Significance of Her2neu, p53, and Ki67 expression in invasive urothelial carcinoma

INTRODUCTION

Bladder cancer (BC) is one of the most common urological malignancies in males after prostate cancer. Numerous environmental and occupational hazards are considered risk factors for BC.^[1] The risk factors include smoking, schistosomiasis infection, and occupational hazards associated with petroleum products, rubber, paint, and dyes.^[2] Urothelial carcinoma (UC) are divided into three categories based on the presence of metastasis and on the invasion of the bladder wall: Muscle invasive BC (MIBC), that grows in the lamina propria and invades the muscle layer; non-MIBC, which is limited to the urothelium and does not invade the deep muscle layer and unresectable locally advanced or metastatic UC, when it has spread to the abdominal or pelvic wall, to regional lymph nodes, or to distant sites.^[3]

Prognostic factors in UC are the degree of differentiation of the tumor, divergent differentiation of UC, and the depth of invasion into the bladder wall. Invasion of the deep muscularis propria is considered to cause a major decrease in the 5-year survival rate to about 30%.^[4] It is difficult to predict an accurate prognosis based on a single factor, mostly due to the huge variation in inter-observer reporting of grade and stage. Hence, the role of genetic and protein markers as more reliable prognostic factors needs to be researched further.^[5]

Her2neu is a potential therapeutic target for UC. Although Her2neu gene amplification and overexpression are reported in UC, their prognostic and predictive value remains uncertain. Tumor proliferation measured by Ki67 is linked to stage progression and tumor recurrence in UC.^[6] Bladder tumor progression is associated with alterations in the p53 pathway, and p53 overexpression can provide prognostic information and guide further management. Previous studies have reported overexpression of p53 to be an independent predictor of tumor progression and reduced survival.^[7] Correlation of Her2neu, Ki67, and p53 with various clinicopathological parameters such as age, smoking status, histological subtype, grade, and stage of tumor has yielded varying results in the literature. The majority of studies have found a positive correlation between these immunohistochemistry (IHC) expressions and tumor stage and grade. However, few studies show a statistically significant association with grade only. Furthermore, clinical parameters such as age, smoking status, and histological subtype show a positive association only in a handful of cases.^[5,7] Discrepancies in assessment and evaluation, and the cutoff used for these three IHC, are also paramount to correlate accurately.

Thus, the present study aims to address this divergence in the literature and correlate the immunohistochemical expression of Her2neu, p53, and Ki67 markers with predictive clinicopathological parameters in invasive UC.

MATERIALS AND METHODS

This study was conducted at the department of pathology in a tertiary centre in East India. This was a prospective and retrospective observational pilot study. The duration of the study was 18 months. A total of 56 transurethral resection of bladder tumor specimens and four radical cystectomy specimens of invasive UC were studied. Patient consent was taken in all cases, and ethical clearance was granted by the Institutional Ethical Committee. All the cases were categorised into low or high grade based on the World Health Organization classification of Urothelial tumors, 5^th edition. Pathological staging of these tumors was performed as per the TNM AJCC 8^th edition. IHC for p53, Ki67, and Her2neu was performed in all these cases using the manual peroxidase-antiperoxidase method. Non-invasive urothelial neoplasms and benign urothelial lesions were taken as the control population. Primary antibodies for HER2 were DAKO (Monoclonal Mouse Anti-Human Ks 20.8), for p53 are Biogenex (Monoclonal Mouse Anti-Human D07), and for Ki67 are Biogenex (Monoclonal Mouse Anti-Human MIB-1). It was ensured that control for each IHC was run satisfactorily before interpretation.

IHC expression of Her-2-neu, p53, and Ki67 was correlated with clinicopathological parameters such as age, sex, smoking history, grade, and stage of tumor. Statistical analysis was performed using SPSS software version 11.5 (SPSS, Chicago, IL, USA), and the Chi-square test was used to compare the variables. A p <0.05 was taken as statistically significant.

Inclusion criteria

All histologically confirmed cases of invasive UC were included in the study [Figure 1].

