Sage Journals: Discover world-class research

Abstract

Introduction

This retrospective study aimed to investigate the correlation between TP53 identified via next-generation sequencing (NGS) and p53 expression in colorectal adenocarcinoma (CRC), as assessed by immunohistochemistry (IHC). Additionally, we characterized p53 IHC staining patterns and sought to determine the optimal threshold for p53 expression as a surrogate for TP53 mutation status.

Methods

In this retrospective cohort analysis, we included 294 archived surgically resected CRC specimens from patients who did not receive preoperative chemotherapy were analyzed. All data were collected from pathology database and electronic medical records. TP53 mutations were identified using NGS, and p53 expression was evaluated by IHC. The correlation between mutation status and IHC staining patterns was assessed, and sensitivity and specificity were calculated.

Results

The TP53 mutation rate was 78.2%, comprising missense (68.4%), nonsense (12.4%), frameshift (11.0%), and splice-site (8.3%) mutations. Missense mutations were associated with nuclear p53 staining, while frameshift mutations mostly showed loss of expression. Nonsense and splice-site mutations exhibited diverse patterns, including loss of expression, nuclear staining with/without cytoplasmic staining, or cytoplasmic staining alone. Among cases with loss of p53 expression, the TP53 mutation rate was 88.9%. When the proportion of strong p53-positive cells exceeded 55%, the missense mutation-positivity rate increased significantly (P < 0.05). The sensitivity and specificity of p53 IHC in predicting TP53 mutations were 92.3% and 94.8%, respectively.

Conclusions

CRC predominantly exhibited missense TP53 mutations. p53 IHC revealed diverse expression patterns, including overexpression, complete loss, cytoplasmic staining, and normal-type patterns. Strong p53 expression (>55%) correlated closely with TP53 missense mutations, supporting IHC as a reliable surrogate. However, cases showing loss of p53 expression should undergo gene sequencing to confirm mutation status.

Keywords

colorectal adenocarcinoma immunohistochemistry next-generation sequencing threshold

Introduction

TP53 is a critical tumor suppressor gene and one of the most frequently mutated genes in human malignancies. Located on the short arm of chromosome 17 (17p13.1), it consists of 11 exons and 10 introns and encodes a 53-kDa nuclear phosphoprotein composed of 393 amino acids. As a transcription factor, p53 contains three key functional domains: the N-terminal transactivation domain (amino acids 1-93), the central DNA-binding domain (amino acids 98-298), and the C-terminal oligomerization domain. Notably, the DNA-binding domain is a mutation hotspot.^1,2 Mutant TP53 lacks the tumor-suppressive function of the wild-type protein and promotes malignant progression through a “gain-of-function” mechanism. In colorectal adenocarcinoma, TP53 mutations are closely linked to tumor proliferation, invasion, metastasis, and remodeling of the tumor microenvironment (eg, lymphocyte infiltration). The mutation types and sites significantly influence both chemosensitivity and prognosis.^3,4

Currently, TP53 sequencing poses challenges due to high costs and technical complexity. As a result, immunohistochemistry (IHC) is commonly used in clinical practice to detect mutant p53. While IHC offers advantages such as simplicity and rapid turnaround time, the absence of standardized interpretation criteria for staining patterns has led to controversy regarding the correlation between p53 IHC results and TP53 mutation status.^5–7 This inconsistency remains particularly understudied in colorectal cancer, with limited systematic evidence.

To address this gap, we analyzed p53 IHC staining patterns and next-generation sequencing (NGS) results from 294 surgically resected colorectal adenocarcinoma specimens. Our aim was to characterize abnormal p53 expression patterns, establish a positive IHC threshold based on the percentage of nuclei with strong positivity, and evaluate whether IHC-derived p53 expression can serve as a reliable surrogate for TP53 mutation status. Our findings may help define clinical indications for molecular testing and support treatment-related decision-making, particularly in routine clinical settings.

Materials and Methods

Clinical Data

This was a retrospective diagnostic accuracy study conducted at The First Affiliated Hospital of Fujian Medical University. The reporting of this study conforms to STROBE guidelines.⁸ This single-center, retrospective study was conducted after obtaining approval from the hospital's ethics committee (approval no. [2015084-2]). Written informed consent was obtained from all participants, which included permission for the use of their clinical data and tissue samples in the current and future biomedical research. All patient data were de-identified to ensure anonymity.

We searched our institutional pathology database to identify all patients with colorectal adenocarcinoma who underwent surgical resection between May 2021 and December 2023. Patient selection was based on the availability of complete datasets. The inclusion criteria were: (1) postoperative histopathological confirmation of colorectal adenocarcinoma according to the fifth edition of the World Health Organization Classification of Digestive System Tumors; (2) no preoperative neoadjuvant therapy; (3) availability of complete results for immunohistochemical testing of mismatch repair proteins (MLH1, MSH2, MSH6, and PMS2) and PCR-based verification of microsatellite loci (BAT-25, BAT-26, NR-21, NR-24, and MONO-27); and (4) concurrent completion of both NGS for TP53 and p53 IHC.

