Pathology Focused Review of Morphologic Subtypes and Molecular Variants of Urothelial Carcinoma with an Emphasis on Clinical/Treatment Relevance

1. Introduction / Review Abstract

Urothelial carcinoma (UC) has significant morbidity, mortality, and remains the most financially costly carcinoma to manage and treat^{1, 2}. This review will cover special morphologic features of UC that may be noted by the pathologist and any subsequent significance in terms of clinical management or treatment considerations as mentioned or recommended in the latest WHO 2022 classification of GU tumors³. Many important potentially therapy altering morphologic findings can be consistently identified and reported on routine microscopic examination of hematoxylin and eosin (H&E) stained slides. Furthermore, there has been a rapid advancement of molecular diagnostics and tailored therapies throughout oncology, and we will briefly highlight some of these as they relate to the management of UC. We will actively attempt to limit the discussion of histologic descriptions or pathologic diagnostic criteria of these entities and focus rather on the recognition of their importance/implication for clinicians who must make clinical management decisions based upon these findings. The 2022 consensus opinions of the International Society of Urologic Pathology (ISUP) recommend that pathologists always report divergent differentiation / histologic subtypes regardless of amount (even <5%) and that the report should also include enumeration of the volume of each type Finally, the importance of open lines of communication with the pathologists who review clinical specimens as well as their practice and reporting methods cannot be overstated.

2. Pathology Specific Considerations in Urothelial Carcinoma (Summarized in Table 1)

Table 1:

Summary of urothelial carcinoma histologic subtypes and divergent differentiations.

Histologic Subtype / Divergent Differentiation	Predicted Clinical Behavior (Compared to Conventional Pure Urothelial Carcinoma)	Comments on Important Features	Additional Testing Considerations	Clinical Management Considerations
Plasmacytoid	Inferior	Frequent LVI, positive margins, peritoneal involvement	Truncating somatic CDH1 alterations	Inferior cisplatin response, Consider early surgery in pT1
Sarcomatoid	Inferior	May have heterologous elements confused for sarcoma		Consider early surgery in pT1
Micropapillary	Inferior	Frequently Her2 Amplified	Her2 IHC/FISH	Consider early surgery in pT1, Questionable role for anti-Her2 therapy, Poor BCG Response
Nested	Equivalent, possibly inferior	Low-grade histology easily mistaken for benign lesions	TERT promoter status may help prove carcinoma	Equivocal diagnosis on bland nested lesions may require (deeper) re-biopsy
Lipid-Rich	Limited data, likely inferior	Very uncommon
Lymphoepithelioma-like	Conflicting data, possibly equivalent	No EBV association	PDL1	Very high rates of PDL1 expression, may benefit from PDL1 targeted therapy
Clear Cell (UC)	Equivalent	Many confusing/overlapping naming conventions with kidney, clear cell of urinary tract, and GYN associated tumors	Urothelial differentiation can be supported with IHC	Best to confirm exactly what type of "clear cell" tumor if unclear
Giant Cell	Limited data, likely inferior	Typically admixed with other components
Poorly Differentiated	Dependent upon ultimate lineage		Panel of IHC or genomic testing may be necessary to prove urothelial lineage	Clinical behavior and response to therapy will be most strongly related to cell of origin
Squamous Differentiation (UC)	Equivalent	Most common form of divergent differentiation (30–40%)		Greater percentage of squamous differentiation suggests decreased cisplatin response
Glandular Differentiation (UC)	Equivalent	Second most common form of divergent differentiation (15–20%)	Genomic characterization may help distinguish histologically ambiguous cases	Consider also urachal, prostate, colon, etc.
Trophoblastic Differentiaiton (UC)	Limited data, likely inferior	Morphologic and tumor marker (B-HCG) overlap with choriocarcinoma	Serum B-HCG may be helpful monitoring tool	Correlate with age to help distinguish UC (older) from choriocarcinoma (younger)

