Diagnostic Usefulness of p53 Immunostaining in Gastric Cancer and Dysplasia: A Real-world Clinical Experience

Abstract

Background/Aim

Gastric cancer and its precancerous lesions represent a significant public health concern. A subset of gastric cancers exhibits mutations in the TP53 gene, often accompanying distinctive morphologic alterations. This study aimed to assess the diagnostic efficacy of p53 immunostaining in real-world clinical settings.

Patients and Methods

A retrospective analysis was conducted on 50 cases of gastric tumors and tumor-like lesions, wherein p53 immunostaining played a pivotal diagnostic role. The staining pattern of p53 was examined in conjunction with clinicopathologic parameters.

Results

Mutant p53 staining pattern demonstrated a significant association with high-grade nuclear atypia (p<0.001), high-grade dysplasia, and tubular adenocarcinoma (p<0.001), as well as microsatellite instability status (p=0.034). Furthermore, the diagnostic utility of p53 immunostaining was evident in scenarios where: 1) biopsy specimens contained few tumor cells, 2) pathologic evaluation of resection margins was limited by cauterization artifacts, and 3) distinction between low-grade and high-grade gastric dysplasia was challenging.

Conclusion

P53 immunostaining can be helpful for the diagnosis of gastric tumor and tumor-like lesions, and accurate pathologic margin evaluation, particularly in lesions demonstrating intestinal-type differentiation and some degree of nuclear atypia.

Gastric cancer is a common cause of death worldwide. In South Korea, gastric cancer has been one of the most common cancer types. Its annual incidence per 100,000 persons was 59.7 in 2016 and age-adjusted death rate was 9.6 per 100,000 individuals in 2017 (1). Despite the recent progress in precision medicine, targetable genetic alterations specific to gastric cancer have not been found yet (2). Gastrectomy or endoscopic removal of gastric cancer and its precursor lesion is still the gold standard of treatment (3). Accurate histologic examination and margin evaluation are thus critical for patient care.

The Cancer Genome Atlas (TCGA) project have molecularly classified gastric cancer into four subtypes: EBV-positive, microsatellite-unstable, genomically stable, and chromosomally unstable (4). Of note, the chromosomally unstable molecular subtype is strongly associated with the intestinal subtype by Laurén histological classification. The chromosomally unstably subtype, histologically intestinal subtype, frequently harbors genetic alteration of TP53 (5).

TP53, located in chromosome 17p13.1 (6), is a tumor suppressor gene that regulates DNA repair, apoptosis, cell cycle progression, and cellular senescence (7). It is the most frequently altered gene in malignancy and a key molecule in carcinogenesis (7,8). Identification of TP53 mutation can be thus used for the classification of tumors (9), prediction of cancer recurrence (10-12), and prognostic stratification (13-16).

Meanwhile, the evaluation of nuclear and structural atypia is the cornerstone of histologic examination of tumors. Incorporation of the degree of nuclear and structural atypia to tumor grading has been applied to multiple organs, including breast (17), prostate (18), and stomach (19,20). Of note, genetic alteration of TP53 frequently accompanies morphologic changes of tumor cells. Usually, TP53-altered tumor cells demonstrate histologic changes, such as chromatin condensation, hyperchromasia, and nuclear size enlargement (21-24).

Although next-generation sequencing is currently used in many hospitals to detect the targetable alterations, it requires large amount of DNA input, prolonged turnaround time, and higher costs compared to other molecular tests (25,26). Immunohistochemistry is still one of the most important diagnostic tools in daily practice due to its convenience, fast turnaround time, and relatively lower-cost (27,28).

This study investigated the diagnostic utility of p53 staining for the evaluation of gastric cancer and dysplasia.

