Prognostic impact of eligibility for adjuvant immunotherapy in locally advanced urothelial cancer

Abstract Objective To evaluate the effect of postoperative pathological findings related to the eligibility of adjuvant immunotherapy on oncologic outcomes in patients with localized and locally advanced muscle‐invasive bladder carcinoma (MIBC) and upper tract urothelial carcinoma (UTUC). Patients and methods We retrospectively evaluated 1082 patients treated with radical cystectomy (n = 597) and nephroureterectomy (n = 485) between January 2000 and April 2021. Patients were divided into two groups: pT3‐4 or pN+ without neoadjuvant chemotherapy and ypT2‐4 or pN+ treated with neoadjuvant chemotherapy (trial‐eligible group) or others (trial‐ineligible group). The primary outcome was the effect of trial eligibility for adjuvant immunotherapy on disease‐free survival (DFS) and overall survival (OS). Secondary outcomes included the additional effect of lymphovascular invasion (LVI) status to the clinical trial criteria on prognosis and a risk model development. Results The median ages of the patients were 69 and 72 years in the MIBC and UTUC groups, respectively. Fifty‐two percent of patients met the trial inclusion criteria. Trial eligibility was significantly associated with poor DFS and OS among patients with MIBC and UTUC. LVI‐positive status was significantly associated with poor prognosis among patients in the trial‐eligible group. A very high risk (LVI+ or pN+ among the pT3‐4 or ypT2‐4) was significantly associated with poor prognosis. Conclusion A total of 52% of patients were eligible for adjuvant immunotherapy. Trial eligibility was significantly associated with a poor prognosis. LVI+ and pN+ may play a key role in candidate selection for adjuvant immunotherapy.


| INTRODUCTION
Localized or locally advanced urothelial carcinoma (UC) is a lifethreatening disease with a high recurrence and mortality rate (5-year survival rate: 50%-60%). [1][2][3] Although radical cystectomy (RC) or nephroureterectomy (RNU) is the standard-of-care first-line treatment, patient prognosis is limited even when using neoadjuvant chemotherapy (NAC) and extended pelvic lymph node dissection. [4][5][6][7][8][9][10] Adjuvant chemotherapy is an alternative strategy to improve survival, but the administration of toxic chemotherapy in all patients is not feasible because of the advanced age, renal impairment, and frailty in patients with UC. The CheckMate 274 trial demonstrated a benefit in disease-free survival (DFS) with adjuvant nivolumab therapy in patients at high risk of muscle-invasive UC and may become a standard of care in the future. 11 In that study, postoperative pT3-4/ ypT2-4 or pN+ was used as an inclusion criterion for the high-risk group, but the validity of this criterion in clinical practice remains unclear. Also, there is an urgent need for the proportion of patients who are eligible for adjuvant immunotherapy in clinical practice.
Conversely, the primary endpoint was not met in the similar IMvigor 010 study, which evaluated the effect of adjuvant atezolizumab after radical surgery. 12 In that study, the authors found no significant difference in DFS between atezolizumab and observation (median 19.4 vs. 16.6 months, respectively; hazard ratio 0.89; p = 0.2446). 12 Although there is no clear reason for these controversial results, the outcome might have potentially been affected by some key confounding factors. Of the inclusion criteria of those phase III studies, the patient's lymphovascular invasion (LVI) status was not included in the definition of high-risk disease. Because LVI status is one of the established pathological risk factors for poor prognosis in patients with UC, [13][14][15][16][17][18][19] we hypothesize that it might play a key role in the selection of potential candidates for adjuvant immunotherapy.
Firstly, we evaluate the effect of trial eligibility for adjuvant immunotherapy on prognosis in patients with localized and locally advanced muscle-invasive bladder carcinoma (MIBC) and upper tract urothelial carcinoma (UTUC) in a real-world practice. We subsequently evaluate the additional effect of LVI status to the clinical trial criteria on patient prognosis and develop a risk model that includes LVI status.

| Design and ethics statement
We conducted this retrospective, multicenter study in accordance with the Declaration of Helsinki. The study was approved by the ethics committee of the Hirosaki University School of Medicine (2019-099) and all hospitals in this study. Written consent was not obtained in exchange for public disclosure of study information (optout approach).

| Eligibility of cisplatin-based chemotherapy
Because of the population difference, we used the modified cisplatinineligible criteria of Galsky et al. 24 Using the original criteria, a patient defined as cisplatin ineligible would meet at least one of the following criteria: ECOG PS > 1, creatinine clearance <60 ml/min or eGFR <50 ml/min/1.73 m 2 , grade >1 hearing loss, grade >1 neuropathy, and/or New York Heart Association (NYHA) Class III heart failure. In addition, we defined the marginal criteria as being ECOG PS 1, eGFR 50-60 ml/min/1.73 m 2 , NYHA Class II heart failure, and age >80 years. Patients with two or more marginal factors (such as ECOG PS 1 and eGFR 55 ml/min/1.73 m 2 ) were classified as a cisplatin ineligible.

