TP53 disruptive mutation predicts platinum‐based chemotherapy and PD‐1/PD‐L1 blockade response in urothelial carcinoma

TP53 mutation is one of the most common genetic alterations in urothelial carcinoma (UrCa), and heterogeneity of TP53 mutants leads to heterogeneous clinical outcomes. This study aimed to investigate the clinical relevance of specific TP53 mutations in UrCa. In this study, a total of eight cohorts were enrolled, along with matched clinical annotation. TP53 mutations were classified as disruptive and nondisruptive according to the degree of disturbance of p53 protein function and structure. We evaluated the clinical significance of TP53 mutations in our local datasets and publicly available datasets. The co‐occurring events of TP53 mutations in UrCa, along with their therapeutic indications, functional effects, and the tumor immune microenvironment, were also investigated. TP53 mutations were identified in 49.7% of the UrCa patients. Within this group, 25.1% of patients carried TP53Disruptive mutations, a genetic alteration correlated with a significantly poorer overall survival (OS) when compared to individuals with TP53Nondisruptive mutations and those with wild‐type TP53. Significantly, patients with TP53Disruptive mutations exhibit an increased probability of responding favorably to PD‐1/PD‐L1 blockade and chemoimmunotherapy. Meanwhile, there was no noteworthy distinction in OS among patients with varying TP53 mutation status who underwent chemotherapy. Samples with TP53Disruptive mutations showed an enriched APOBEC‐ and POLE‐related mutational signature, as well as an elevated tumor mutation burden. The sensitivity to immunotherapy in tumors carrying TP53Disruptive mutation may be attributed to the inflamed tumor microenvironment characterized by increased CD8+T cell infiltration and interferon‐gamma signaling activation. In conclusion, UrCa patients with TP53Disruptive mutations have shown reduced survival rates, yet they may respond well to PD‐1/PD‐L1 blockade therapy and chemoimmunotherapy. By distinguishing specific TP53 mutations, we can improve risk stratification and offer personalized genomics‐guided therapy to UrCa patients. © 2024 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.


Introduction
Urothelial carcinoma (UrCa), which includes urothelial carcinoma of the bladder (UCB) and upper tract urothelial carcinoma (UTUC), is a major cause of morbidity and mortality worldwide, causing over 200,000 deaths annually [1].The survival of patients with UrCa is constrained by our current approaches to staging, surgery, and systemic treatment.Approximately half of patients succumb to this aggressive disease within 5 years, and the use of neoadjuvant and/or adjuvant chemotherapy provides only modest benefits for overall survival (OS) [2].Hence, there is an urgent need for significant progress in enhancing clinical management for these patients.Immunotherapy and targeted therapy with new agents directed at specific molecular pathways represent a promising avenue for achieving substantial progress [3].Large-scale collaborations utilizing next-generation sequencing (NGS), such as The Cancer Genome Atlas (TCGA), have already contributed to our understanding of the molecular mechanisms underlying UrCa [4].Progress in identifying and comprehending genetic alterations across the spectrum of urothelial malignancies offers hope for improved outcomes through personalized tumor molecular profiles and individualized therapy options.
The tumor suppressor gene TP53 plays a central role in the development and progression of UrCa [5].It plays biologically important roles in cell-cycle regulation, which is one of the three primary pathways frequently altered in UrCa [6].The spectrum of TP53 mutation is extensive and varies across various malignancies, including nonsense mutation, missense mutation, insertion, frameshift, and deletion, for example.Approximately 80% of these mutations occur in the DNA binding domain, with the majority being missense mutations [7,8].The crucial question for clinicians is whether all this knowledge is applicable in the management of UrCa.Previous studies suggested that TP53 alterations were enriched in certain molecular subtypes of bladder cancer and correlated with the survival of patients [9].However, in a phase III study, neither the prognostic value of p53 nor the benefit of chemotherapy in patients with p53-positive tumors was confirmed [10].Crystal structural analyses suggest that the effects of TP53 mutations on DNA binding domain may vary, leading to variable effects on tumor behavior [11].Investigations across different tumor types also suggest that the heterogeneity of TP53 mutations is associated with equally diverse clinical outcomes [12,13].Therefore, it is necessary to determine the clinical associations of specific TP53 mutations to better understand their relevance for personalized treatment approaches in UrCa.
In this study, we classified TP53 mutations into two categories based on the predicted degree of protein structure disturbance in p53-DNA complexes: disruptive and nondisruptive [13].TP53 Disruptive mutation included all DNA sequence changes that introduced a STOP sequence, leading to the disruption of p53 protein production or any DNA sequence alteration that occurred within the L2 or L3 DNA-binding domains of p53 (codons 163-195 or 236-251) involving replacement of an amino acid from one polarity/charge category with an amino acid from another category.While TP53 Nondisruptive mutation included any DNA sequence alteration that occurred outside the L2 or L3 DNA-binding domains or inside the L2 or L3 DNA-binding domains involving the replacement of an amino acid from one polarity/charge category with an amino acid from the same category [13].Thus, disruptive mutations are likely to cause loss of activity of the p53 protein, while nondisruptive variants may retain the functional properties of wild-type (WT) p53 [14].In this study, we carried out a comprehensive analysis of the clinicopathological, genomic, and immunophenotypic characteristics of the TP53-mutant UrCa, and evaluated its impact on chemotherapy, PD-1/ PD-L1 blockade therapy, and chemoimmunotherapy in patients.