2.1: Urothelial Carcinoma Histologic Subtypes

Pathologists and clinicians alike have long recognized the importance of muscle invasive bladder cancer as a trigger point for escalated aggressive intervention, and yet several other features can be detected at the time of initial diagnosis that may offer additional valuable prognostic or diagnostic information. UC subtypes are, by definition, high-grade and diagnosed by their consistent morphologic appearance or growth pattern. Histologic subtypes are present in approximately 20% of UC cases⁴. The World Health Organization (WHO) Classification of Urothelial Tumors, the International Collaboration on Cancer Reporting (ICCR), and College of American Pathologists (CAP) require or strongly recommend that the diagnostic pathologist not only comment on the presence but also estimate the quantity of these subtypes due to their potential impact on treatment or prognosis^{3, 5}. However, not all newly “required” changes are immediately adopted into daily practice and we highlight that nearly half (44%) of cases initially verified by community pathologists did not document histologic subtypes upon re-review at an academic center with sub-specialty GU pathologists⁴. Another study examined the change in practice over time at the same institution with roughly similar findings, n=1211 cystectomy cases from 1980–2005 were re-reviewed in 2013. In this study, 33% of pure UC cases were reclassified as containing variant histology on this decades later re-review – with significant impact to resulting survival curves ⁶. These discrepancies highlight the importance of a close and vocal working relationship between the pathologist and clinician as well as a familiarity with pathologist’s current reporting patterns and careful interpretation of historical data considering the steady advancement of medical knowledge.

Plasmacytoid, sarcomatoid, and micropapillary subtypes are considered high risk in non-muscle invasive UC and current NCCN guidelines recommend early cystectomy for pT1 lesions with these features⁷. The other histologic subtypes are suggested to be more aggressive but with more controversial or ambiguous management recommendations.

Plasmacytoid (pUC):

Compared to conventional UC, pUC is an aggressive subtype that presents at higher stage with increased risk of lymphovascular invasion, positive lymph nodes, and a nearly 5x higher risk of positive margins at the time of cystectomy⁸. There is also a strong propensity for growth along tissue planes with early involvement of the peritoneum and frequent peritoneal carcinomatosis ⁹. This subtype invades as single cells, sometimes without a dominant mass, and may be challenging to identify by imaging or intraoperatively – hence the increased risk of positive surgical margins at excision. Given the inability to accurately preoperatively stage disease early cystectomy should be considered in pT1 cases. pUC has inferior response to chemotherapy. Studies with neoadjuvant platinum-based regimens have shown inferior response, both in terms of complete pathologic response rates and overall survival, as compared to conventional UC¹⁰. A majority of pUC (84%) demonstrate somatic truncating CDH1 alterations which correlate well with aberrant E-cadherin immunohistochemical expression (73%)^{11, 12}. These molecular and IHC findings may be very helpful in diagnostically challenging cases but are certainly not required (or even necessary) for routine diagnosis or recognition of this important subtype.

Sarcomatoid (sUC):

sUC is an aggressive subtype that histologically mimics sarcoma and may have heterologous elements mimicking differentiated sarcomas such as osteosarcoma, rhabdomyosarcoma, chondrosarcoma, or angiosarcoma. This component is more likely to be admixed with other aggressive subtypes but can also appear in a pure form which may be particularly diagnostically challenging for the pathologist to reach a carcinoma diagnosis. Keratin IHC may prove valuable in establishing a carcinomatous lineage, GATA3 IHC may help establish urothelial origin. sUC is an aggressive subtype with worse overall survival, but with similar neoadjuvant response rates, as compared to pure UC¹³. Early cystectomy should be considered in pT1 cases.

Micropapillary (mUC):

mUC is yet another aggressive subtype with marked propensity for early lymphovascular invasion. Early cystectomy should be considered in pT1 cases. ERBB2 (Her2) has been shown to be frequently overexpressed/amplified in mUC at a rate of approximately 3.0–3.5x that compared to conventional UC^{14, 15}. While earlier investigations into directed anti-Her2 therapies against conventional UC were not particularly promising, the higher levels of protein expression as well as higher rate of Her2 positivity in mUC may be a promising target in the near future¹⁵. Additional larger studies on the role of HER2 directed therapy in mUC are needed and this admittedly remains a controversial area.

