2. Neoadjuvant Treatment
2.1. Neoadjuvant Chemotherapy (NAC)
Currently cisplatin-based chemotherapy before surgery is the standard of care in the management of MIBC. NAC should be recommended for all patients with T2 to T4 or N1 MIBC who are eligible for cisplatin
[3][9]. Two randomized clinical trials and meta-analysis have shown an OS benefit with this strategy compared with up-front RC. In addition, alternative chemotherapy regimens have been evaluated in the neoadjuvant setting.
Two phase 3 randomized clinical trials have shown a longer OS with cisplatin-based chemotherapy in the neoadjuvant setting. The SWOG-8710 trial included 317 patients with T2-T4aN0 MIBC. They were randomized to three cycles of neoadjuvant methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) followed by surgery compared to RC alone
[10]. A benefit in OS was demonstrated, with a median OS (mOS) of 77 months in the neoadjuvant group vs. 46 months in the surgery alone group (
p < 0.06). In addition, an increase in the pathologic complete response, defined as pT0N0, was observed with neoadjuvant chemotherapy (38% vs. 15%;
p < 0.001).
A modified MVAC regimen (dose-dense MVAC (dd-MVAC)) has also been tested in two phase 2 clinical trials. Choueiri et al. evaluated four cycles of dd-MVAC as neoadjuvant therapy in 39 patients with MIBC in a single-arm phase 2 trial
[11]. The primary endpoint was pathologic response (PaR) defined by pathologic downstaging to ≤pT1N0. The PaR was 49% (80% CI, 38 to 61), with a 26% of pT0N0. Grade 3 or higher toxicity was observed in 10% of patients, with no neutropenic fevers or treatment-related death. Plimack et al. administered three cycles of neoadjuvant accelerated MVAC in a phase 2 trial with 44 patients
[12]. This entry reported 38% of pCR and 52% of downstaging to non-muscle invasive disease. In addition, most of the patients (82%) experienced only grade 1–2 treatment-related toxicities. There were no treatment-related deaths. Currently, dd-MVAC has replaced classic MVAC because of its better tolerance, shorter duration and higher pCR rate.
Cisplatin plus gemcitabine (CG) is the standard first-line treatment in metastatic UC due to a phase 3 trial that showed a similar progression-free survival (PFS) and OS compared to MVAC with a better safety profile
[13]. Despite the lack of randomized trials with preoperative CG, these data have been extrapolated to the neoadjuvant setting, and CG is commonly used prior to surgery in MIBC patients. Some retrospective studies have described similar outcomes with this regimen as neoadjuvant therapy
[14][15][16].
Various meta-analyses have demonstrated a benefit in OS with cisplatin-based chemotherapy vs. up-front surgery in patients with MIBC. GETUG-AFU V05 VESPER trial is a phase III randomized study that compares dd-MVAC vs. CG in patients with MIBC as perioperative therapy
[17][18]. GETUG-AFU V05 VESPER trial is a phase III randomized study that compares dd-MVAC vs. CG in patients with MIBC as perioperative therapy
[19]. A total of 88% of patients received chemotherapy in the neoadjuvant setting. The primary endpoint was PFS at 3 years. In the neoadjuvant group, PFS at 3 years was significantly longer in the dd-MVAC arm (66% vs. 56%, HR 5 0.70 [95% CI, 0.51 to 0.96],
p = 0.025). In addition, dd-MVAC reached a higher local control rate (pCR, tumor downstaging or organ confined) 63% vs. 50% (
p = 0.021).
2.2. Single Inmmunotherapy Agents
Several trials to date have addressed the benefit of immunotherapy agents in the perioperative treatment of MIBC. The PURE-01 trial was an open-label, single-arm, phase-2 trial that included 50 patients with a predominant (at least 50%) UC histology and clinical T3b or less of an N0 stage tumor. Patients were included regardless of their cisplatin eligibility, and 92% of included patients were cisplatin-eligible. Patients were administered three cycles of pembrolizumab before surgery. The primary endpoint of the trial was a pCR rate of 25%. This endpoint was achieved as pCR in the study was 42%. A total of 52% of patients achieved the down-staging of the primary tumor
[20]. Updated survival analyses of this trial evidenced a 1 and 2 year event-free survival (EFS) of 84.5% and 71.7%, respectively
[21]. Only 6% of patients had grade 3 immune-related adverse events (irAEs), with only one patient that required treatment discontinuation. The ABACUS trial included 95 patients with MIBC (T2-T4aN0M0) who were ineligible or refused neoadjuvant cisplatin treatment. These patients received two cycles of Atezolizumab before RC. Of these 95 patients, 85 underwent surgery. The pCR rates were 31% in all patients and 37% of PD-L1-positive patients. In an updated report of the trial results, 2-year disease-free survival (DFS) was 68%. In patients who achieved pCR, the 2-year DFS improved to 85%. Higher T stage at baseline and at cystectomy and node-positive correlated with poor DFS. The 2-year OS rate was 77%
[22]. The results of the cisplatin-ineligible arm of the AURA trial (Oncodistinct-004) have recently been presented in ASCO 2022 meeting. This arm included patients with cT2-T4aN0-2M0 UC that were randomized to avelumab monotherapy or paclitaxel (P) + gemcitabine (G) + avelumab. The pCR was achieved in 16% of patients in the P + G+ avelumab arm, compared to 36% in the avelumab monotherapy arm. Downstaging showed a similar trend between arms. Ongoing studies with single immunotherapy approaches will help to clarify the role of neoadjuvant single immunotherapy agents in this population.