| Surgical procedures
RC or RNU was performed using the previously described basic technique. 25,26 Briefly, patients with MIBC underwent RC, urinary diversion (orthotopic ileal neobladder construction, ileal conduit diversion, and cutaneous ureterostomy) and standard pelvic lymph node dissection. In patients with UTUC, we performed open or laparoscopic RNU, which includes the removal of the kidney, ureter, and ipsilateral bladder cuff. We managed the distal ureter using the extravesical approach. We performed regional lymph node dissection only when an obvious lesion was observed on imaging study findings. 27

| Outcomes
We divided the patients into two groups: pT3-4 or pN+ without NAC and ypT2-4 or pN+ treated with NAC (trial-eligible group) or others (trial-ineligible group). A DFS event was defined as the length of time from primary treatment to recurrence or death. An OS event was defined as the length of time after primary treatment to last follow-up or any cause of death. The primary outcome was the effect of trial eligibility for adjuvant immunotherapy on the DFS and OS. In the case of UTUC, superficial recurrences of bladder tumors were not included in the visceral DFS. Secondary outcomes included the additional effect of LVI status to the clinical trial criteria on prognosis, risk model development, and a comparison of the Harrell's concordance index (cindex) 28 and net benefit 29 between the base model (pT3-4/ypT2-4 or pN+) and the LVI model (pT3-4/ypT2-4 and pN+ or LVI+) in patients with MIBC and UTUC.

| Baseline characteristics
The median ages of the patients with MIBC and UTUC were 69 and 71 years, respectively. Table 1

| Primary outcome
We found a significant difference in DFS and OS between the trial-eligible and -ineligible groups among patients with MIBC  Figure S1) and UTUC ( Figure S2).

| Secondary outcomes
Multivariable Cox regression proportional hazards model showed a significant effect of LVI and pN+ for DFS and OS in both patients with MIBC and patients with UTUC (Table 2). In the trial-eligible group of MIBC patients, 41% and 18% of patients were LVI+ or pN+ and LVI+ and pN+, respectively ( Figure S3A). The duration of DFS was significantly shorter in patients with LVI+ (p < 0.001, Figure S3B) or pN+ (p < 0.001, Figure S3C). In the trial-eligible group of patients with UTUC, 44% and 7.8% of patients were LVI+ or pN+ and LVI+ and PN+, respectively ( Figure S3D). The duration of DSF was significantly shorter in patients with LVI+ (p < 0.001, Figure S3E) or pN+ (p < 0.001, Figure S3F).  (pT3-4 or ypT2-4, and pN0 and LVIÀ), and very high-risk (pT3-4 or ypT2-4, and pN+ or LVI+) ( Table 3). There was a significant difference in DFS (p < 0.001, Figure 3A) and OS (p < 0.001, Figure 3B) between the high-risk and very high-risk groups in the trial-eligible group of MIBC. Similarly, we observed a significant difference in DFS (p < 0.001, Figure 3C) and OS (p < 0.001, Figure 3D) between the high-risk and very high-risk groups in the trial-eligible group of UTUC.  this result to mean that we can reduce unnecessary treatment in 7.5 per 100 patients with a 45% of the risk for tumor relapse.

| DISCUSSION
This study investigated the proportion of trial-eligible patients for adjuvant immunotherapy and its impact on prognosis in localized or locally advanced UC. Because the CheckMate 274 and IMvigor 010 trials included selected patients, 11,12 we need to recognize the selection biases in those patients to translate the outcomes from clinical trial to practice. We observed that more than half (n = 543/1082,  [13][14][15][16][17][18][19] We found that more than half of the patients in this cohort had LVI+ or pN+ (very high-risk) ( Figure S4A,D). In addition, when we simply compared LVI status in the trial-eligible group, the median DFS was significantly worse in patients with LVI+ than in those with LVIÀ, which was similar to that of patients with pN+ ( Figure S4B,C,E,F). Results of the multivariable Cox regression proportional hazards model showed that LVI+ or pN+ were significant factors for poor prognosis. We subsequently developed a risk model that included LVI status (base model plus LVI status: LVI model) and observed a clear difference in prognosis ( Figure 3). Several limitations in this study need to be acknowledged. First, because of the retrospective study design, we could not control for selection bias and other unmeasurable confounders. Second, the statistical analysis might be underpowered because of the small sample size. Third, analyses under a single population are a problem for generalization. Also, this was an observational study presenting the wellknown outcomes, and those were not beyond expectations. Nonetheless, this study presents the clinical implications of the eligibility of adjuvant immunotherapy and its impact on prognosis in localized or locally advanced UC. Further studies are required to determine the optimal strategies for the transition from surgical treatment to adjuvant immunotherapy.

| CONCLUSIONS
A total of 52% of patients were potentially eligible for adjuvant immunotherapy. Trial eligibility was significantly associated with a poor prognosis. LVI+ and pN+ may play a key role in the selection of candidates for adjuvant immunotherapy.
The present retrospective, multicenter study was performed in accordance with the ethical standards of the Declaration of Helsinki and was approved by the ethics review board of Hirosaki University School of Medicine (authorization No. 2019-099) and all hospitals.
Pursuant to the provisions of the ethics committee and the ethics guidelines in Japan, written informed consent is not required for public disclosure of study information in the case of a retrospective and/or observational study using materials, such as existing documents (opt-out approach).

AUTHORS CONTRIBUTIONS
Yuki Miura is responsible for data collection.