The final cohort comprised 294 patients. The cohort had a median age of 65 years (range: 28-88) and a male-to-female ratio of 176:118. NGS identified TP53 mutations in 230 patients (78.2%), while the remaining 64 (21.8%) had wild-type TP53.

Analysis of p53 Expression Using IHC

All specimens underwent routine tissue processing, dehydration, paraffin embedding, and sectioning at a thickness of 4 µm. For each case, one representative hematoxylin and eosin (H&E)-stained slide showing diagnostic features was selected, and the corresponding paraffin blocks were used for p53 IHC staining. Staining was performed using the EnVision two-step method with an anti-p53 antibody (DO-7, 1:500 dilution; Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd) on a Roche automated staining platform, following the manufacturer's protocol. Five random fields at 20× magnification (300 × 300 µm) were evaluated per case. Each field was assessed for p53 staining intensity (categorized as loss of expression, weak, moderate, strong, or heterogeneous), subcellular localization (nuclear with or without cytoplasmic staining, or cytoplasmic staining alone), and the percentage of positive cells. Final results were averaged across the fields. The interpretation criteria for p53 immunohistochemical staining are classified into the following five distinct patterns.² (1) Nuclear expression: characterized by strong nuclear positivity without cytoplasmic expression in tumor cells. (2) Nuclear expression with cytoplasmic expression : characterized by nuclear positivity with strong cytoplasmic expression in tumor cells . (3) Loss of expression: defined as a complete lack of nuclear staining in all tumor cells, with concurrent positive staining in an internal control (eg, normal mucosa, stromal cells). (4) Cytoplasmic: marked by significant cytoplasmic staining accompanied by an absence of nuclear staining. (5) Wild-type: exhibits a variable (heterogeneous) nuclear staining of weak to moderate intensity. The background microenvironment (stromal cells, lymphocytes) shows normal p53 expression and can serve as an internal control. Tonsil tissue was used as an external control, which required that over 40% of the cells in the germinal centers exhibited varying degrees of staining, with occasional strongly positive staining cells (Supplementary Figure 1A). For cases with significant heterogeneity, hotspot areas were evaluated. Two pathologists independently reviewed all slides. The inter-rater agreement for IHC scores was R = 0.996, P < 0.0001. The positive/negative assessments were consistent in 97.3% of cases. Discrepant cases (where only one rater scored the sample as positive) accounted for only 2.7% of the total. For details, please refer to (Supplementary Figure 1B and Table S1).

Immunohistochemical Analysis of Mismatch Repair (MMR) Proteins

Formalin-fixed, paraffin-embedded (FFPE) tissue sections (4-μm thick) from primary colorectal cancer specimens were utilized for MMR protein analysis. IHC was performed on an automated staining platform (eg, Ventana BenchMark XT) according to the manufacturer's instructions. Briefly, after deparaffinization and heat-induced epitope retrieval with EDTA buffer (pH 8.0), sections were incubated with primary antibodies against MLH1 (clone M1, Ventana), PMS2 (clone EPR3947, Ventana), MSH2 (clone G219-1129, Ventana), and MSH6 (clone 44, Ventana). The staining was visualized using the UltraView Universal DAB Detection Kit (Ventana), followed by counterstaining with hematoxylin. The assessment of MMR protein expression was conducted by two pathologists in a blinded manner. Nuclear localization of the staining was considered positive. The presence of any discernible nuclear staining in tumor cells, irrespective of intensity, was interpreted as retained expression. The complete absence of nuclear staining in tumor cells, while adjacent non-neoplastic cells (eg, stromal fibroblasts or lymphocytes) served as positive internal controls, was defined as loss of expression. Cases showing loss of expression in one or more of the four MMR proteins were classified as mismatch repair deficient (dMMR). Cases with preserved nuclear expression of all four proteins were classified as mismatch repair proficient (pMMR).

Detection of Microsatellite Instability

Genomic DNA was isolated from FFPE tumor tissues and paired normal mucosa using the DNeasy Blood & Tissue Kit (Qiagen). MSI status was determined using the MSI Analysis System (Version 1.2, Promega), which co-amplifies five mononucleotide repeat markers (BAT-25, BAT-26, NR-21, NR-24, and MONO-27) and two pentanucleotide control markers. PCR was performed according to the manufacturer's protocol. The fluorescently labeled PCR products were analyzed by capillary electrophoresis on an ABI 3500xl Genetic Analyzer (Applied Biosystems). Data were collected and fragment sizes were determined using GeneMapper Software v5.0.