Regarding other histologies, after an amendment in March 2018, the PURE-01 trial allowed the inclusion of predominant variant histologies (VH). An update of the results of the trial, including these patients (
n = 19), has recently been published. A substantially lower pCR rate was found (16%) in this subgroup. Responses were achieved in squamous cell carcinoma and lymphoepithelioma-like variant patients, while no pathological responses were noted in the other variants. However, researchers suggest that responses were related to tumor biomarkers as responding VH patients were enriched in higher TMB and CPS patients
[23]. In the ABACUS trial, only patients with predominant urothelial carcinoma were included, so data in this subgroup of patients is scarce. The ABACUS-2 trial is currently ongoing and recruiting patients in two different arms: One for urothelial carcinoma of the upper tract and the other for rarer histological subtypes (NCT04624399). In addition, the EV-303/KEYNOTE-905 trial will compare pembrolizumab monotherapy vs. pembrolizumab + enfortumab vedotin (EV) vs. up-front surgery (NCT03924895).
2.3. Combination of Inmmunotherapy Agents
Anti-PD(L)1 agents have been usually combined with anti-CTLA4 drugs to potentiate each other. In UC, this combination in the neoadjuvant scenario have been investigated in several trials. In the NABUCCO trial, the combination of nivolumab and ipilimumab was addressed. A total of 24 patients with stage III UC were treated in this trial with a defined sequence of ipilimumab (3 mg/kg in days 1 and 22) and nivolumab (3 mg/kg in days 22 and 43) followed by RC in the cohort 1. As a feasibility trial, the main endpoint was the possibility to resect within 12 weeks of starting the treatment, showing a 96% resection rate. A total of 46% of the patients achieved a pCR and 58% showed no invasive residual tumor. Grade 3–4 immune-related adverse events (irAEs) were present in 55% of patients
[24]. The aim of NABUCCO cohort 2 was to find an optimal dose of preoperative nivolumab and ipilimumab. Patients in cohort 2a received two cycles of nivolumab at 1 mg/kg with ipilimumab 3 mg/kg followed by a third cycle of nivolumab at 3 mg/kg. In the cohort 2b, the treatment was nivolumab at 3 mg/kg in combination with ipilimumab 1 mg/kg and an extra cycle of nivolumab at 3 mg/kg. The pCR in cohort 2a was higher in comparison with cohort 2b (43% vs. 7%)
[25]. In addition, a higher G3-4 irAEs were observed in cohort 2a patients (33% vs. 20%). Another trial also investigated different nivolumab and ipilimumab schedules in cisplatin-ineligible patients.
The DUTRENEO trial included a population of 61 cisplatin-eligible patients with cT2a-T4aN0-1 tumors. The patients included in this trial had a tumor pro-inflammatory interferon-𝛾 (INF-𝛾) immune signature assessed and divided into “hot” and “cold” tumors. Patients with “hot” tumors were randomized to standard cisplatin-based chemotherapy or to three cycles of immunotherapy, while “cold” patients received chemotherapy. This approach failed to select the patients more responsive to immunotherapy. Patients with “hot” tumors showed a similar pCR with immunotherapy or chemotherapy (34.8% vs. 36.4%, respectively). In addition, pCR in patients with “cold” tumors treated with chemotherapy was 68.8%
[26]. A combination study with durvalumab and tremelimumab in cisplatin-ineligible high-risk patients (high-risk features were defined by T3-T4 tumors, hydronephrosis, pathologic variant urothelial carcinoma, vascular and lymphatic invasion, and/or high-grade upper tract disease) showed an overall pCR of 37.5% and downstaging of 58%. In VH, pCR was 57%, although there were only seven patients with VH included
[27].