Nested (nUC):

nUC has a low-grade appearance that mimics normal features of the urothelial tract – yet has a very aggressive clinical behavior discordant from its histologic appearance with a high likelihood of metastasis at initial presentation ¹⁶. The appearance of nUC by its very nature is challenging for pathologists to recognize and diagnose, particularly on superficial biopsy. A pathology report suggesting equivocal features of nUC should prompt considerable clinical concern and consideration of possible re-biopsy with more generous (deeper) sampling. In particularly challenging cases the morphology may be essentially uninformative and molecular methods examining TERT promoter status may be required to reach a definitive diagnosis¹⁶.

Lipid-rich (lrUC):

A subtype with probable poor prognosis and increased cytoplasmic lipid; data on management differences is much more limited or less convincing than other subtypes¹⁷.

Lymphoepthelioma-like (lelUC):

A mimic of the nasopharyngeal tumor of the same name, but unique in that LELUC carry no association with EBV¹⁸. There is some controversy in the literature regarding its overall prognosis, however studies suggest markedly increased PDL1 expression (93% of cases) and the potential for treatment benefit with PDL1 inhibitors in the appropriate clinical context¹⁹.

Clear cell (ccUC):

We must stress the distinction between ccUC (also known as: glycogen rich UC) from clear cell adenocarcinoma of the urinary tract which is more aggressive and molecularly distinct³. This is an area with overlapping naming that can lead to confusion and clarification with the pathologist should be considered if unclear in the report. To be clear, ccUC should be managed as a urothelial carcinoma.

Giant Cell (gcUC):

gcUC classically presents as a minor component within conventional UC, it is very uncommon, and suggests a worse prognosis with aggressive behavior²⁰.

Poorly Differentiated (pdUC):

pdUC lacks morphologic features that can be recognized by the pathologist and requires immunohistochemical markers to point to a cell of origin and lineage³. Many sensitive and specific IHC stains exist for this purpose, however, many truly poorly differentiated carcinomas will lose these markers of differentiation and the pathologist may be unable to definitively classify these lesions. In this setting, comprehensive molecular testing and imaging correlation should be considered as critical adjuncts.

It is helpful to highlight a few pitfalls with pdUC regarding involvement of the bladder by dedifferentiated tumors from adjacent organs. In our experience at a tertiary academic center overreading community pathology diagnoses prior to treatment we have seen relatively frequent misdiagnoses of poorly differentiated prostate carcinoma, HPV+ cervical carcinoma, and poorly differentiated colorectal carcinoma mistaken for UC. These were admittedly advanced and challenging cases from a pathologic perspective, however in most cases additional history (remote history of radiated prostate carcinoma) and/or additional physical examination (fungating and bleeding cervical mass) were clinically known but not communicated to the pathologist. The diagnosis of a muscle invasive UC in patients with prior pelvic/lower abdominal carcinomas and particularly in women at an atypically young age (40–50’s) should prompt suspicion to rule out other tumors. Cervical carcinoma can be particularly challenging as the morphology and IHC profile can overlap significantly with UC – a thorough review of systems, past medical history, and review of PAP smear and/or bimanual GYN examination should be considered by the clinical treatment team.

2.2: Urothelial Carcinoma with Divergent Differentiation

It is common for UC to assume morphology that may resemble other tissues – squamous, glandular (vs urachal), and/or trophoblastic. When classically appearing UC or precursor lesions are absent (non-invasive papillary UC or UC in situ) the pathologist will be unable to give a specific diagnosis and can only be descriptive – Ex: “Muscle invasive carcinoma with glandular features.” In these situations, additional IHC staining is typically not helpful. These tumor express markers consistent with their appearance. Thankfully, the most common presentation is that only a fraction of the UC tumor will express divergent differentiation. The type and relative volume may result in variable clinical management. If not currently doing so, pathologists should be encouraged by their clinical counterparts to report divergent differentiation and estimate a relative percentage of the tumor displaying non-classical UC features. Ex: “High grade urothelial carcinoma with focal squamous differentiation (10% of tumor volume).”