The chromatograms of tumor DNA were compared with those from the matched normal DNA for each marker. The presence of novel peaks (indicating shifts in allele sizes) in the tumor sample that were not present in the normal sample was defined as instability for that marker. A case was classified as microsatellite instability-high (MSI-H) if instability was observed in ≥2 of the 5 mononucleotide markers. Cases with 0 or 1 unstable markers were classified as microsatellite stable (MSS).

NGS

All samples underwent manual macrodissection of FFPE sections to ensure a minimum tumor cell fraction of 20% prior to nucleic acid extraction. Targeted NGS was performed using the Colorectal Cancer NGS Testing Kit Panel (Amoy Diagnostics Co., Ltd, Xiamen,China). This panel comprehensively covers the coding regions of 200 genes, including TP53, KRAS, APC, BRAF, NRAS, PIK3CA genes relevant to CRC), and other genes frequently altered in solid tumors. Genomic DNA was extracted from FFPE tissues using the QIAamp DNA FFPE Tissue Kit (Cat. no. 56404; Qiagen, Hilden, Germany). DNA concentration and purity were assessed using a Nanodrop Lite spectrophotometer and a Qubit 3.0 fluorometer with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific), ensuring sample quality for library preparation (A260/A280 ratio: 1.8-2.0; total DNA ≥50 ng). DNA was fragmented via ultrasonic shearing (Covaris M220, USA), followed by end repair and adapter ligation to create indexed pre-libraries. Hybridization capture was performed using a human 200-gene mutation detection panel (reversible terminator-based sequencing) to generate the final libraries. Fragment distribution of both pre- and final libraries was assessed using an Agilent 2100 Bioanalyzer to confirm quality (main peak: 150-200 bp, with no significant adapter dimer contamination). Paired-end sequencing (2 × 150 bp) was conducted on an Illumina NextSeq 550Dx high-throughput sequencer. Mutation data were analyzed using a bioinformatics pipeline, and all sequencing results were independently reviewed by two molecular pathologists to ensure accuracy.

Statistical Analysis

Statistical analyses were performed using SPSS 18.0 software. Receiver operating characteristic (ROC) curves were constructed to assess the diagnostic performance of p53 IHC expression in detecting TP53 mutations and to calculate sensitivity and specificity at different thresholds. The optimal cutoff value was determined using Youden's Index, defined as: sensitivity + specificity − 1. Chi-square tests were used to evaluate the relationships between IHC results, mutation types, and expression patterns. Two-tailed P-values < 0.05 were considered statistically significant.

Results

TP53 Mutation Status

The TP53 mutation rate in sporadic colorectal adenocarcinoma was 78.2% (230/294), with clinically significant TP53 mutations and variants of uncertain significance accounting for 74.2% (218/294) and 4.1% (12/294), respectively. Among the clinically significant mutations, missense mutations were the most frequent (68.4%), followed by nonsense (12.4%), frameshift (11.0%), and splice-site mutations (8.3%). Of the 12 cases with variants of uncertain significance, the mutation types included frameshift (seven cases), missense (four cases), and splice-site (one case). Of the TP53 mutations, 84.3% (194/230) were located in the DNA-binding domain, followed by the C-terminal oligomerization domain (11/230) and the N-terminal transactivation domain (6/230). Additionally, 19 mutations occurred in intronic regions (Supplementary Figure 1C and Table S2). TP53 gene status in sporadic colorectal adenocarcinoma was strongly associated with abnormal p53 IHC staining patterns (P < 0.000), but showed no significant correlation with patient age, sex, tumor differentiation, or lymph node metastasis. Additionally, among all cases, MSI-H accounted for 3.7% (11/294). TP53 mutations were more common in pMMR tumors (79.9%), while their mutation frequency was significantly lower in dMMR tumors (36.4%) (Table 1 and Table S3).

Table 1.

Association of TP53 Gene Phenotype with Clinical Pathological Factors and p53 IHC Expression Patterns in Colorectal Carcinoma.