2.4. Combination of Inmmunotherapy and Chemotherapy
As the standard of care in the perioperative treatment of MIBC is nowadays cisplatin-based chemotherapy, several trials have addressed if the addition of immunotherapy could improve the results of chemotherapy alone. The HCRN GU 14-188 was a phase 1b/2 trial that included patients with T2-T4aN0M0 BC. The study included two cohorts of patients, one for patients eligible for treatment with cisplatin (43 patients) and another for ineligible patients (37 patients). In the cisplatin-eligible cohort, patients received four cycles of cisplatin + gemcitabine overlapped with five cycles of pembrolizumab. In this cohort, the pCR rate was 44.4%
[28]. In the cisplatin-ineligible cohort, patients received three cycles of weekly gemcitabine and overlapped with five cycles of pembrolizumab. The pCR in this population was 45.2%
[29]. The LCCC1520 trial included 39 cisplatin-eligible patients with a pCR of 36%.
Another study with the introduction of Atezolizumab before four cycles of full dose CG and a consolidation dose of Atezolizumab showed an encouraging 41% of pCR. After a median follow up of 16.5 months, no relapse in patients who achieved a <ypT2N0 was seen. It is important to note that all patients underwent RC
[30]. The BLASST-1 was a phase 2 single-arm trial that investigated neoadjuvant CG for four cycles with nivolumab. The cT2a-T4 tumors and node positive patients were allowed. The primary objective of the study was ≤ypT1N0 pathological response, observed in 65.8% of patients. The pCR rate was 34%
[31]. At a median follow-up of 15.8 months, 12-month RFS rate was 85.4% and PFS was 83%
[32]. Durvalumab has also been tested in combination with cisplatin and gemcitabine in the SAKK 06/17 trial that included node-positive patients, with a pCR of 34% and an EFS at 2 years of 76.1% that was the primary endpoint of the trial
[33].
The other regimen of neoadjuvant chemotherapy, which is being tested in combination with immunotherapy, is MVAC. Cohort 1 of the AURA trial assessed the combination of avelumab with CG and dd-MVAC
[34]. The pCR was 32% and 43% and the downstaging to ≤ypT1N0 was 57% and 64% with CG and dd-MVAC, respectively, in combination with avelumab. In addition, dd-MVAC will be tested in combination with durvalumab in a phase 2 trial that is currently recruiting (NEMIO trial)
[35]. Other immunotherapy agents that are being investigated in combination with MVAC are nivolumab and pembrolizumab (RETAIN-2; NCT04506554 and NCT04383743, respectively).
Other chemotherapies are being evaluated in combination with ICI. The SWOG GAP trial is a phase 2 trial that will include 196 cisplatin-ineligible patients and will compare the combination of carboplatin with gemcitabine and avelumab to observation before surgery (NCT04871529). The combination of tislelizumab, an anti-PD-1 antibody, with Nab-paclitaxel will be addressed in a phase 2 trial of MIBC not extended to regional lymph nodes. In this trial, the primary objective will be the CR rate, but patients will be able to undergo transurethral resection of bladder tumor (TURBT) instead of RC. The results of the cohort in which TURBT was performed have been published. A total of 22 patients were included in this cohort; a 77% pT0, 4.5% pTa, 13.63% pT1 and 4.5% pTis was observed. The recurrence-free survival rate of these patients after one year was 82%
[36]. However, patients treated with TURBT continued to receive medication after the procedure, and this may confound the trial outcome and a longer follow-up is needed.
Definitive results from ongoing phase 3 studies will establish the role of chemo-immunotherapy in the perioperative setting. The KEYNOTE 866 is a phase 3 randomized trial of neoadjuvant chemotherapy with perioperative pembrolizumab or placebo in cisplatin-eligible patients. Patients will receive four cycles of cisplatin + gemcitabine in combination with pembrolizumab or placebo, followed by RC plus pelvic lymph node dissection (PLND), and postoperative 13 cycles of adjuvant pembrolizumab or placebo
[37]. The NIAGARA trial is an ongoing phase 3 trial with a similar design to KEYNOTE 866, which will address the efficacy of durvalumab in combination with CG in the neoadjuvant setting and durvalumab in the adjuvant
[38]. The ENERGIZE trial is another phase 3 trial that will compare the perioperative combination of chemotherapy with nivolumab with or without linrodostat mesylate, an IDO inhibitor
[39].