Materials and methods
This study enrolled eight independent UrCa cohorts, including three local patient cohorts and five public datasets.The detailed inclusion and exclusion criteria for these cohorts are outlined in supplementary material, Figure S1A, and the patient composition for the subgroup analysis was summarized in supplementary material, Figure S1B.

Patient inclusion
For the Zhongshan Hospital Affiliated to Fudan University (ZSHS) cohort, 135 patients who received radical cystectomy at Zhongshan Hospital between 2002 and 2014, with formalin-fixed paraffin-embedded (FFPE) tumor tissue microarray (TMA) and clincal data available, were enrolled ultimately.The clinical characteristics of patients are provided in supplementary material, Table S1.For the ZSHS metastatic urothelial carcinoma (mUrCa) cohort, an additional 19 metastatic urothelial carcinoma (mUrCa) patients treated with chemotherapy, immune checkpoint inhibitors (ICIs), or both were included with panel sequencing data.For the Fudan University Shanghai Cancer Center (FUSCC) cohort, 101 UrCa patients treated in FUSCC with the FUSCC urothelial carcinoma (FUSCC-UrCa) panel [15] sequencing data were enrolled.This study was approved by the Clinical Research Ethics Committee of Zhongshan Hospital and Fudan University (No. B2015-030).Written informed consent was obtained from each patient.

Public data sets
For the Memorial Sloan Kettering Cancer Center (MSKCC) cohort [16], 1,146 cases who were pathologically diagnosed as having UrCa and with clinical data were included ultimately.For the TCGA cohort [4], the clinical and genomic information of 379 patients diagnosed with bladder cancer was obtained using TCGA-Assembler version 2.0.6 in July 2021 and was included for analysis ultimately.For the MSK-IMPACT cohort [17], among those treated with ICIs, 199 patients diagnosed with UrCa were included.The clinical and genomic information of MSKCC and MSK-IMPACT cohort was acquired from cBioPortal for Cancer Genomics.For the IMvigor210 cohort [18], 274 mUrCa patients treated with PD-L1 inhibitor atezolizumab with matched clinical data were included.The clinical and RNA sequencing (RNA-seq) data were obtained using the IMvigor210CoreBiologies R package from http:// research-pub.gene.com/IMvigor210CoreBiologies.For the NCT03179943 cohort [19], 20 patients treated with guadecitabine and atezolizumab, with RECIST response data, were ultimately included.