Clinical Pathological Factors	TP53 Gene Phenotype (N = 294)			TP53 mutation types
	Wild (N = 64)	Mutation (N = 230)	P-value	Missense (N = 153)	Nonsense (N = 27)	Frameshift (N = 31)	Splice-site (N = 19)
Age (years)
<59	22 (23.2%)	73 (76.8%)	0.690	46 (63%)	6(8.2%)	13 (17.8%)	8 (11%)
≥59	42 (21.1%)	157 (78.9%)		107 (68.2%)	21 (13.4%)	18 (11.5%)	11(7%)
Sex
Male	36 (20.5%)	140 (79.5%)	0.505	98 (70%)	16 (11.4%)	17 (12.1%)	9 (6.4%)
Female	28 (23.7%)	90 (76.3%)		55 (61.1%)	11 (12.2%)	14 (15.6%)	10 (11.1%)
Differentiation
Well + Moderate	53 (20.5%)	205 (79.5%)	0.173	139 (67.8%)	23 (11.2%)	25 (12.2%)	18 (8.8%)
Poor	11 (30.6%)	25 (69.4%)		14 (56%)	4 (16%)	6 (24%)	1 (4%)
Lymph node metastasis
Positive	29 (25.2%)	86 (74.8%)	0.251	55 (64%)	11 (12.8%)	13 (15.1%)	7 (8.1%)
Negative	35 (19.6%)	144 (80.4%)		98 (68.1%)	16 (11.1%)	18 (12.5%)	12 (8.3%)
MMR status			0.003
dMMR/MSI-H	7 (63.6%)	4 (36.4%)		3 (75%)	/	/	1(25%)
pMMR/MSS	57(20.1%)	226 (79.9%)		150 (66.4%)	27 (11.9%)	31 (13.7%)	18(8%)
IHC staining patterns			0.000
Nuclear expression	3 (1.7%)	173 (98.3%)		153 (88.4%)	4 (2.3%)	9 (5.2%)	7 (4%)
Nuclear expression with Cytoplasmic	/	4 (100%)		/	2 (50%)	/	2 (50%)
Cytoplasmic	/	2 (100%)		/	1 (50%)	/	1 (50%)
Loss of expression	6 (11.1%)	48 (88.9%)		/	20 (41.7%)	21 (43.8%)	7 (14.6%)
Wild-type	55 (94.8%)	3 (5.2%)		/	/	1 (33.3%)	2 (66.7%)

IHC, immunohistochemistry, dMMR, mismatch repair deficient, pMMR, mismatch repair proficient, MSS-H, microsatellite instability-high, MSS, microsatellite stable

Assessment of p53 Expression Using IHC

Among the 294 cases, 240 (84.6%) exhibited varying degrees and patterns of p53 immunostaining, while 54 cases (18.4%) showed complete loss of p53 expression (Figure 1). The p53 staining pattern was assessed in relation to the type of TP53 mutation. All missense mutations (149 cases) demonstrated nuclear staining. Of the frameshift mutations (31 cases), 67.7% (21/31) showed loss of expression and 29.0% (9/31) exhibited nuclear staining. Among nonsense mutations (27 cases), 74.1% (20/27) showed loss of expression, 22.2% (6/27) showed nuclear staining with or without cytoplasmic staining, and 3.7% (1/27) showed cytoplasmic staining alone. Splice-site mutations (19 cases) displayed all three staining patterns. Missense and splice-site mutations were significantly associated with strong nuclear staining, whereas nonsense and frameshift mutations were predominantly associated with loss of expression. Abnormal cytoplasmic staining was observed in only two cases—one with a nonsense mutation and one with a splice-site mutation (Table 1).

Figure 1.

Correlation between TP53 mutation status and p53 protein expression patterns: A. Colorectal adenocarcinoma (H&E staining). B. Nuclear overexpression without cytoplasmic staining (TP53 missense mutation, exon5. c. 473G > Ap. R158H). C. Nuclear staining with cytoplasmic staining (TP53 splice-site mutation, intron6c.673-2A>G). D. Abnormal cytoplasmic expression (TP53 splice-site mutation, intron8c.920-2del). E. Complete loss of expression (TP53 nonsense mutation, exon5c.378C > A p.Y126*). F. Weak to moderate heterogeneous wild-type expression (TP53 wild-type). H&E = hematoxylin and eosin.

Additionally, cases showing loss of p53 expression had a significantly higher TP53 mutation rate, at 88.9% (48/54). In cases with TP53 variants of uncertain significance, regardless of mutation type (frameshift, missense, or splice-site), p53 IHC staining consistently showed nuclear localization.

Optimization and Validation of a 55% Cutoff for p53 IHC as a Robust Surrogate for Detecting TP53 Mutations in CRC