UC with Squamous Differentiation

Squamous differentiation (SqDif) is the most common type of divergent differentiation and may be seen in as many as 30–40% of cases^{6, 21–23}. On limited biopsy or TUR specimens a tumor may appear entirely squamous, however it is important to realize this does not preclude a UC with extensive SqDif because a diagnosis of pure squamous cell carcinoma of the urothelial tract can only be made at the time of definitive cystectomy (and is much more uncommon). UC with SqDif does not necessarily portend a worse prognosis than classical UC when stage matched, however extensive SqDif does reliably predict poor response to neoadjuvant cisplatin-based chemotherapy with 5x lower rates of downstaging and 2.5x lower rates of complete pathologic response ^{21, 22, 24, 25}.The omission of neoadjuvant therapy and earlier surgical intervention could be considered by the clinical team in such situations. It is unclear exactly what percentage of the tumor displaying SqDif by morphology may be significant in these situations as various cutoffs have been used. In challenging or equivocal cases, this “basal-squamous subtype” can be reliably detected by molecular sequencing methods and detection of the molecular signature corresponds well to the predicted response to therapy. This method of comprehensive genomic profiling prior to definitive therapy may provide additional information to the clinical team in terms of anticipated therapeutic response of tumors with SqDif²⁵.

UC with Glandular Differentiation

Glandular differentiation (GDif) is the second most common type of divergent differentiation and present in up to 20% of UCs ^{6, 21, 22}). Perhaps the most important consideration when glandular differentiation is detected is the overlap with adjacent carcinomas, namely urachal adenocarcinoma, colorectal carcinoma, and prostatic adenocarcinoma. IHC profiling is only variably helpful as UCs with GDif can express markers indistinguishable from some of these other tumor types. If pure glandular differentiation is diagnosed in the bladder, clinical and radiologic exclusion of bladder dome-based tumors (urachal adenocarcinoma) and prior/concurrent colorectal/prostatic adenocarcinoma should be considered. Confirming with the pathologist that the tumor expresses at least some classic urothelial morphology or more ideally urothelial precursor lesions – papillary UC or UC in situ - is very helpful in establishing urothelial lineage. In cases of particularly histologically ambiguous features genomic studies have been shown helpful in distinguishing GDif from true adenocarcinomas ²⁶. In a similar fashion to its SqDif counterpart, UC with GDif does not necessarily suggest more aggressive disease when staged matched with classical UC ^{21, 22}.

UC with Trophoblastic Differentiation

Trophoblastic differentiation (TDif) is much less common (<5% of UC’s) and is an attempt by the tumor to recapitulate the cytotrophoblast and synciotrophoblast cells seen in choriocarcinoma. UC with TDif may be mistaken for choriocarcinoma and vice versa, however typically the age and demographic information can be very helpful in this distinction with UC with TDif (disease of elderly) and choriocarcinoma (disease of young adults). Both choriocarcinoma and UC with TDif can produce significant amounts of beta-HCG and this is not helpful in distinguishing them. However, once a UC with TDif has been definitively diagnosed, monitoring with serum beta-HCG could be considered as an additional (less invasive) disease monitoring technique^27–29.

2.3: Upper Tract Urothelial Carcinoma (UTUC)

UC can arise at any point along the urinary tract – from urethra to the collecting system of the kidney. UTUC’s, those that arise above the bladder in the ureter and renal collecting system have unique features and challenges for pathologist and clinician alike. Monitoring and sampling the upper tract is technically challenging given the deep location, and biopsies from this area are typically significantly less substantial and low volume. Already faced with the complication of small samples, pathologists must also contend with the relatively more muted or subtle atypia that develops in the upper tract urothelium. The morphologic threshold for high grade urothelial carcinoma is relatively lower in this region and yet the stakes (typically nephroureterectomy for UC-CIS or worse) are quite high. Additional frequent complications in this area include impacted calculi and reactions to injury such as nephrogenic adenoma all of which may impart an atypical appearance.