TP53 mutation classification criteria
TP53 mutations were divided into disruptive and nondisruptive mutations according to the degree of disturbance of p53 protein function and structure, as previously described [13].(1) The TP53 Disruptive mutations included all DNA sequence changes that introduced a STOP sequence, leading to the disruption of p53 protein production or any DNA sequence alteration that occurred within the L2 or L3 DNA-binding domains of p53 (codons 163-195 or 236-251) involving the replacement of an amino acid from one polarity/charge category with an amino acid from another category.TP53 homozygous deletions were also treated as TP53 Disruptive mutations, given their similar biological function, leading to the disruption of p53 protein production.This inclusion applied to six cohorts with available copy number variant (CNV) data, including MSKCC, TCGA, MSK-IMPACT, IMvigor210, ZSHS mUrCa, and FUSCC cohorts.Notably, only the NCT03179943 cohort did not provide CNV data for the patients.(2) The TP53 Nondisruptive mutations included any DNA sequence alteration that occurred outside the L2 or L3 DNA-binding domains or inside the L2 or L3 DNA-binding domains involving replacement of an amino acid from one polarity/charge category with an amino acid from the same category.Refer to the Supplementary materials and methods for additional details of TP53 mutation classification.

p53 immunohistochemistry (IHC)
IHC using p53 antibody (Catalogue No. YM6900, Immunoway, Plano, TX, USA) was conducted following established protocols for the ZSHS cohort [20], and the cases were classified into one of three subgroups: (1) p53 WT: tumors with 1% to 80% of nuclear staining in tumor cells, typically with varying intensity; (2) p53 null: no tumor cell nuclear staining with a positive internal control; and (3) p53 overexpression (OE): tumors with uniform and intense nuclear staining in at least 80% of estimated tumor cell nuclei.Typical images of p53 IHC were presented in supplementary material, Figure S2.

IHC and assay methods
Serial tissue sections fixed in formalin and embedded in paraffin were used to construct TMAs, and the antibodies used for IHC staining are summarized in supplementary material, Table S2.For the ZSHS cohort, IHC staining for specific immune cells, such as CD8 + T cells, CD4 + T cells, macrophages, and other molecules, was carried out according to previous protocols [21].Tertiary lymphoid structures (TLSs) were qualified and quantified using both H&E and CD3/CD20 double IHC staining employing a published scale [22].As previously, the infiltration density of immune cells was evaluated as the mean value of cells/HPF under three randomized high-power field (HPF, magnification Â200) independently by two pathologists who were blind to the clinical information with NanoZoomer-XR (Hamamatsu Photonics, Hamamatsu, Japan) and Image Pro plus 6.0 (Media Cybernetics, Rockville, MD, USA) digitally.

Processing of genomic and transcriptomic data
For transcriptomic analyses, RNA-seq data were mined along with the process of acquiring clinical information and were normalized through the formula log 2 (FPKM+1).The infiltration score of immune cells, T cell-inflamed signature, T effector signature, IFNγ-related gene signature, and other immune-related signatures were calculated by single sample Gene Set Enrichment Analysis (ssGSEA) algorithms based on related gene expression, as previously reported (supplementary material, Table S3) [23][24][25][26][27][28][29].Moreover, immune scores were evaluated by an estimate method [30].For genomic analyses, panel DNA sequencing was conducted in our local ZSHS mUrCa and FUSCC cohorts.

Statistical analyses
Kaplan-Meier analysis and log-rank tests were performed to conduct survival analyses.Univariate and multivariate analyses were conducted using the Cox proportional hazards regression model.Multivariate analysis was adjusted for tumor stage defined by the American Joint Committee on Cancer (AJCC), age, gender, tumor mutation burden (TMB), and DNA-damage response (DDR) gene panel, which could be confounders alongside TP53 mutational status in the MSKCC and TCGA combined cohort.A similar analysis was also performed in the ZSHS cohort after adjusting for stage, age, and gender.Pearson's χ 2 test and Fisher's exact test were applied to compare categorical variables.Statistical p values were computed using Mann-Whitney U-test, and detailed statistical tests are described in the corresponding figure legends.The statistical analysis was two-tailed, and p < 0.05 was considered statistically significant.All analyses were conducted using IBM SPSS Statistics version 20.0 (IBM Corp., Armonk, NY, USA) and R version 4.1.2software.