Close

Figure 1: (a) Low grade lamina invasive urothelial carcinoma. Arrowhead highlights sheets of tumour cells [Hematoxylin & eosin, (H&E, 200×)]. (b) High grade deep muscle invasive urothelial carcinoma. Blue arrows highlight sheets of high-grade tumour cells. Black circle highlights the deep muscle layer (H&E, 200×).

Export to PPT

Interpretation of immunostaining

Her2neu

A score of 0 (absent staining or weak, incomplete membranous staining in ≤10% of tumor cells) or 1+ (weak, incomplete membranous staining in >10% tumor cells) was taken as negative. A score of 2+ (weak-to-moderate complete membranous staining ≥10% of tumor cells or strong membranous staining in <10% of tumor cells) was taken as equivocal, and a score of 3+ (strong complete membranous staining in >10% tumor cells) was considered positive [Figure 2].

Close

Figure 2: (a-d) Her2neu immunohistochemistry scoring performed by ASCO/CAP guidelines: (a) Score 0 (Negative) (200×), (b) Score 1+ (Negative) (200×), (c) Score 2+ (Equivocal) (200×), (d) Score 3+ (Positive) (200×). ASCO/CAP: The American Society of Clinical Oncology and the College of American Pathologists

Export to PPT

P53

Tumors with patchy nuclear expression were considered as p53 wild type. Tumors with absent or complete nuclear expression were considered p53-mutated type [Figure 3].

Close

Figure 3: p53 immunohistochemistry expression: (a) Wild type (Negative) (200×), (b) Null phenotype (Positive) (200×), (c) Mutated phenotype (Positive) (200×).

Export to PPT

Ki67

A low Ki67 nuclear expression of ≤20% in tumor cells was defined as negative, and a high Ki67 nuclear expression of >20% was positive [Figure 4].

Close

Figure 4: (a and b) (200×) Ki67 immunohistochemistry expression: (a) Low Ki67 expression (Negative) (200×), (b) High Ki67 expression (Positive) (200×).

Export to PPT

RESULTS

The age of the study population ranged 36-84 years. The majority of the patients were ≥60 years of age (66.66%; 40/60). The mean age of patients was 61 years. 96.66% of the study population were male (58/60), and 50% (30/60) of the patients were smokers. Most of the cases (73.33%; 44/60) were high-grade UC, and 26.66%; 16/60 were low-grade UC.

DISCUSSION

Her2neu plays a pivotal regulatory role in cell growth, differentiation, and survival. The significance of Her2neu overexpression has been demonstrated in breast and gastric cancers. It can be a potential therapeutic target in UC. However, there are conflicting results regarding the oncogenic capacity of Her2neu overexpression in UC. In our study, there was no significant association between Her2neu expression and the age of the study population or their sex. This was in concordance with studies by Ibrahim et al.,^[1] Rosli et al.,^[8] and Sarwala et al.^[9] Furthermore, there was no significant association between smoking status and Her2neu expression. Very few studies have drawn a correlation between the smoking status of patients and Her2neu expression. One of the notable studies is by Sarwala et al., who noted a significant association with p <0.0001.^[9] A statistically significant association was noted between the grade of invasive UC and Her2neu expression in our study (p <0.0001). However, there was no significant correlation between stage and Her2neu expression in tumors. There is conflicting data in the literature in this regard. In a retrospective study of 39 cases of invasive UC, Khaled El Gehani et al.^[10] noted a significant association between Her2neu expression and tumor stage, but no association with tumor grade.^[10] The majority of studies showed results like those observed in our study.^[11-13] Few other studies showed a significant association between Her2neu and pathological stage as well as tumor grade.^[14-16] These heterogeneous data on the prognostic significance of Her2neu may also be attributed to the difference in IHC antibodies used and the lack of a standardized scoring criterion.