Patients and Methods

Study population. We performed a retrospective cohort study on histologically confirmed cases of gastric cancer (n=915), gastric dysplasia (n=431), and indefinite for dysplasia (n=13) [with 541 biopsy, 480 endoscopic submucosal dissection (ESD)/endoscopic mucosal resection (EMR), and 338 surgical procedures]. In total, 50 cases in which p53 staining was used for diagnostic purposes from 2021 to 2022 at the Department of Pathology of Kyungpook National University Chilgok Hospital were included in this analysis. We mainly performed p53 staining on gastric tumors and tumor-like lesions showing intestinal-type differentiation, with some degree of nuclear atypia, based on the result of TCGA gastric cancer classification and previous studies (4,29,30). The clinicopathological data, including age, sex, tumor location, microsatellite instability (MSI) test, and Epstein-Barr Virus Encoded RNA (EBER) in situ hybridization result, of the cases were retrieved from the electronic medical records of the hospital. The study was conducted in accordance with the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Kyungpook National University of Chilgok Hospital (No. KNUCH 2024-02-015). The requirement for written informed consent from the patients was waived because of the retrospective nature of the study.

Pathological evaluation. Biopsy and surgical samples underwent fixation in 10% neutral-buffered formalin before being embedded in paraffin blocks. The blocks were then sliced into sections measuring 4 μm in thickness and subsequently stained using hematoxylin and eosin. The histologic review of these slides was carried out by two gastrointestinal pathology specialists (MSK and ANS) independently. The diagnoses and classification of gastric tumors and tumor-like lesions were determined based on the criteria outlined in the fifth edition of the World Health Organization Classification of Tumors (31). Nuclear atypia was categorized into low-grade or high-grade, based on the previous studies (20,32). Ambiguous cases showing nuclear atypia between low-grade and high-grade were designated as ‘intermediate’.

P53 immunohistochemistry. The FFPE sections underwent deparaffinization and rehydration using xylene and alcohol. Subsequently, these sections were treated with a primary antibody against p53 (mouse monoclonal, clone DO7, 1:300, NCL-p53-CM5p, Novocastra, Newcastle, UK). Using an ultraView Universal DAB Detection Kit (Ventana Medical Systems, Tucson, AZ, USA), the slices were chromogenically visualized, followed by the counterstaining with hematoxylin. Evaluation of the stained slides was conducted by JHP and MSK. For p53 immunostaining interpretation, “mutant” pattern was assigned if either ≥60% of tumor cells exhibited strong nuclear expression or complete absence of staining, whereas “wild-type” pattern was given to tumor cells showing p53 expression levels between these two (1-59% staining), as previously documented (33,34).

TP53 sequencing. Discrepancies between TP53 sequencing results and p53 immunostaining patterns have been reported, particularly with the wild-type and complete negative p53 staining patterns (35,36). Six cases (three with complete negative p53 staining and three with wild-type pattern staining) were thus selected for the TP53 Sanger sequencing. A representative section of the FFPE block was used for TP53 sequencing. The QIAamp DNA FFPE Tissue Kit (Qiagen, Valencia, CA, USA) was used to extract DNA from the FFPE tissues, based on the manufacturer’s protocols. The TP53 exons (2-11) were amplified using the PyroMark PCR Kit (Qiagen, Hilden, Germany) using different primers. Amplification reaction was performed for 35 cycles at 95˚C for 30 s, 60˚C for 30 s, and 72˚C for 1 min. Despite multiple attempts at polymerase chain reaction (PCR) amplification, one sample with complete negative staining was not successfully amplified. Samples were then run on 1.5% agarose gels to ensure the quality and length of the amplified regions. The PCR products were purified using the ExoSAP-IT™ Express PCR Product Cleanup (Thermo Fisher, Waltham, MA, USA). Sanger sequencing was performed using the BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher). For each reaction, 20-100 ng/μl of each PCR product was used. DNA sequences were subjected to capillary electrophoresis on ABI PRISM 3730XL Analyzer (96 capillary types, Thermo Fisher). Variant analysis was performed using the Variant Reporter Software version 2.1 (Applied Biosystems, Waltham, MA, USA).