ROC curve analysis was conducted to compare p53 IHC results with NGS findings. A total of 240 cases with positive p53 expression were analyzed, comprising 52 wild-type cases, 125 cases with missense mutations, 15 cases with splice-site mutations, 22 cases with nonsense mutations, and 26 cases with frameshift mutations. Different mutation types were associated with varying levels of p53 protein expression. The median p53 protein expression was 90% for missense mutations, 0% for nonsense mutations, 0% for frameshift mutations, and 10% for splice-site mutations (Figure 2A). Using a cutoff value of 55% for p53-positive expression (Figure 2B), the strongest correlation between IHC and NGS for TP53 missense mutations cases was observed, with a sensitivity of 92.3% and specificity of 94.8% (P < 0.05). The positive predictive value was 98.2%, and the negative predictive value was 78.6% metrics at the 55% threshold (Table S4). Based on this, IHC positivity was defined as strong nuclear staining in >55% of tumor cells, and negativity as strong nuclear staining in ≤55%. The application of these criteria to the IHC-positive group (n = 171) showed that 168 cases (98.3%) harbored TP53 mutations as confirmed by NGS, while the remaining three were wild-type. In contrast, within the IHC-negative group (n = 69), TP53 mutations were detected in 14 cases (20.9%), and 55 cases were wild-type (Figure 2C). No significant correlation was observed between the intensity of p53 protein expression and the TP53 mutation type (missense vs nonsense) (P > 0.05).

Figure 2.

Performance of p53 immunohistochemistry with a defined positivity threshold for screening TP53 mutation status: A. Distribution of the p53 percentage positivity for each mutation type. B. The receiver operating characteristic (ROC) curve indicates that a 55% positive cell count is the optimal cutoff value, showing the best concordance between the immunohistochemical (IHC) and next-generation sequencing (NGS) results with regard to determining the TP53 gene missense mutation status. C. Grouped bar chart of cases.

Among the 54 cases that showed a complete absence of protein expression, the distribution of mutation types was as follows: six were wild-type, 20 had nonsense mutations, 21 had frameshift mutations, and seven had splice-site mutations. The complete absence of protein expression by IHC demonstrated a sensitivity of 62.3% (48/77) and a specificity of 90.6% (58/64) for predicting the corresponding genetic alterations. The positive and negative predictive values were 88.9% and 66.7%, respectively. For details, please refer to Appendix Table S5.

Discussion

As one of the most critical tumor suppressor genes, TP53 is the most frequently mutated gene in human malignancies, with mutation rates exceeding 50% across all tumor types. Additionally, studies have shown that the TP53 mutation status in CRC is negatively correlated with MMR status.^9,10 Existing data have shown distinct patterns in the distribution of TP53 mutation types: missense mutations are the most common (60-75%), followed by frameshift mutations (∼25%) and nonsense mutations (∼15%).^1,11 The TP53 mutation rate in sporadic CRC is approximately 40%–50%^3,4; however, systematic studies analyzing the frequency distribution of specific TP53 mutation types and their associated protein expression patterns in CRC remain limited.

Current evidence suggests that different TP53 mutation types are correlated with characteristic p53 IHC expression patterns. Common patterns include missense mutation-associated overexpression, complete loss of expression, abnormal cytoplasmic localization, and wild-type expression. The missense mutation pattern (also termed the “overexpression pattern”) is typically characterized by diffuse, strong nuclear staining due to the accumulation of mutant p53 protein. The loss-of-expression pattern, often observed in frameshift, nonsense, or splice-site mutations, typically shows a complete absence of staining, suggesting the presence of truncated or unstable proteins. Abnormal cytoplasmic expression—marked by cytoplasmic staining—is likely associated with p53 mislocalization. The wild-type pattern exhibits weak-to-moderate heterogeneous nuclear staining, indicative of normal p53 protein levels. Notably, a unique “basal overexpression pattern” has been reported in certain tumor types (eg, vulvar squamous cell carcinoma), where intense staining is localized to tumor cells in the basal layer, possibly reflecting terminal differentiation.¹² These findings emphasize the tumor type-specific nature of p53 IHC patterns, and in clinicopathological practice, interpretation should consider tumor context to avoid misclassification.

The results of our study revealed a TP53 mutation rate of 78.2% in sporadic colorectal adenocarcinoma. TP53 mutations were more common in pMMR tumors (79.9%), while their frequency was significantly reduced in dMMR tumors (36.4%), further confirming the negative correlation between TP53 mutation status and MMR status in CRC. The distribution of mutation types was distinct, with missense mutations (68.4%), nonsense mutations (12.4%), frameshift mutations (11.0%), and splice-site mutations (8.3%). Notably, different mutation types were associated with distinct p53 protein expression patterns, reflecting their varying impacts on protein expression and stability. Specifically, nonsense or frameshift mutations that lead to the production of truncated proteins generally do not result in stable proteins, resulting in a negative IHC result. In contrast, many missense mutations produce conformationally stable mutant proteins with an extended half-life, which accumulate in the nucleus and lead to strong positive IHC staining. Therefore, directly comparing IHC and sequencing results without considering the underlying mutation type may lead to misinterpretation.^9,13,14 Additionally, we identified two cases of abnormal cytoplasmic expression, both confirmed by molecular testing to carry TP53 mutations: (1) a mutation in the C-terminal nuclear export signal domain (c.1000G > T), and (2) a mutation in the nuclear localization signal domain (c.818G > A). These observations are consistent with previous reports in high-grade serous tubo-ovarian carcinoma and endometrial cancer.^6,15,16 When interpreting cytoplasmic staining results, strict use of positive controls is essential to distinguish true expression from diffusion artifacts caused by overstaining. Our findings further confirm the correlation between TP53 mutations and p53 expression patterns, offering new insights into the molecular mechanisms behind atypical expression in CRC. The rare cytoplasmic expression pattern, linked to domain-specific mutations, provides valuable evidence regarding the molecular basis of p53 mislocalization.