2.5. Combination of Inmmunotherapy and Other Systemic Therapies
2.5.1. Combination of Immunotherapy with Antibody–Drug Conjugates
Antibody–drug conjugates (ADCs) are composed by a monoclonal antibody that targets a tumor-associated antigen linked to a cytotoxic payload. Enfortumab Vedotin (EV) is an ADC that targets the membrane Nectin-4, a member of a family of related immunoglobulin-like adhesion molecules, and this receptor is overexpressed in approximately 60% of samples of BC. The antibody is conjugated to monomethyl auristatin E, a microtubule-disrupting agent
[40]. As this compound has shown promising results in the metastatic setting
[7], EV is currently being investigated in the perioperative setting in combination with pembrolizumab in cis-eligible (EV-304/KEYNOTE-B15 trial, NCT04700124) and ineligible (EV-303/KEYNOTE-905 trial, NCT03924895) patients and in combination with durvalumab and tremelimumab in cis-ineligible patients (VOLGA trial, NCT04960709). Sacituzumab govitecan (SG) is another ADC target to tumor-associated calcium signal transducer 2 (TROP-2)-expressing cancer cells, which is conjugated with SN-38, an anti-topoisomerase 1, as the cytotoxic compound. This ADC has shown positive results in advanced UC
[8] and randomized phase III trials are currently recruiting in these patients (TROPICS-04, NCT04527991). A phase 2 trial for cis-ineligible patients in the neoadjuvant setting combining SG with pembrolizumab is currently recruiting
[41].
2.5.2. Other Targeted Agents
Fibroblast growth factor receptor (FGFR) is a family of tyrosine kinase receptors constituted by four members: FGFR1–FGFR4. Multiple ligands of the FGF family can bind this receptor, activating downstream transduction intracellular signaling pathways. Enrichment in FGFR3 expression has been shown in luminal subtype tumors and this subtype may represent up to 20% of pT2 tumors
[42]. The FGFR inhibition in advanced UC have shown promising results
[6], and it supports the development of clinical trials in the neoadjuvant (NCT04228042) and adjuvant settings (PROOF-302 trial), focused in patients not eligible for cisplatin
[43]. Furthermore, data from early phase clinical trials suggest that ICIs may increase responsiveness to FGFR inhibitors, and this may represent a rationale to develop trials that combine FGFR inhibitors with ICIs in the neoadjuvant setting of UC in the future
[44].
Poly ADP-ribose polymerase (PARP) has a key role in DNA repair and can act as a transcription modulator of genes involved in chromatin remodeling and gene transcription. The decreased expression of genes associated with homologous recombination repair (HRR) pathway results in sensitivity to treatment with PARP inhibitors, causing synthetic lethality
[45]. NEODURVARIB was a phase 2 trial that assessed the value of the addition of Olaparib to durvalumab for two cycles in cT2–T4aN0 cisplatin-ineligible patients. The primary endpoint of the study was pCR that was 44.5%. Combination was well tolerated, with adverse events grade 3–4 detected in only 8.3% of cases
[46].
An overexpression of hypoxia inducible factor (HIF-1) and vascular endothelial growth factor (VEGF) has been associated with poor prognosis and metastatic spread in urothelial carcinoma. Vessel density is associated with vascular invasion, recurrence and shorter survival in invasive UC
[47]. Cabozantinib, a VEGFR2 and MET inhibitor, has demonstrated immunomodulatory results, a decrease in the number of myeloid-derived suppressor cells and regulatory T cells and an increase in PD-1 expression in regulatory T-cells. This may lead to a less immunosuppressive stroma and may be able to potentiate anti-PD1 immunotherapy effects
[48]. To explore this hypothesis, the phase 2 ABATE trial will include cT2–T4aN0/xM0 cisplatin-ineligible patients. The primary objective of the study is to detect more than 20% of <pT2 response
[49]. Methionine aminopeptidase 2 inhibition has been shown to inhibit the proliferation of human microvascular endothelial cells and tumor cells
[50], and it has been hypothesized that this effect could be synergistic with PD-1 inhibition. The ANTICIPATE trial is a phase I/II trial that will test the combination of APL-1202 with Tislelizumab as a neoadjuvant therapy in MIBC
[51].
CD38 is a membrane receptor expressed in tumor-associated macrophages and inhibits lymphocyte T CD8 function via adenosine receptor signaling. CD38 overexpression has been associated with resistance to ICIs
[52], and CD38 blockade has been shown to suppress bladder cancer growth in vivo
[53]. Daratumumab, an anti CD38 antibody, is being investigated in the neoadjuvant setting in MIBC (NCT03473730).