TP53 Disruptive mutation predicts poor OS in urothelial carcinoma
To investigate the correlation between OS and TP53 status, a total of 1,660 UrCa patients from three cohorts were analyzed.Of 1,525 patients with DNA sequencing and clinical data from the MSKCC and TCGA combined cohort, 363 (23.8%) and 395 (25.9%) patients presented TP53 Disruptive mutations and TP53 Nondisruptive mutations, respectively.As illustrated, patients with TP53 Disruptive mutations (median OS 54.9 months) had the worst prognosis compared to those with TP53 Nondisruptive mutations (median OS 97.1 months) or TP53 WT (median OS 92.9 months) (Figure 1A, log-rank p = 0.007).In the ZSHS cohort, we conducted additional survival analysis using p53 IHC data.As previously TP53 disruptive mutation in urothelial carcinoma 141 described [13,20], IHC staining was utilized to distinguish p53 null staining, p53 OE, and p53 WT.Of 135 patients from the ZSHS cohort, 30 (22.2%) were grouped into the p53 null subgroup.Similar to the TP53 Disruptive mutation subgroup, the p53 null subgroup patients had a significantly worse OS (median OS 50.0 months) compared with p53 OE (median OS 106.0 months) or p53 WT patients (median OS 94.0 months) in the ZSHS cohort (Figure 1B, log-rank p = 0.048).Notably, p53 IHC staining patterns (p53 null, p53 OE, and p53 WT) were found to correspond concordantly with the TP53 Disruptive mutation, TP53 Nondisruptive mutation, and TP53 WT status, as determined previously by DNA sequencing [20,[31][32][33].Additionally, we also conducted a more extensive analysis of the correlation between TP53 mutational status and p53 protein expression by IHC in the FUSCC cohort.The result showed that the overall concordance rate between p53 protein expression and TP53 mutational status was 92.1% (supplementary material, Figure S3A).Among the cases with mutated TP53, cases with disruptive mutations were generally p53 null (90.9%), and cases with nondisruptive mutations were generally p53 OE by IHC (95.2%) (supplementary material, Figure S3B).This collective evidence indicates that p53 IHC performs effectively as a surrogate test for TP53 mutations in UrCa.Additionally, univariate and multivariate Cox regression analysis showed that TP53 Disruptive mutation could serve as a potential independent prognostic factor for survival outcomes in the MSKCC and TCGA combined cohort (supplementary material, Figure S4A).A similar analysis was also performed in our local ZSHS cohort for additional validation.After adjusting for stage, age, and gender as confounders, p53 null also correlated with a worse prognosis (supplementary material, Figure S4B).

Correlation between chemotherapy benefit and TP53 status in urothelial carcinoma
Cisplatin-based chemotherapy has served as the standard of care in UrCa for the past 50 years.We next explore whether TP53 or p53 status could indicate chemotherapeutic benefit.Patients harboring either TP53 Disruptive mutations or TP53 Nondisruptive mutations could not enjoy OS benefits from platinum-based chemotherapy in the MSKCC and TCGA combined cohort (supplementary material, Figures S4C and S5A).Then we compared patients from the MSKCC and TCGA combined cohort who survived for over 5 years (defined as long-term responders) and succumbed within 6 months (defined as refractory responders).Notably, we found that only 22% and 17% of refractory responders harbored TP53 Disruptive mutations and TP53 Nondisruptive mutations, while 40% and 17.5% of long-term responders harbored TP53 Disruptive mutations and TP53 Nondisruptive mutations, respectively (Figure 1C).Consistent with the previous trial, the predictive value of p53 was not confirmed in patients treated with platinum-based chemotherapy from our local ZSHS cohort (supplementary material, Figures S4D and S5B).By redefining patients who succumbed within 12 months in the ZSHS cohort as refractory patients, we found that a higher proportion of refractory patients exhibited p53 WT while 21.2% and 39.4% of long-term responders gained p53 null and OE status, respectively (Figure 1D).
To further investigate the correlation between chemotherapy benefit and TP53 status, we enrolled another 11 mUrCa patients treated by chemotherapy.The results showed that mUrCa patients with TP53 Disruptive mutations had an OS similar to that of TP53 WT patients, while the TP53 Nondisruptive mutation was not detected in the ZSHS mUrCa chemotherapy subcohort (supplementary material, Figure S5C).