Targeted NGS. To further compare the p53 immunostaining results and TP53 mutational status, 22 additional cancer specimens were collected retrospectively from 3, 9, and 10 patients with gastric, ovarian, and endometrial cancer, respectively. The specimens were used in both p53 immunostaining and targeted NGS during diagnostic work-up at the Department of Pathology, Kyungpook National University Chilgok Hospital, in 2022 (From January to December). In summary, cancer-associated genetic alterations were identified using targeted next-generation sequencing (NGS) with a customized cancer panel (ONCOaccuPanel, NGeneBio, Seoul, Republic of Korea), following the manufacturer’s guidelines. This panel was designed to detect potential single nucleotide variants in 323 cancer-related genes, as well as insertions/deletions, copy number variations, and potential fusion variants of six genes. For the targeted NGS procedure, DNA was initially extracted from the FFPE sections using the QIAGEN AllPrep DNA/RNA FFPE Kit (Qiagen, Hilden, Germany). Subsequently, 100 ng of the extracted DNA underwent hybridization capture-based target enrichment, followed by paired-end sequencing (2×150 bp) on the MiSeq sequencer (Illumina, San Diego, CA, USA), as per the manufacturer’s protocol. Analysis of single nucleotide variants (SNVs), insertions/deletions (indels), and copy number variations (CNVs) in the samples was conducted using the Burrows–Wheeler Aligner and Genome Analysis Tool Kit. Variants with a minimum depth of coverage of 100× and allele frequencies of at least 3% were included in the analysis. Fusion variants were assessed using STAR-Fusion and FusionCatcher. Finally, variant interpretation and annotation were performed in accordance with the guidelines provided by the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists (37).

Epstein-Barr virus encoded RNA (EBER) in situ hybridization. We conducted EBER in situ hybridization using the INFORM EBV-encoded RNA probe (Ventana Medical Systems) to evaluate the EBV status of the samples, following the manufacturer’s protocol. Each hybridization run included a positive control acquired from EBV-positive nasopharyngeal carcinoma.

MSI Testing. We utilized polymerase chain reaction (PCR) with five National Cancer Institute (NCI) markers (BAT-26, BAT-25, D5S346, D17S250, and S2S123) to determine the microsatellite instability (MSI) status of tumors. Both representative tumor tissues and corresponding normal tissues were utilized for MSI assessment. PCR products were analyzed using a DNA autosequencer (ABI 3731 Genetic Analyzer; Thermo Fisher Scientific). Following the revised Bethesda Guidelines (38), tumors exhibiting at least two markers with unstable peaks were categorized as MSI-high, those with one unstable marker were considered MSI-low, and tumors lacking unstable markers were labeled as microsatellite stable (MSS).

Statistical analysis. The associations between clinicopathological parameters were assessed using the chi-square test. p-Values <0.05 was considered statistically significant. Statistical computations were conducted utilizing the R software.

Results

Patient cohort. The median age of the patients was 72 years (range=52-86 years). There were 45 males and 5 females. Among them, 42 biopsy specimens and 8 ESD/EMR specimens were included. On histologic examination, four cases were indefinite for dysplasia, 10 cases were intestinal-type dysplasia, low-grade, nine cases were intestinal-type dysplasia, high-grade, 25 cases were tubular adenocarcinoma, and two cases were poorly cohesive carcinomas. Fourteen cases showed low-grade nuclear atypia, 31 cases showed high-grade nuclear atypia, and five cases showed intermediate grade nuclear atypia. On p53 staining, 17 patients presented with wild-type expression and 33 with mutant-type expression. Among them, 26 patients had diffuse and strong positive nuclear staining, and the remaining seven presented with complete negative staining. Among the 12 cases that underwent EBER in situ hybridization and MSI testing, all cases were EBV-negative, and two cases were MSI-high.

Clinical utility of p53 immunohistochemistry in the diagnosis of gastric cancer and dysplasia. The results of p53 immunostaining in association with clinicopathologic parameters are shown in Table I. Among clinicopathological parameters, the mutant pattern of p53 staining showed significant associations with high-grade nuclear atypia (p<0.001), high-grade dysplasia, and tubular adenocarcinoma (p<0.001), as well as with microsatellite stable (MSS) status (p=0.034).

Table I. p53 immunohistochemistry result in association with clinicopathologic parameters.

EBV: Epstein-Barr virus; EMR: endoscopic mucosal resection; ESD: endoscopic submucosal dissection; MSI: microsatellite instability; MSS: microsatellite stable; N/A: not applicable.

Upon reviewing the slides, p53 staining performed for the diagnostic purpose was categorized under the following three circumstances; 1) to evaluate biopsy specimens only containing a small amount of lesion (n=28), 2) to evaluate resection margin status (n=6), 3) to determine the degree of gastric dysplasia (n=16).