Other studies are aiming to find an adequate partner for immunotherapy as the PECULIAR trial or NCT03978624. Both will combine pembrolizumab with epigenetic modifiers, such as Epacadostat or Entinostat. The other approach is the combination of ICIs with a replication-competent oncolytic adenovirus in cisplatin-ineligible patients with MIBC (NCT04610671).
3. Adjuvant Treatment
3.1. Chemotherapy
Several studies have investigated the role of adjuvant chemotherapy (AC) in MIBC
[54].
The EORTC 3099 trial evaluated AC (four cycles of CG, high-dose MVAC, or conventional MVAC) versus deferred six cycles cisplatin-based CT after surgery. A total of 284 patients with pT3-T4 or N+ M0 UC were randomized
[55]. Although the primary endpoint of OS was not reached, AC showed significantly better DFS compared with deferred CT. However, this trial is controversial due to the design and the primary endpoint selected.
A total of 142 patients were included in a Spanish clinical trial that compared AC (paclitaxel, gemcitabine and cisplatin) with observation in resectable high-risk MIBC
[56]. AC arm had a longer OS and DFS compared with observation arm, but this trial was prematurely finished because of the poor recruitment. An Italian clinical trial assessed the use of AC (CG) versus observation in patients with MIBC
[57]. This trial also had poor recruitment, only 200 of the planned 600 patients were included and the difference in OS and DFS between the AC arm and the observation arm was not found.
Several systematic reviews and meta-analyses have been published, analyzing the data of adjuvant clinical trials in MIBC
[54]. A meta-analysis of 10 clinical trials (1183 patients) showed a benefit of cisplatin-based adjuvant chemotherapy on OS (HR = 0.82, 95 % CI = 0.70–0.96,
p = 0.02)
[58]. An absolute improvement in survival from 50% to 56% was also observed at 5 years. In addition, cisplatin-based chemotherapy improved DFS (HR = 0.71, 95% CI = 0.60–0.83,
p < 0.001), locoregional recurrence-free survival (HR = 0.68, 95% CI = 0. 55–0.85,
p < 0.001) and metastasis-free survival (HR = 0.79, 95% CI = 0.65–0.95,
p = 0.01).
Theoretically AC offers some advantages over NAC: early definitive treatment, such as RC; more accurate pathological staging and prognostic factor; and avoidance of overtreatment of some patients. However, AC has some limitations: patients treated with radical surgery are more probable to be ineligible for cisplatin-based AC because of impaired renal function or performance status
[59].
3.2. Immunotherapy
Three randomized phase 3 clinical trials have evaluated the use of adjuvant immunotherapy (IT) in MIBC: IMvigor010, CheckMate 274 and AMBASSADOR
[60][61].
The Imvigor010 trial compared atezolizumab with observation as adjuvant treatment in patients with high-risk MIBC in an open-label study
[62]. A total of 807 patients were included with ypT2-4a or ypN+ tumors following NAC or pT3-4a or pN+ tumors if NAC was not administrated. Patients who had not been treated with NAC should decline or were ineligible for cisplatin-based adjuvant therapy. A total of 1200 mg of Atezolizumab was administered every three weeks for up to one year. DFS was the primary endpoint of this trial. No significant differences were found in DFS, 19.4 months in the atezolizumab arm vs. 16.6 months in the observation arm (HR: 0.89, 95% CI 0.74–1.08;
p = 0.24).
Nivolumab as adjuvant treatment in patients with high-risk MIBC was evaluated in CheckMate 274 trial, a randomized, double-blind phase 3 clinical trial
[63]. A total of 709 patients were included with ypT2-4a or ypN+ after NAC or pT3-4a or pN+ tumors if NAC was not administrated. Patients who had not been treated with NAC should decline or were ineligible for cisplatin-based AC. Inclusion criteria were similar in Imvigor010 and CheckMate 274. However, in the CheckMate trial, patients were randomized to receive 240 mg nivolumab every two weeks up to one year or a placebo. In this trial, the co-primary endpoints were DFS in all randomized patients and in patients with tumor PD-L1 expression ≥1%. In the whole population analysis, the DFS was 20.8 months in the nivolumab arm and 10.8 months in the control arm. The percentage of patients alive and disease free at 6 months was 74.9% in nivolumab group vs. 60.9 % in placebo group (HR: 0.70; 98.22 % CI, 0.55–0.90). Among patients with a PD-L1 expression ≥1%, the DFS at 6 months was 74.5% and 55.7% in the nivolumab and placebo groups, respectively (HR, 0.55; 98.72% CI, 0.35 to 0.85;
p < 0.001).