At present, TP53 testing—including NGS and p53 IHC—plays an important role in the diagnosis and management of CRC, as the results are closely tied to therapeutic decision-making and prognosis. However, clinical observations have revealed that p53 protein expression does not always align with TP53 mutation status, indicating a need for further investigation. In ovarian and endometrial cancers, p53 IHC has shown strong predictive value for TP53 mutations. For instance, studies on ovarian cancer report that three abnormal expression patterns (overexpression, loss, and cytoplasmic staining) yield 96% sensitivity and 100% specificity for predicting TP53 mutations.² However, the optimal threshold for defining nuclear overexpression varies by tumor type—eg, >80% for high-grade serous ovarian carcinoma¹⁷ and >10% for isocitrate dehydrogenase (IDH)-mutant gliomas.¹⁸ To date, no standardized cutoff exists for colorectal adenocarcinoma. Our study demonstrates that the performance of p53 immunohistochemistry using a 55% positivity cutoff is critically dependent on mutation type. As the assay detects aberrant protein accumulation, it is inherently unable to identify mutations that lead to protein loss—a limitation reflected in the fact that approximately 80% of such mutations in our cohort were not classifiable. By contrast, for gain-of-function missense mutations that confer protein stability (missense mutations), our work is the first to establish in colorectal adenocarcinoma that the 55% threshold provides optimal concordance with NGS, yielding a sensitivity of 92.3% and a specificity of 94.8%.

Despite a strong overall correlation between TP53 mutations and aberrant p53 expression, 17 of the 240 cases (7.1%) exhibited discordant results. These included 14 cases with detected TP53 mutations but negative p53 IHC, and three cases with wild-type TP53 status but positive p53 IHC. The observed discordances can be attributed to a range of technical, biological, and methodological factors. For cases with wild-type TP53 but positive p53 protein expression, possible reasons include: false positivity due to antibody cross-reactivity or non-specific staining; incomplete coverage of all TP53 exons, regulatory regions, or failure to detect epigenetic alterations by NGS; intratumoral heterogeneity leading to mutations confined only to certain regions; or TP53-independent regulatory mechanisms, such as MDM2/MDM4 overexpression or other genetic alterations that indirectly stabilize p53 protein.^12,19

In cases with TP53 mutations but negative p53 protein expression, multiple factors contribute to the potential mechanisms. First, the status of the TP53 alleles is critically important—if a missense mutation occurs without loss of heterozygosity (LOH) in the second allele, cells may still express wild-type p53 protein; nonsense or frameshift mutations generally do not produce stable proteins, either of which can lead to false-negative IHC results. Second, antibodies may have limited ability to recognize C-terminal mutants or conformationally altered p53 proteins.²⁰ Additionally, while NGS can detect low-frequency mutations, IHC requires protein expression to reach a certain threshold for detection; sampling bias due to tumor heterogeneity may also contribute. Other mechanisms include: co-occurring genetic events such as MDM4 amplification or PPM1D overexpression, which promote the degradation of mutant p53 protein; specific mutation types (eg, splice-site mutations generating Δ40p53 isoforms) that result in unstable proteins unrecognizable by antibodies^21–23; or antigen epitope masking caused by overfixation during sample pre-processing. These mechanisms collectively explain discordant gene–protein results in TP53 testing and offer guidance for refining clinical testing protocols.

Our findings indicate that using a 55% cutoff, p53 IHC serves as an excellent tool for detecting TP53 mutations that lead to protein accumulation (primarily missense mutations), with a positive result (diffuse strong positivity) being highly reliable. However, this method cannot detect null mutations such as nonsense or frameshift mutations. We found that the complete absence of protein, as detected by IHC, had a sensitivity of only 62.3% and a specificity of 90.6% for detecting protein-truncating mutations. A negative IHC result does not rule out TP53 pathway inactivation, as these mutation types typically result in an IHC-negative pattern. For cases with a high clinical suspicion, direct genetic sequencing is recommended, particularly for IHC-negative tumors with highly malignant morphological features. Moreover, this study focused on patients who had not received preoperative neoadjuvant chemotherapy. Therefore, for patients who underwent neoadjuvant treatment, it remains unclear whether TP53 mutation status was altered by therapy and whether such changes are associated with drug resistance. Additionally, as our research concentrated on postoperative resection specimens, we have conducted preliminary work to assess whether this interpretation criterion is also applicable to small preoperative biopsy specimens (with limited fields of view). However, a satisfactory conclusion has not yet been reached. We acknowledge that the present study is based on a quantitative analysis and a single-center dataset. Future work should be extended to multi-center, large-scale cohorts to further validate the standardization of quantitative interpretation across different laboratories.