Figure 1 shows representative cases of small biopsy samples wherein p53 staining was used to determine the diagnosis. Mutant pattern p53 staining was particularly helpful for small biopsy specimens that contained only few cancer or dysplastic cells, needing to be distinguished from reactive gastric epithelial cells. Meanwhile, the diagnosis of indefinite for dysplasia, the term which can be applied to cases showing some degree of nuclear or cytologic atypia but lacking definitive histologic features, was supported by a wild-type p53 staining result.

Figure 1. Representative cases in which p53 staining was performed for diagnostic purposes in small biopsies. (A) and (B): Case 1 (tubular adenocarcinoma). (A) Although tumor cells (arrows) showed hyperchromasia and nuclear enlargement, only a few tumor cells were sampled. In addition, the tumor cells were admixed with many inflammatory cells, requiring distinction from reactive gastric epithelial change. (B) On p53 immunostaining, the tumor cells showed strong nuclear immunoreactivity. On subsequent endoscopic submucosal resection, the diagnosis of tubular adenocarcinoma was confirmed. (C) and (D): Case 2 (gastric dysplasia, high-grade). (C) On histologic examination, only a single strip of gastric epithelium (arrow) showed suspicious dysplastic change. (D) Strong p53 nuclear positivity supported the diagnosis of high-grade dysplasia. Subsequent endoscopic submucosal resection confirmed the lesion as high-grade dysplasia. (E) and (F): Case 3 (indefinite for dysplasia). (E) The biopsy specimen demonstrates some atypical glands. Gastric epithelium shows some distorted glands with enlarged, hyperchromatic nuclei. However, the lesion was too small and heavily infiltrated with inflammatory cells. (F) p53 immunostaining showed wild-type expression (F). Follow-up biopsy was performed twice by six-month sampling intervals and only atrophic gastric epithelium was found. Original magnifications: (A)-(B) and (E)-(F), ×100; (C)-(D), ×200.

Figure 2 represents ESD/EMR cases that p53 immunostaining played a crucial role in determining the resection margin status. Histologic evaluation of ESD/EMR resection margin can be hampered by the cauterization artifact during the procedure. In such circumstances, mutant pattern p53 immunostaining can be used for the resection margin assessment.

Figure 2. Representative cases wherein p53 immunostaining was used for the resection margin assessment. (A)-(C): Case 1. (A) Histologic diagnosis was tubular adenocarcinoma. The infiltrating glands demonstrate complex branching. The neoplastic tumor cells show hyperchromasia and size enlargement. (B) Accurate margin evaluation was difficult by histologic examination alone due to the cauterization artifact. (C) Cauterized tumor cells showed diffuse nuclear positivity for p53 staining, which confirmed the margin involvement by the tumor. (D)-(F): Case 2. (D) The histologic diagnosis was also tubular adenocarcinoma, and (E) resection margin status was also difficult to evaluate due to the cauterization artifact. (F) Diffuse p53 nuclear positivity on cauterized tumor cells also confirmed the margin involvement by the tumor cells in this case. Original magnifications: (A) and (D), ×100; (B)-(C) and (E)-(F), ×40.

Figure 3 shows gastric dysplasia cases showing borderline degree of nuclear and cytologic atypia between low-grade and high-grade. In such cases, mutant pattern p53 immunostaining was the supporting evidence of high-grade dysplasia, whereas wild type p53 expression favored the diagnosis of low-grade dysplasia.

Figure 3. Representative cases wherein p53 staining was used to distinguish between low-grade and high-grade dysplasia. (A)-(B): Case 1, (C)-(D): Case 2. Histologic examination of case 1 (A) and case 2 (C) shows unequivocal neoplastic changes of gastric epithelium. However, the distinction of low-grade and high-grade dysplasia is not always clear based on histologic examination alone, considering nuclear atypia and architectural distortion. Strong nuclear p53 expression (B) and complete absence of p53 staining (D) supported the diagnosis of high-grade dysplasia. (E)-(F): Case 3. The lesion consists of dysplastic intestinalized epithelium (E). However, the degree of cytological atypia was intermediate between low-grade and high-grade. Architectural atypia showing moderate level of gland distortion and back-to-back glandular rearrangements were also observed. Wild-type p53 staining in this case supported the diagnosis of low-grade dysplasia. Original magnifications: (A) and (E)-(F), ×100; (B), ×40; (C)-(D), ×200.