TP53 Disruptive mutation predicts PD-1/PD-L1 blockade and chemoimmunotherapy benefit in urothelial carcinoma
We then explored the link between TP53 status and therapeutic response to immune checkpoint inhibitors (ICIs) alone or in combination with chemotherapy in UrCa.By analyzing patients treated with ICIs from the combined urothelial carcinoma immunotherapy cohort, we found that patients with TP53 Disruptive mutations (median OS: not reached) showed significantly better OS compared to those with TP53 Nondisruptive mutations (median OS: 24.0 months) or TP53 WT (median OS: 15.0 months) (Figure 2A, log-rank p = 0.017; supplementary material, Figure S6A,B for single cohort analysis).Moreover, 31.5% of patients with TP53 Disruptive mutations achieved either complete response (CR) or partial response (PR) following treatment with ICIs, while the response rates were 23.9% for patients with TP53 Nondisruptive mutations and 22.0% for patients with TP53 WT tumors (Figure 2B and supplementary material, Figure S6C-E).Significant therapeutic response was observed in a clinical case of a 70-year-old patient (FUSCC-P077) diagnosed with mUrCa harboring a TP53 Disruptive mutation (p.H178fs).Following 26 cycles of pembrolizumab, a majority of the previously affected systemic metastatic lymph nodes displayed remission, as indicated by PET/CT scans (Figure 2C).
Analyzing patients treated with chemoimmunotherapy from combined urothelial carcinoma chemotherapy plus ICI cohort, we observed that patients with TP53 Disruptive mutations (median OS: not reached) exhibited significantly prolonged OS compared to those with TP53 Nondisruptive mutations (median OS: 20.1 months) or TP53 WT (median OS: 23.2 months) (Figure 2D, Log-rank p = 0.095; supplementary material, Figure S7A,B for single cohort analysis).Among patients with assessed therapeutic response, those with TP53 Disruptive mutations achieved the highest response rate (CR/PR rate), 37.5%, following treatment with chemoimmunotherapy (Figure 2E and supplementary material, Figure S7C-E).A notable therapeutic response was indicated in a clinical case involving a 77-year-old patient (FUSCC-P103) from the FUSCC cohort with TP53 Disruptive mutations (p.D182N, p.R280T, p.Q331X) who received a long-term clinical response after three cycles of gemcitabine and carboplatin (GC) plus tislelizumab (Anti-PD-1).Both primary lesion in bladder and metastatic lymph nodes achieved CR after chemoimmunotherapy (Figure 2F).Neither of the exemplar cases was deficient mismatch repair (dMMR) or microsatellite instability-high (MSI-H) (supplementary material, Table S4).Similar results were also observed in eight UrCa patients treated upon immunotherapy from the ZSHS mUrCa immunotherapy subcohort (supplementary material, Figure S7F).

Correlation between genomic feature and TP53 status in urothelial carcinoma
The impact of TP53 status on the genomic feature of UrCa patients remains unclear.Based on multicohort analyses, we noticed that TP53 Disruptive mutations led to elevated TMB (median TMB 11.0 mut/MB) compared to TP53 Nondisruptive mutations (median TMB 9.0 mut/MB) and TP53 WT (median TMB 6.0 mut/MB) (Figure 3A and supplementary material, Figure S8A).Elevated tumor neoantigen burden (TNB) was also observed in the TP53 Disruptive mutation subgroup (supplementary material, Figure S8B).As different mutational processes may have different propensities for neoantigen generation, we evaluated the mutational signatures and found a higher level of APOBEC-and POLE-related mutational signature in patients with TP53 Disruptive mutations (Figure 3B).
We then conducted additional investigations to determine the mutation rates of key genes associated with TP53 status in UrCa.ERBB2, CDK12, E2F3, and KMT2D mutations co-occurred with TP53 Disruptive mutations while RB1 was the top co-mutated gene with TP53 Nondisruptive mutations.Murine double minute 2 (MDM2), fibroblast growth factor receptor 3 (FGFR3), and genes lying in 9p21, including CDKN2A, CDKN2B, and MTAP, were mutual exclusive to TP53 mutations (Figure 3C and supplementary material, Figure S8C).Focusing on the distribution of typical CNVs, we observed that patients with TP53 Disruptive mutations had the lowest rate of homozygous deletion at 9p21.3 and the highest rate of focal amplification at 11q13.3 (Figure 3D).In cases where The red circle highlights a target lesion and its evolution.CR, complete response; OS, overall survival; PR, partial response; UrCa, urothelial carcinoma; mUrCa, metastatic urothelial carcinoma; FUSCC, Fudan University Shanghai Cancer Center; ZSHS, Zhongshan Hospital Affiliated to Fudan University; TP53 Dis mut, TP53 Disruptive mutation; TP53 Nondis mut, TP53 Nondisruptive mutation; WT, wild type; ICI, immune checkpoint inhibitor; GC, gemcitabine plus cisplatin.