Conclusion

In summary, this study systematically investigated the relationship between TP53 mutations and p53 protein expression in colorectal adenocarcinoma. We found that nuclear p53 positivity in >55% of tumor cells strongly predicted TP53 missense mutation status, with a sensitivity of 92.3% and a specificity of 94.8%. In cases showing complete p53 expression loss, the TP53 mutation rate was 88.9%, supporting the use of supplementary NGS in such scenarios. While IHC cannot fully replace molecular testing, our results demonstrate that p53 IHC serves as a reliable and cost-effective surrogate for TP53 missense mutations in resource-limited settings. Collectively, these findings offer valuable insights for improving diagnostic accuracy and optimizing treatment strategies and resource allocation in colorectal cancer care.

Supplemental Material

sj-doc-1-tct-10.1177_15330338261420099 - Supplemental material for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer

Supplemental material, sj-doc-1-tct-10.1177_15330338261420099 for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer by Jingjing Wu, Haifeng Yu, Shanshan Huang and Xiangna Chen in Technology in Cancer Research & Treatment

Supplemental Material

sj-tif-2-tct-10.1177_15330338261420099 - Supplemental material for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer

Supplemental material, sj-tif-2-tct-10.1177_15330338261420099 for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer by Jingjing Wu, Haifeng Yu, Shanshan Huang and Xiangna Chen in Technology in Cancer Research & Treatment

Supplemental Material

sj-doc-3-tct-10.1177_15330338261420099 - Supplemental material for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer

Supplemental material, sj-doc-3-tct-10.1177_15330338261420099 for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer by Jingjing Wu, Haifeng Yu, Shanshan Huang and Xiangna Chen in Technology in Cancer Research & Treatment

Supplemental Material

sj-xlsx-4-tct-10.1177_15330338261420099 - Supplemental material for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer

Supplemental material, sj-xlsx-4-tct-10.1177_15330338261420099 for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer by Jingjing Wu, Haifeng Yu, Shanshan Huang and Xiangna Chen in Technology in Cancer Research & Treatment

Supplemental Material

sj-xls-5-tct-10.1177_15330338261420099 - Supplemental material for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer

Supplemental material, sj-xls-5-tct-10.1177_15330338261420099 for A Retrospective Study of the Correlation Between Next-Generation Sequencing and Immunohistochemical Detection of TP53 in Colorectal Cancer by Jingjing Wu, Haifeng Yu, Shanshan Huang and Xiangna Chen in Technology in Cancer Research & Treatment

Footnotes

ORCID iD

Jingjing Wu

Consent to Participate

Written informed consent was obtained from all participants involved in the study.

Author Contributions

Conceptualization, J.W.; Methodology, S.H. and H.Y.; Investigation, S.H. and H.Y.; Data curation, H.Y and X.C; Formal analysis, H.Y. and X.C; Writing – original draft, J.W.; Writing – review & editing, J.W.; Visualization, J.W. and X.C. All authors have read and agreed to the published version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China, Fujian Provincial Health and Construction Commission, Young and middle-aged backbone personnel Training Project, (grant number 81902340, 2020GGA052).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of The First Affiliated Hospital, Fujian Medical University (approval no. [2015084-2]).

Supplemental Material

Supplemental material for this article is available online.

References

Köbel

Piskorz

Lee

, et al. Optimized p53 immunohistochemistry is an accurate predictor of TP53 mutation in ovarian carcinoma. J Pathol Clin Res. 2016;2(4):247–258. doi:10.1002/cjp2.53

Multiple Applications of P53 Immunohistochemistry in Gynecologic Pathology: Proceedings of the Joint Meeting of the International Society of Gynecological Pathologists at the 2020 USCAP Annual Meeting (109th United States and Canadian Academy of Pathology Annual Meeting (USCAP).