Comparison of the p53 immunostaining pattern and TP53 sequencing results. The p53 immunostaining patterns were compared with the sequencing results. Three patients with wild-type p53 staining in this cohort did not harbor an oncogenic TP53 mutation on Sanger sequencing. However, pathogenic TP53 mutations were not detected in two patients with the complete negative p53 immunostaining pattern. Due to the limited number of patients who underwent Sanger sequencing, the p53 immuno-histochemical staining results were further validated using targeted NGS. Results showed that seven patients with the wild-type p53 immunostaining pattern did not harbor TP53 gene mutation. Eight patients with a diffuse and strong pattern on p53 immunostaining harbored oncogenic missense TP53 mutations. Six patients with the complete negative staining pattern harbored a pathogenic TP53 gene mutation (five with truncating mutation and one with oncogenic missense TP53 mutation). One patient with the cytoplasmic p53 immunostaining pattern had a splice site mutation, which is consistent with a previous study (35) (Table II).

Table II. Comparison of the p53 immunostaining patterns and TP53 sequencing results.

IHC: Immunohistochemistry; NA: not applicable; NGS: next-generation sequencing.

Discussion

We present the diagnostic utility of p53 staining in real-world cases in the diagnosis of gastric adenocarcinoma, its precursor lesion, and tumor-like lesions. The importance of TP53 gene in the tumorigenesis have been comprehensively studied (7). Pathogenic mutations of TP53 are thus used for the prognostic stratification of hematologic malignancies (39,40). Also, loss of function of TP53 is obligatory in several malignancies, such as high-grade serous carcinoma of the ovary (41) and small cell lung carcinoma (42).

Oncogenic mutation of TP53 frequently accompanies morphologic changes of tumor cells. Freed-Pastor et al. (2012) showed that mutant p53 depletion leads to phenotypic conversion to invasive morphology (43). Several other studies have linked TP53 mutation with epithelial-mesenchymal transition and tumor invasiveness (44,45). The association of cytologic and architectural atypia with TP53 mutation in gastrointestinal tract cancer have also been suggested (22,29,46). Gastric dysplasia is the precursor of gastric adenocarcinoma. Gastric dysplasia is graded as low-grade or high-grade depending on the degree of nuclear atypia, cytologic atypia, mitotic activity, and cytoplasmic differentiation (32). TP53 mutations frequently occur in high-grade dysplasia, whereas they are not common in low-grade dysplasia (46,47).

In line with previous studies, this study also found a strong association between mutant pattern p53 staining and tubular adenocarcinoma, high grade dysplasia, and high-grade nuclear atypia. Therefore, aberrant expression of p53 immunostaining in these lesions can be used as an important diagnostic tool. We recommend three diagnostic indications of p53 staining in stomach lesions (Figure 4).

Figure 4. Diagnostic indications for p53 staining in gastric tumors and tumor-like lesions.

Biopsy sampling using upper gastrointestinal endoscopy is widely accepted as the gold standard method for the detection of gastric cancer and its precursor lesion (48). However, histologic examination for cancer screening is sometimes disturbed by inadequate samples (49). Of note, when only few tumor cells can be sampled, the presence of strong nuclear expression or complete absence of immunostaining suggests the lesion to be neoplastic (at least high-grade dysplasia), even if only a few atypical cells are sampled. However, wild-type p53 does not rule out the possibility of gastric adenocarcinoma or dysplasia (46,50). Integrative diagnosis should be made by combining histologic examination, p53 staining result, and endoscopic findings.

ESD/EMR is a minimally invasive endoscopic procedure for the en block resection of a gastric tumor. ESD/EMR is currently the standard procedure for early gastric cancer and its precursor lesion (51). Accurate post-operative pathologic margin evaluation is very important. Several previous studies demonstrated the association of resection margin status with patient outcome (52-55). Although pathologic margin evaluation mainly relies on histologic examination, accurate histologic examination is sometimes hampered by the cauterization artifact (56). In such circumstances, aberrant p53 immunostaining in cauterized tumor cells can be successfully applied to confirm the resection margin involvement. Since TP53 mutation frequently occurs in the intestinal subtype in TCGA molecular classification and much less in other subtypes (4), this approach is particularly useful if the lesion shows intestinal-type differentiation.