144
K Jin, J Xu, X Su et al patients with TP53 WT or TP53 Nondisruptive mutations showed limited response to chemotherapy or immune checkpoint inhibitors, alternative therapeutic options and the presence of actionable alterations were explored.For instance, based on data from the OncoKB database, additional targets such as FGFR3 were identified, offering potential therapeutic avenues (supplementary material, Figure S8D).

TP53 Disruptive mutation facilitates immune activation in urothelial carcinoma
Immunophenotyping was performed to gain further biological insights into the immunologic features across TP53 status.Along with the increased neoantigen burden, patients harboring TP53 Disruptive mutations have the highest immune score, functional signatures including T-cell-inflamed signature, IFNγ-related signature, T effector signature, and TLS signatures in TCGA cohort.Moreover, there is an increase in the infiltration scores of effector cells, including activated CD8 + T cells, cytotoxic lymphocytes, type 1 helper T (Th1) cells, type 1 macrophage (M1), natural killer (NK) cells, and effector molecules, including granzyme A (GZMA), granzyme B (GZMB), and interferon-gamma (IFN-γ) in patients harboring TP53 Disruptive mutations (Figure 4A and supplementary material, Figure S9A).Immune suppressive cells like regulatory T (Treg) cells, type 2 helper T (Th2) cells, type 2 macrophages (M2), and neutrophils were found to have decreased in the TP53 Disruptive mutation subgroup (supplementary material, Figure S9A).Similar results were also found in the IMvigor210 cohort (supplementary material, Figure S9B).In line with findings from the TCGA and IMvigor210 cohorts, patients with p53 null status were associated with a higher proportion of TLS infiltration and an inflamed tumor microenvironment in the ZSHS cohort (Figure 4B,C).In addition, higher infiltration of CD8 + T cells and IFN-γ-producing cells was observed in patients with p53 null (Figure 4D).IHC analysis further validated the increased immune effector cells and immune checkpoints in patients with p53 null (supplementary material, Figure S9C,E).TP53 disruptive mutation in urothelial carcinoma 145 Conversely, immune suppressive cells exhibited decreased infiltration in the p53 null subgroup (supplementary material, Figure S4D,F).Moreover, patients with p53 null had the highest expression of PD-L1 in the ZSHS cohort (Figure 4E).