UTUC while very rare, is the third most common Lynch Syndrome (LS) associated carcinoma, and the most recent American Urologic Association Guidelines (2023) strongly recommend universal mismatch repair (MMR) screening (either by MMR IHC or microsatellite instability testing) on all suspected UTUCs with referral to genetics or confirmatory germline testing as appropriate³⁰. UTUC is nearly 9x more likely to demonstrate MMR altered IHC than its classical UC counterpart³¹. Nearly 5% of all tested UTUCs are ultimately found to be LS associated as compared to 2.0–3.5% of colorectal and endometrial carcinomas ³². Mismatch repair testing (MMR IHC) for the genes PMS2, MLH1, MSH2, and MSH6 is readily available to most pathology labs and should be performed universally in the setting of UTUC. The detection of MMR protein loss has been associated with favorable response to PDL1 immunotherapies and may have monitoring implications for the patient’s immediate family members³³.

2.4: Small Cell Carcinoma (“Neuroendocrine Differentiation”)

Small cell carcinoma (SmCC) is rare in the urothelial tract, in the consensus molecular classification of muscle-invasive UC only 3% of UC contained genomic signatures in the small cell / neuroendocrine-like category³⁴. SmCC may occasionally exist in its pure form but usually contains other non-small cell (urothelial) components from which it derived³⁵. SmCC has the worst prognosis and overall survival of the molecular defined subtypes of UC³⁴. There is also an important distinction between true SmCC (with or without a background of conventional UC component) and UC with focal immunohistochemical neuroendocrine differentiation. Many tumors (including UC) may express focal neuroendocrine markers by immunohistochemistry while lacking the classical morphologic features of SmCC. Importantly, neuroendocrine IHC expression alone does not correlate with response to SmCC platinum therapy or disease aggressiveness. Rather, it is the initial morphologic features of SmCC that seem to best correlate with disease biology; 72% of cases molecularly defined in the SmCC/NE-like category were identified initially by morphologic SmCC findings³⁴. Multiple studies with substantial clinical follow-up have shown that neoadjuvant chemotherapy for localized (M0) SmCC has significant pathologic downstaging and profound impact on overall survival, neoadjuvant chemotherapy should be strongly considered in the context of localized (M0) urothelial carcinoma with SmCC differentiation^{36, 37}. M1 disease reliably predicts poor survival even with neoadjuvant therapy³⁶.

3. Treatment Considerations Based on Molecular Genetic Testing

3.1: The landscape of UC genetics and targets

The Cancer Genome Atlas (TCGA) has characterized UC as a high mutation rate disease with several significantly mutated genes and copy number alterations involving the cell cycle, RTK/RAS/PI3K, chromatin remodeling, and DNA repair pathways (Table 2).³⁸ While several trials for UC using cell cycle and receptor tyrosine kinase inhibitors, PI3Kinase and mTOR inhibitors, PARP inhibitors and epigenetic therapies have had limited success ^39–44, some have resulted in FDA approval of targeted therapies such as immune checkpoint inhibitors (ICI) , erdafitinib, enfortumab-vedotin (EV) and sacituzumab-govitecan (SG). ^45–50

Table 2:

Select significantly mutated genes, copy number changes, fusions and canonical pathways affected in UC Adapted from Robertson et al (2017 TCGA study).

Gene mutations (frequency %)	Copy number amplifications (amp), deletions (del), fusions (fus) (frequency %)	Canonical pathways (frequency %)
TP53 (48) KMT2D (28) KDM6A (26) ARID1A (25) PIK3CA (22) KMT2C (18) RB1 (17) EP300 (15) STAG2 (14) FGFR3 (14) ATM (14) CREBBP (12) FAT1 (12) ELF3 (12) ERBB2 (12) SPTAN (12) KMT2A (11) ERBB3 (10)	E2F3 amp (12) PPARG amp (6) MDM2 amp (6) CDKN2A del (22) RB1 del (4) FGFR3 fus (2) PPARG fus (3)	p53/cell cycle pathway (89) RTK/RAS/PI3K pathway (71) Histone modifying (52) SWI/SNF (26) DNA damage (16) Cohesin complex (14) Oxidative stress (10) Alternative splicing (7)

UC presents most frequently as non-muscle invasive bladder cancer (NMIBC). FGFR3 gene alterations are present in 70% of low-grade NMIBC usually Ta (papillary lesions). High-grade NMIBC is associated with TP53 mutations and is genomically more unstable with a more significant potential to progress to escalating grade levels, invasion, and metastatic disease with the cumulative acquisition of oncogenic driver genomic alterations. ⁵¹

