared to HER2 negative patients in CRC [23]. These results are however in contrast with other studies in which a negative correlation of HER2 positivity was observed with survival [24]. However, Schuell et al. and Kruszewski et al. observed no correlation between HER2 expression and survival in CRC patients [13,14]. Kavanagh et al. also demonstrated that overexpression of HER2 was not a predictor of disease-free and overall survival in CRC patients [11].

Presence of intracellular HER2 might have therapeutic implications in treatment with targeted therapy. Hence, present study evaluated the utility of HER2 in patients treated with adjuvant therapy, and, observed a trend of better RFS and OS with positive HER2 expression suggesting that HER2 immunopositivity could be a useful prognosticator in predicting better treatment response in CRC patients. There are reports that if cytoplasmic HER2 is actively involved in colorectal carcinogenesis, the administration of lapatinib or other intracellular HER2 -targeting compounds could be a breakthrough in the treatment of especially CRC those who have failed to respond to adjuvant chemotherapy [17]. However, a major limitation of the current study is that mutational results of KRAS and BRAF have not been included and that HER2 is responsible for cell survival and proliferation via signaling through the RAS–RAF–ERK and PI3K–PTEN–AKT pathways [25]. So, present results were unable to elucidate the exact relationship of HER2 expression with therapeutic response in CRC. Moreover, we found that HER2 protein was indeed present in patients with CRC but very few cases showed strong expression enough to consider for its therapeutic target.

Now, it is an established fact that investigation of MSI status in CRC is warranted for three reasons: first, as a potential screening tool for HNPCC, second as a prognostic marker and, finally, as a potential predictor of response to chemotherapy. Therefore, to improve understanding of CRC further, investigation of mismatch repair gene MLH1 could be important, which is most often known to arise from epigenetic silencing in sporadic cancer. Hence, present study focused on exploring the prevalence of MLH1 protein expression and its prognostic value in CRC patients.

In current study, only 29% of 82 cases showed positive cytoplasmic MLH1 expression while 71% of the patients lacked MLH1 expression. Lack of protein expression as a surrogate for MSI is clinically important to patients. Cytoplasmic staining in the absence of nuclear staining for MLH1 has been previously described in only 1 CRC case among the studied patients [26]. Moreover, in a study conducted on endometrial cancer patients by Berends et al, immunohistochemical staining in three patients showed clear cytoplasmic, but no nuclear staining for MLH1 in tumor cells while in normal cells, a normal nuclear staining pattern was observed. Apart from the possibility of a technical artifact, it would be hypothesized that the MLH1 protein was incompletely synthesized in the cytoplasmic ribosomes due to a mutation (as yet undetected) in the MLH1 gene that prevents transport to the nucleus [27].

In relation to clinical parameters, in current study a trend of higher incidence of negative MLH1 expression was observed in females as compared to male patients. Similar correlation of MLH1 expression with gender was obtained in a study on CRC patients conducted by Soreide et al. [28]. However, Kishi et al, in 97 surgical specimens after chemotherapy, observed that MLH1 expression did not show any correlation with clinicopathological features, such as gender, age, tumor location, histological grade, tumor length, tumor depth, and lymph node metastasis in esophageal cancer [29]. Further, present study showed that negative MLH1 expression was significantly higher in mucinous adenocarcinoma as compared to adenocarcinoma (p=0.021). It is known that mucinous histological types of tumors are usually of aggressive phenotype compared that of adenocarcinomas. Hence, the finding probably indicates that loss of MLH1 protein expression could be useful in identifying patients with aggressive phenotype.

Few studies are available regarding prognostic value of MLH1 expression in CRC. Present study reported better OS and RFS in patients with negative MLH1 expression; however, the data was not statistically significant. In accordance, recently Huang et al. demonstrated that the 5-year-survival rate was significantly lower for MLH1 overexpression (77.5% versus 94.4%, p=0.039) in colon cancer patients [30]. Kruschewski et al. showed that significant differences for overall and recurrence-free survival were not seen for patients with MLH1-negative carcinomas, although there was a tendency for longer overall survival (72 vs. 63 months) [31]. Moreover, aberrant MMRP expression may have an effect on the prognosis of stage III CRC patients [32]. Further, Wilczak et al. described that MMR (MSH6, MLH1 and PMS2) overexpression was linked to poor outcome in prostate cancer. It is interesting that MSI positive (MLH1 negative) CRCs seem to have a better prognosis, because they are often less differentiated and larger than MSI negative cases (features high proliferation and low survival) [33]. The reasons for the putative survival advantage are not known, but plausible explanations include a selfdestructive effect of numerous mutations accumulating in the cell genome, possibly mutating genes necessary for the viability of the malignant clone. Resulting mutant proteins may also be transferred to the cell membrane to evoke an immune reaction against the tumor [34].

In a recent report, it was observed that simultaneous genetic instability of MLH1 and MSH2 might increase the negative predictive value of HER2 overexpression [5]. Hence, present study for the first time evaluated intercorrelation between HER2 and MLH1 protein expression in CRC and observed significant positive correlation between them. Additionally, patients coexpressing both HER2 and MLH1 proteins had significantly better disease outcome as compared to those with absence of both the proteins (data not shown). Thus, it suggests that coexpression of HER2 and MLH1 could be useful in discriminating high/low risk group of CRC patients.

Conclusion

In conclusion, absence of cytoplasmic HER2 protein expression could be useful biomarker to classify high risk group of CRC patients with poor clinical outcome. Although MLH1 protein expression alone had no prognostic value in studied patients, combine use of HER2 and MLH1 protein expression may probably emerge as a biomarker to predict clinical outcome in CRC patients.

References

open access journals, open access scientific research publisher, open access publisher
Select your language of interest to view the total content in your interested language

Viewing options

Flyer image
journal indexing image

Share This Article