Discussion
In UrCa, TP53 mutation is the most common genomic alteration [4].Despite previous studies revealing the association of TP53 mutation with a worse prognosis in UrCa, its heterogeneity has been overlooked until now [34].Herein, we classified TP53 mutations into disruptive and nondisruptive types and found that patients with TP53 Disruptive mutations had a worse OS compared with patients with TP53 WT or TP53 Nondisruptive mutations in UrCa.Our findings align with previous reports that emphasized distinct clincal outcomes associated with different TP53 mutations in head and neck squamous cell carcinoma and diffuse large B-cell lymphoma (DLBCL) [13,35].Remarkably, UrCa patients with TP53 Disruptive mutations show a higher likelihood of responding positively to PD-1/PD-L1 blockade and chemoimmunotherapy.Additionally, despite 40% of long-term responders to chemotherapy in the MSKCC and TCGA combined cohort, there was no significant difference in OS among patients undergoing chemotherapy with varying TP53 mutation statuses.These findings provide insights into more precise patient stratification in the future.Given the variations in TP53 mutations, a comprehensive investigation is necessary to ensure their optimal utilization as biomarkers for treatments.Our study represents the first systematic evaluation of the correlation between disruptive and nondisruptive TP53 mutations and the response of UrCa patients to chemotherapy, immunotherapy, and chemoimmunotherapy.Platinumbased chemotherapy is currently considered the standard of care as a first-line treatment for advanced or metastatic UrCa.However, a considerable proportion of patients are unresponsive to this form of therapy [36].Cisplatin causes damage to DNA by creating crosslinks within and between DNA strands.We hypothesize that tumors with mutations in DNA damage repair (DDR) genes are more sensitive to cisplatin due to their inherent deficiency in DNA repair.The frequency of DDR mutations is high and strongly correlated with response to chemotherapy in bladder cancer patients [37].Studies have demonstrated that mutations in DDR genes, such as ATM, RB1, or FANCC, are associated with partial response, progression-free survival, and OS in patients undergoing cisplatin-based chemotherapy [38].Patients with at least one mutation had an 85% 5-year survival rate, compared to 45% for patients without any mutations [39].Furthermore, the DDR gene ERCC2 has been linked to pathological response to cisplatinbased neoadjuvant chemotherapy [40,41].It has been observed that tumors with ERCC2 mutations exhibit a distinct trinucleotide mutation signature, which suggests that transcription-coupled repair is the primary mechanism affected in ERCC2-mutated cancers [42].With regard to TP53 Disruptive mutation, it could drastically reduce the DNA repair capacity of tumor cells and lead to the response to platinum-based chemotherapy [43].This might be partly supported by our findings that more long-term responders were observed in patients carrying TP53 Disruptive mutations, although there was no significant difference in OS between patients with different TP53 mutation statuses receiving chemotherapy.It is of clinical use to determine the association between TP53 Disruptive mutations and the response to treatments that target DNA damage repair pathways [44].
In our analysis, we found that the TMB level and PD-L1 expression, which have been established as reliable predictive biomarkers for ICIs in mUrCa [36], displayed distinct distributions among samples with varying TP53 status.In numerous cancer types, TP53 mutations have been shown to be associated with increased genetic instability [45].Our findings revealed that samples harboring TP53 Disruptive mutations exhibited a high TMB level accompanied by the enrichment of APOBEC-and POLE-related mutational signatures.The KMT2D mutation is significantly enriched in the samples harboring TP53 Disruptive mutations, consistent with the finding that KMT2D deficiency sensitizes UrCa tumors to ICI [46].In contrast, the genomic alteration profiles of the other two subgroups were markedly distinct.RB1 mutations co-occurred with the TP53 Nondisruptive mutations, whereas the frequencies of CDKN2A/B alterations and homozygous deletion of 9p21.3 were significantly increased in TP53 WT patients, which confer a cold tumor immune microenvironment and primary resistance on ICIs [47].As expected, a robust inflamed immune microenvironment was demonstrated in patients with TP53 Disruptive mutations, manifested by enriched infiltration of CD8 + T cells and activation of IFN-γ signaling.Significantly, patients with TP53 Disruptive mutations exhibited a higher level of PD-L1 expression compared to other patients.These factors may account for the responsiveness to immunotherapy in patients harboring TP53 Disruptive mutations.
Although our findings were derived from a retrospective analysis, they provide evidence in favor of tailored treatment strategies that take into account the TP53 status of the patient.Our data indicate that patients with TP53 Disruptive mutations might be the most suitable candidates for ICIs and chemoimmunotherapy.We also found that ERBB2 alterations were substantially increased in patients with TP53 Disruptive mutations, so HER2 antibody-drug conjugate could be added as a rational alternative strategy in this subpopulation [48].For patients with TP53 WT or TP53 Nondisruptive mutations retaining p53 expression, MDM2 inhibitors are the most extensively pursued therapy approach to prevent p53 degradation [49].We also found an increase in FGFR3 alterations in patients with TP53 WT, suggesting that FGFR inhibitors could be considered to enhance the prognosis of these patients [50].
Our study has several limitations.First, our analysis of retrospective cohorts is subject to sampling biases and institutional differences as to how patients are evaluated and genomically profiled.In publicly accessible datasets, comprehensive treatment details are only available for a limited subset of patients.Consequently, the therapeutic analysis may not encompass all patients who have indeed undergone treatment.Additionally, assessment of p53 protein expression levels using IHC in the ZSHS cohort does not fully mimic the TP53 sequencing data, although concordance between TP53 mutation status and IHC reached an overall agreement of 92.1% in our research.
To sum up, our study highlighted the potential of categorizing TP53 mutations as either disruptive or nondisruptive in determining the prognosis of UrCa and expanded the repertoire of biomarkers that can be utilized for treatment selection.These findings could have far-reaching implications for clinical practice, the design of clinical trials, and our understanding of the underlying mechanisms that drive UrCa oncogenesis.