p53 expression in invasive UC was correlated with age, sex, and smoking status of the study population. There was no significant association with any of these parameters. This agrees with many previously published studies.^[17-20] We observed that p53 expression has a significant association with tumor grade and no association with pathological stage. Our studies were in concordance with studies by Vasudha and Mahadevappa^[17] and Wu et al.^[19] Studies by Koyuncuer et al. and Missaoui et al. showed conflicting results. These authors observed a significant correlation of p53 expression with tumor stage and not with tumor grade.^[7,21] Vetterlin et al. showed p53 positivity in 43/83 (51.8%) high-grade cases and 1/18 (5.55%) low-grade cases (p = 0.001), which was similar to our study.^[18] Mohapatra et al. found no significant association between tumor stage and p53 expression (p = 0.154).^[20] In a study of 110 UC cases, Thakur et al. observed high p53 expression in 72.4% (42/58) of pT1 cases, 73.9% (17/23) of pTa cases, and 86.2% of pT2 cases, which showed a significant association of p53 expression with tumor stage.^[22] It is suggested that p53 aberrant expression has a poor prognosis in UC. In the molecular pathogenesis of UC, p53 alteration is characteristic of the tumorigenesis of high-grade muscle-invasive carcinoma.^[23] The conflicting results may be due to employing different techniques for p53 testing and varied criteria for interpretation of p53 immunostaining. In a study of 413 cases of UC, Asgari et al.^[24] concluded that high-grade and low-grade UC can be differentiated based on intensity and percentage of cells staining for p53.^[24]

There was no significant correlation observed between Ki67 expression and the age of patients in our study. Similar findings were noted in the past studies conducted by Ibrahim et al. (p = 0.73), Vetterlein et al. (p = 0.9), and Jawad et al. (p >0.005).^[1,18,12] In the present study, Ki67 showed no significant association with the sex of the patient (p = 0.53). Many previously published studies agreed with our observation.^[12,19,25] We observed no correlation between the smoking status of patients and Ki67 expression, and neither did any other study in the literature. We observed a significant association between the grade of tumor and Ki67 expression; however, there was no significant association noted between Ki67 expression and the stage of tumor. Tyagi et al. showed that high Ki67 was associated with high-grade tumors, and the finding was statistically significant.^[25] Hasan et al. studied 62 cases, of which 34 (54.8%) cases were high-grade and 28/62 (45.2%) cases were low-grade UCs. They found a significant association between high Ki67 and high-grade tumors (p <0.005).^[26] Jawad et al. observed that high-grade UC exhibited more Ki67 expression than low-grade tumors (p = 0.014).^[12] Contrary to the results of our study, Sarwala et al. and Missaoui et al. showed that Ki67 expression was significantly associated with the pathological stage of the tumor (p <0.0001 and p <0.001, respectively).^[9,21] However, a study by Hasan et al.,^[26] showed no significant association between Ki67 expression and pathological tumor stage.^[26]

Our study showed a significant association between Her2neu and Ki67 expression (p = 0.0007). Ibrahim et al.,^[1] showed no significant association in this regard.^[1] According to Sarwala et al., coexpression of both markers correlated well with tumor grade and muscle invasion. Coexpression was superior to a single marker to predict tumor aggression, progression, and prognosis.^[9] In our study, there was a significant association between the expression of p53 and Ki67 with p <0.0001. Wang et al.,^[27] noted that there was a significant correlation between the expressions of Ki67 and p53.^[27] The evaluation of Ki67 and p53 immunoexpression by Thakur et al.^[22] reported a significant association with tumor stage and histological grade individually as well as in combination.^[22] In the present study, there is a significant association between Her2neu and p53 expression (p = 0.004). This agreed with Ponnaboina et al., who studied 88 suspected cases of UC.^[28] The authors acknowledge the small sample size, short duration of study, and lack of long-term patient follow-up. Hence, larger prospective studies will help validate current observations.

CONCLUSIONS

Tumor grade is a major prognostic factor in UC; however, lack of a clear cutoff for distinguishing low- from high-grade carcinomas has led to variability in application of the criteria and sometimes poor interobserver reproducibility. In our study, Her2neu, p53, and Ki67 overexpression were significantly associated with tumor grade, with the overexpression more prevalent in high-grade UC, reflecting the aggressiveness of the tumor cells. Hence, these antigens have the potential to function as surrogate markers for tumor grade. Coexpression of p53, Her2neu, and Ki67 can be used as a potential prognostic marker.