The molecular studies of patients with metastatic UC that have progressed on platinum-based chemotherapy reveal TP53, ERBB2, PI3KCA, FGFR3, and ARID1A alterations in 66%, 29%, 27%, 24%, and 22% of all patients, respectively. TP53 mutated tumors are associated with the worst prognosis. ⁵²

A transcriptome-based international consensus classification for UC into 6 molecular subtypes, each with specific characteristics, was developed by the Bladder Cancer Taxonomy Group as a guide to evaluating disease biology (Table 3).³⁴ A single-sample classifier is available as a research tool for correlative tissue studies with defined therapies and clinical settings.

Table 3:

Characteristics of consensus subclasses of UC³⁴.

Molecular subclass	luminal papillary	luminal nonspecified	luminal unstable	stroma-rich	basal/squamous	neuroendocrine-like
Positively enriched bladder cancer signatures	Urothelial differentiation Ta pathway FGFR3 coexpressed genes	Urothelial differentiation	Urothelial differentiation Cell cycle activity	Smooth muscle	Basal Differentiation Myofibroblasts Interferson response Keratinization Late cell cycle	Neuroendocrine differentiation Cell cycle activity Late cell cycle
Tumor microenvironment enrichment		Fibroblasts		Stromal infiltration Fibroblasts	Immune infiltration Cytotoxic lymphocytes Monocytic lineage
Associated regulon activity	RARG, FGFR3, RXRA, FOXA1, ERBB3, ERBB2, PPARG, GATA3	PPARG+	ERBB3, ERBB2, PPARG, ESR2	RARA, RARB, PGR, FGFE1	KLF4, STAT3, FOXM1, EGFR, HIF1A, GATA6 TP63
Enriched gene mutations and copy number alterations	KDM6A mutations, FGFR3 mutation, fusion/ amplification, CDKN2A deletions	ELF3 mutations	TP3 and ERCC2 mutations. PPARG amplifications and fusions, E3F3/SOX4 amplifications		TP53, RB1 mutations	TP53, RB1 mutations and deletions
Significant associations with histologic variants, morphology, CIS and smooth muscle infiltration	Papillary	Micropapillary, CIS			Squamous, sarcomatoid, smooth muscle infiltration	Neuroendocrine
Clinical and prognostic associations	Under 60 years, T2, best survival	Over age 80 years, T2, Male			Women, T3/T4	Worst survival
Predicted Therapies and targets	FGFR3 -targeted therapies	Pan fibroblast, TGF-beta	Radiotherapy, Cell cycle, DNA replication	Pan fibroblast TGF-beta, hypoxia	Immune-checkpoint inhibition EGFR, Hypoxia	Radiotherapy, Cell Cycle, DNA replication

3.2: Predictors of response to treatment and specific targeted therapies

BCG:

In a retrospective study of 62 high-grade T1 UC samples from patients treated with BCG, DNA damage response and repair (DDR) gene mutations (ERCC2, BRCA2) were associated with higher TMB and no recurrence. TP53, ATM, ARID1A, AHR, and SMARCB1 mutations, and focal copy-number gain in CCNE1 and CDKN2A deletion were enriched in patients with progressive disease and recurrence. Pathologic stage pT1b was associated with focal amplifications of E2F3/SOX4, PVRL4, CCNE1, and TP53 alterations. ⁵³

Cisplatin:

In a retrospective study of a 100 patients with locally advanced and metastatic urothelial carcinoma, patients with somatic alterations in DDR genes were noted to have more favorable risk disease and better progression-free and overall survival⁵⁴. In a mixed cohort of 58 patients from two phase II trials using cisplatin-based neoadjuvant chemotherapy, ATM, RB1 and FANCC gene mutations were associated with a pathologic complete response, and longer overall and disease specific survival⁵⁵. In a comprehensive multi-omics analysis of 300 muscle-invasive bladder cancer patients treated with cisplatin-based chemotherapy, genomic instability driven by chromosomal alterations, indels, signature 5 mutations, and BRCA2 mutations, and immune cell infiltration and PD-1 protein expression correlated with response while basal/squamous gene expression subtype was associated with poor response.⁵⁶