Gastric dysplasia, the precursor lesion of gastric adenocarcinoma, is graded into low-grade and high-grade (32,57). In low-grade dysplasia, tumor cells usually show only mild to moderate nuclear atypia and mild architectural distortion. The nuclei are elongated and maintain polarity. In contrast, high-grade dysplasia shows prominent nuclear atypia, nuclear size enlargement, and commonly prominent nucleoli. Complex glandular atypia showing marked distortion is often observed. Although current WHO classification uses a 2-tiered system for gastric dysplasia, in real-world cases, there are dilemmatic cases showing intermediate differentiation between low-grade and high-grade dysplasia, which can cause inter-observer variation (35,58). Since TP53 mutations frequently occur in high-grade dysplasia, p53 immunostaining can be helpful for the distinction of low-grade and high-grade dysplasia in such cases.

We would like to again emphasize that diagnostic application of p53 staining is most successful when applied to tumors showing intestinal morphology. Since alterations of TP53 frequently accompany morphologic changes of tumor cells, we also recommend p53 staining only when tumor cells show at least an intermediate degree of nuclear atypia. In addition, wild-type p53 immunostaining can exhibit a wide range of nuclear reactivity, from near absence of staining to high-wild type expression (35). Accurate interpretation of p53 immunostaining is thus very important for determining the diagnosis.

Study limitations. First, some pathogenic alterations of TP53, especially truncating mutations or splice site mutations, can result in wild-type p53 staining pattern (35,59), although the number of such cases is small. Thus, the prediction of the TP53 mutational status based on immunohistochemistry alone can be an issue. Indeed, in this cohort, two cases of complete negative p53 staining did not harbor pathogenic TP53 alterations on Sanger sequencing. However, in these two cases, the biopsy samples of the patients contained a relatively small amount of tumor cells. Thus, tumor cellularity might have led to the false-negative results on Sanger sequencing. Otherwise, pathogenic TP53 mutations could have occurred in other sites such as the splice acceptor or donor region, which cannot be covered by selected exon sequencing. Indeed, when we validated the p53 immunostaining pattern using targeted NGS, which covers not only the whole exons but also the partial introns of the TP53 gene, immunostaining of p53 predicted a TP53 mutational status in all but one case (complete negative immunostaining and oncogenic missense TP53 sequencing result). Hence, p53 immunostaining can be an effective and convenient tool for predicting TP53 mutational status. However, a completely negative p53 immunostaining pattern should be interpreted with caution, in combination with histologic examination and endoscopic findings. Second, due to the limited number of cases, our study only included six cases wherein p53 immunostaining was used to evaluate margin involvement. Subsequent studies using a larger cohort should be conducted to validate whether p53 immunostaining for margin evaluation is more effective than histologic examination. Several gastric adenocarcinomas and high-grade dysplasia do not harbor TP53 mutations. However, a wild-type pattern of p53 immunostaining does not guarantee that the lesion will be reactive. Gastric lesion diagnosis should be made based on the combination of histological, p53 immunohistochemistry, and endoscopic findings. Previous studies have discussed the diagnostic value of p53 immunostaining in gastric lesions. However, this article comprehensively evaluated the diagnostic importance of p53 staining for determining gastric cancer and its precursor lesions alone.

In conclusion, p53 staining is a useful diagnostic tool when applied to appropriate indications for the evaluation of small biopsy specimens, the pathologic evaluation of ESD/EMR margin, and the distinction between low- and high-grade dysplasia.

Funding

This work was supported by a Biomedical Research Institute grant, from Kyungpook National University Hospital (2023).

Conflicts of Interest

The Authors declare that they have no competing interests in relation to this study.

Authors’ Contributions

MSK conceived and designed the manuscript. JHP and MSK drafted the manuscript. MSK, ANS analyzed previous articles on aggressive thyroid cancer. JHP, ANS, and MSK reviewed and revised the manuscript carefully. All Authors have read and approved the final manuscript.