Figure 1 .
Figure 1.Correlation between prognosis, chemotherapy benefit, and TP53 status in urothelial carcinoma.(A) Kaplan-Meier analyses of OS for patients with urothelial carcinoma according to TP53 status by DNA sequencing in MSKCC and TCGA combined cohort.(B) Kaplan-Meier analyses of OS for patients with UrCa according to p53 status by IHC in the ZSHS cohort.(C) The swimmer plot describes the OS of each UrCa patient harboring different TP53 mutations who survived for over 5 years or succumbed within 6 months from MSKCC and TCGA combined cohort.(D) The swimmer plot describes the OS of each UrCa patient harboring a different p53 protein expression who survived for over 5 years or succumbed within 6 months from ZSHS cohort.IHC, immunohistochemistry; UrCa, urothelial carcinoma; TP53 Dis mut, TP53 Disruptive mutation; TP53 Nondis mut, TP53 Nondisruptive mutation; WT, wild type; OE, overexpression; OS, overall survival; MSKCC, Memorial Sloan Kettering Cancer Center; TCGA, The Cancer Genome Atlas; ZSHS, Zhongshan Hospital Affiliated to Fudan University.

Figure 2 .
Figure 2. TP53 Disruptive mutation predicts PD-1/PD-L1 blockade and chemoimmunotherapy benefit in urothelial carcinoma.(A) Kaplan-Meier analyses of OS according to TP53 status in mUrCa patients treated with anti-PD-1/PD-L1 immune checkpoint inhibitors from MSK-IMPACT, FUSCC, and ZSHS mUrCa combined cohort.(B) Immunotherapeutic response to immune checkpoint inhibitors according to TP53 status in mUrCa patients from IMvigor210 and FUSCC combined cohort.(C) PET/CT scan showed typical PR imaging alteration in a patient with a TP53 Disruptive mut (p.H178fs) from FUSCC cohort after 26 rounds of pembrolizumab treatment.The white arrows highlight metastatic lymph nodes and their evolution.The red circle highlights a target lesion and its evolution.(D) Kaplan-Meier analyses of OS based on TP53 status in mUrCa patients treated with chemotherapy plus immunotherapy from NCT03179943, FUSCC, and ZSHS mUrCa combined cohort.(E) Clinical response to chemotherapy plus immunotherapy according to TP53 status in mUrCa patients from NCT03179943 and FUSCC combined cohort.(F) PET/CT and abdominal and pelvic CT scans showed typical CR imaging alteration of a patient with TP53 Disruptive mut (p.D182N, p.R280T, p.Q331X) from FUSCC after three cycles of GC plus tislelizumab.The red circle highlights a target lesion and its evolution.CR, complete response; OS, overall survival; PR, partial response; UrCa, urothelial carcinoma; mUrCa, metastatic urothelial carcinoma; FUSCC, Fudan University Shanghai Cancer Center; ZSHS, Zhongshan Hospital Affiliated to Fudan University; TP53 Dis mut, TP53 Disruptive mutation; TP53 Nondis mut, TP53 Nondisruptive mutation; WT, wild type; ICI, immune checkpoint inhibitor; GC, gemcitabine plus cisplatin.

Figure 4 .
Figure 4. TP53 Disruptive mutation facilitates immune activation in urothelial carcinoma.(A) Heatmap illustrating immune functional signatures and immune effector cells of TCGA cohort to detect the correlation between TP53 status and immune microenvironment.(B and C) Correlation between (B) TLS level, (C) immune phenotype, and p53 status in ZSHS cohort.(D) Correlation between CD8 + T cell infiltration (left), IFN-γ + cell infiltration (right), and p53 status in ZSHS cohort.(E) Correlation between PD-L1 level and p53 status in ZSHS cohort.WT, wild type; OE, overexpression; IC, immune cell; TCGA, The Cancer Genome Atlas; ZSHS, Zhongshan Hospital Affiliated to Fudan University; TLS, tertiary lymphoid structures.