Immune checkpoint inhibitors (ICIs): Several immunohistochemistry (IHC) assays for PD-1/PD-L1 are available to correlate with response to specific ICIs but have limited value as these do not constitute absolute or definite biomarkers for ICI response. Tumor mutational burden (TMB), microsatellite instability, and other determinants play a significant role in response to ICIs. For example, in an analysis of tumor characteristics of patients on the IMvigor 210 trials involving patients with platinum-refractory or cisplatin-ineligible UC who were treated with the PD-L1 inhibitor atezolizumab, tumors of the neuroendocrine subtype had a 100% response rate. The neuroendocrine subtype represents a high-risk cohort with advanced disease, and it is hypothesized that the response seen is due to low levels of TGFβ expression and high TMB.⁵⁷ In another study of 60 patients treated with ICIs, tumor DDR gene mutations correlated with a higher response rate, progression-free and overall survival⁵⁸.

Erdafitinib:

FGFR alterations are seen in approximately 20% of metastatic UC. FGFR inhibition with erdaftinib is highly effective in UC with select FGFR2/3 alterations with an objective response rate (ORR) of 40% and is now currently approved for use in FGFR2/3 altered locally advanced and metastatic UC. ⁴⁸ In a randomized, double-blind, placebo-controlled, phase III trial of FGFR inhibitor infigratinib as adjuvant therapy in patients with invasive UC harboring susceptible FGFR3 alterations, FGFR3 alterations were seen in only 19% of all patients screened. The prevalence was higher in UTUC (30%) than bladder cancer (13%). The trial was stopped early and results are awaited⁵⁹.

Antibody Drug Conjugates (ADCs):

EV is an ADC with a high affinity for Nectin-4, an adhesion protein involved in oncogenesis and highly and frequently expressed in UC.⁶⁰ In previously treated locally advanced or metastatic UC, EV has an ORR of 40.6%⁴⁹. and in combination with pembrolizumab in the treatment naïve setting, an ORR of 73.3. ⁶¹ SG, an ADC that targets the oncogenic protein Trop-2, has an ORR of 27% in previously treated locally advanced or metastatic UC. ⁵⁰ Disitamab vedotin (RC48), an ADC drug targeting HER2, is currently under study. In a single-arm study of patients with HER2⁺ (IHC status 3+ or 2+), previously treated, locally advanced or metastatic UC, an impressive ORR of 51.2% is noted. ⁶²

3.3: The role of molecular testing in various stages of UC:

Despite correlation of histopathological features and treatment response with various molecular characteristics in both early and late stages of UC, the role of standard of care clinical molecular testing remains confined to advanced or metastatic UC and currently informs treatment with FGFR inhibitor, erdafitinib, in second or later-line of treatment in the advanced or metastatic disease stage. Clinical molecular testing plays a significant role in matching patients with advanced or metastatic UC to clinical trials. There is emerging evidence of the role of plasma circulating tumor DNA in predicting benefit, and outcomes in patients treated with perioperative immunotherapy, and chemotherapy. The role of these assays in clinical decision making remains to be defined in prospective studies^63–65. For patients receiving perioperative systemic therapy, clinical molecular testing may guide regimen choice in cases of uncertain histology such adeno carcinoma of the bladder versus UC with glandular differentiation. Adenocarcinoma of the bladder closely resembles colorectal adenocarcinoma molecularly whilst UC with glandular differentiation more closely resembles UC molecularly^{26, 66}. We may see targeted therapy with FGFR inhibitors emerge as standard of care in localized UC in the future, in which scenario, clinical molecular testing will move to earlier stages of bladder cancer. Until new biomarker-based indications of treatment emerge, clinical molecular testing of UC is mostly useful in making treatment decisions only in the second- or later- line advanced or metastatic UC.

3.4: Future Directions

The multi-omic characterization of UC has ushered in an era of predictive and prognostic molecular biomarkers and molecularly targeted therapies in all stages of the disease. A composite of histopathology and molecular characterization with DNA, RNA, and proteomic studies of UC will likely guide treatment toward precise personalized therapy in every disease stage in the future.