Zhang

Tong

JHM

, et al. Targeting the oncogenic p53 mutants in colorectal cancer and other solid tumors. Int J Mol Sci. 2019 Nov 28;20(23):5999. doi:10.3390/ijms20235999

Hasbullah

Musa

. Gene therapy targeting p53 and KRAS for colorectal cancer treatment: A myth or the way forward? Int J Mol Sci. 2021 Nov 3;22(21):11941. doi:10.3390/ijms222111941

Raffone

Travaglino

Cerbone

, et al. Diagnostic accuracy of p53 immunohistochemistry as surrogate of TP53 sequencing in endometrial cancer. Pathol Res Pract. 2020 Aug;216(8):153025. doi:10.1016/j.prp.2020.153025

Tessier-Cloutier

Kortekaas

Thompson

, et al. Major p53 immunohistochemical patterns in in situ and invasive squamous cell carcinomas of the vulva and correlation with TP53 mutation status. Mod Pathol. 2020 Aug;33(8):1595–1605. doi:10.1038/s41379-020-0524-1

Rabban

Garg

Ladwig

Zaloudek

Devine

. Cytoplasmic pattern p53 immunoexpression in pelvic and endometrial carcinomas with TP53 mutation involving nuclear localization domains: An uncommon but potential diagnostic pitfall with clinical implications. Am J Surg Pathol. 2021;45(11):1441–1451. doi:10.1097/PAS.0000000000001713

Von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573–577. doi:10.7326/0003-4819-147-8-200710160-00010

Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012 Jul 18;487(7407):330–337. doi:10.1038/nature11252

10.

Nagao

Koshino

Sugimura-Nagata

, et al. The complete loss of p53 expression uniquely predicts worse prognosis in colorectal cancer. Int J Mol Sci. 2022 Mar 17;23(6):3252. doi:10.3390/ijms23063252

11.

Hassin

Oren

. Drugging p53 in cancer: One protein, many targets. Nat Rev Drug Discov. 2023 Feb;22(2):127–144. doi:10.1038/s41573-022-00571-8

12.

Zhang

Liu

Zhang

Feng

. Gain-of-function mutant p53 in cancer progression and therapy. J Mol Cell Biol. 2020 Sep 1;12(9):674–687. doi:10.1093/jmcb/mjaa040

13.

Perez

Shen

Duan

, et al. Modeling the etiology of p53-mutated cancer cells. J Biol Chem. 2016 May 6;291(19):10131–10147. doi:10.1074/jbc.M116.724781

14.

Wang

Nakayama

Hong

Oshima

. Gain-of-Function p53 mutation acts as a genetic switch for TGFβ signaling-induced epithelial-to-mesenchymal transition in intestinal tumors. Cancer Res. 2024 Jan 2;84(1):56–68. doi:10.1158/0008-5472.CAN-23-1490

15.

Bellizzi

. P53 as exemplar next-generation immunohistochemical marker: A molecularly informed, pattern-based approach, methodological considerations, and pan-cancer diagnostic applications. Appl Immunohistochem Mol Morphol. 2023;31(7):507–530. doi:10.1097/PAI.0000000000001144

16.

Singh

Piskorz

Bosse

, et al. P53 immunohistochemistry is an accurate surrogate for TP53 mutational analysis in endometrial carcinoma biopsies. J Pathol. 2020 Mar;250(3):336–345. doi:10.1002/path.5375

17.

Park

Han

Choi

, et al. P53 immunohistochemistry and mutation types mismatching in high-grade serous ovarian cancer. Diagnostics (Basel). 2022 Feb 24;12(3):579. doi:10.3390/diagnostics12030579

18.

Takami

Yoshida

Fukushima

, et al. Revisiting TP53 mutations and immunohistochemistry–A comparative study in 157 diffuse gliomas. Brain Pathol. 2015 May;25(3):256–265. doi:10.1111/bpa.12173

19.

Chen

Zhang

, et al. Mutant p53 in cancer: From molecular mechanism to therapeutic modulation. Cell Death Dis. 2022 Nov 18;13(11):974. doi:10.1038/s41419-022-05408-1

20.

Sabapathy

Lane

. Understanding p53 functions through p53 antibodies. Mol Cell Biol. 2019 Apr 1;11(4):317–329. doi:10.1093/jmcb/mjz010

21.

Huang

Che

Wang

. P53/MDM2 signaling pathway in aging, senescence and tumorigenesis. Semin Cancer Biol. 2024 Jun;101:44–57. doi:10.1016/j.semcancer.2024.05.001

22.

Andrysik

Sullivan

Kieft

Espinosa

. PPM1D Suppresses p53-dependent transactivation and cell death by inhibiting the Integrated Stress Response. Nat Commun. 2022 Dec 1;13(1):7400. doi:10.1038/s41467-022-35089-5

23.

Ota

Nakao

Sawada

, et al. Δ40p53α suppresses tumor cell proliferation and induces cellular senescence in hepatocellular carcinoma cells. J Cell Sci. 2017 Feb 1;130(3):614–625. doi:10.1242/jcs.190736

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.03 MB

4.72 MB

0.02 MB

0.01 MB