PD-1 and PD-L1 inhibitors have revolutionized cancer therapy, which shifted clinical investigations toward immunotherapy in advanced UC. In addition, PD-1 inhibitor appears to exhibit favorable survival outcomes and a comparable safety profile with PD-L1 inhibitor in cancer therapy reported in a systematic review and meta-analysis [9]. Therefore, we investigated the efficacy and safety of domestic PD-1 inhibitor (tislelizumab or sintilimab) alone or in combination with platinum-based chemotherapy as first-line treatment for advanced UC in daily clinical practice.

PD-1/PD-L1 are primarily involved in inhibitory immune signaling, and PD-1 or PD-L1 inhibitors can reactivate T cell-mediated antitumor immunity by blocking the PD-1/PD-L1 immune checkpoint pathway [17]. Regimens that combine PD-1/PD-L1 inhibitors and platinum-based chemotherapy are appealing in clinical practice because platinum-based chemotherapy can induce immune-modulatory effects, thereby enhancing concomitant PD-1 and PD-L1 blockade, and this combination might also be beneficial because of the absence of clinical cross-resistance between these different therapeutic classes [18, 19]. PD-1 or PD-L1 inhibitor monotherapy was shown to be an effective treatment option in daily clinical practice for metastatic UC patients, with a median OS of 7.8 months, a median PFS of 2.8 months, and ORR in 24% and DCR in 40% [20]. Pond et al. [21] also reported the clinical outcomes of 79 platinum-ineligible patients with advanced UC treated with first-line PD-1/PD-L1 inhibitors in the real world, and the ORR was 27.9%, the median OS was 45 weeks and the treatment failure-free survival was 16 weeks. Recently, the KEYNOTE-361 trial did not meet the primary endpoints of superior PFS and OS with first-line pembrolizumab (PD-1 inhibitor) plus chemotherapy versus chemotherapy alone in patients with advanced UC, reporting a median PFS of 8.3 months and median OS of 17.0 months for patients treated with pembrolizumab plus chemotherapy [11]. However, Mori et al. [22] conducted a systematic review and meta-analysis that indicated a superior oncologic benefit from first-line ICI combination therapies in patients with chemotherapy-eligible metastatic UC over standard chemotherapy. Moreover, addition of atezolizumab (PD-L1 inhibitor) to platinum-based chemotherapy as first-line treatment prolonged PFS in patients with metastatic UC in the IMvigor130 trial, with a median OS of 16.0 months and median PFS of 8.2 months [10]. In the JAVELIN Bladder 100 trial of 700 patients with advanced UC, maintenance avelumab plus best supportive care significantly prolonged OS after first-line chemotherapy [23]. Most recently, the CheckMate 901 trial reported that nivolumab plus gemcitabine-cisplatin resulted in significant better outcomes in patients with previously untreated advanced UC than gemcitabine-cisplatin alone, whit a median OS of 21.7 months and median PFS of 7.9 months [24]. In addition, enfortumab vedotin plus pembrolizumab significantly prolonged OS (31.5 months) and PFS (12.5 months) than chemotherapy in patients with untreated locally advanced or metastatic UC [25]. In our study, the median OS was 16.0 months and median PFS was 12.0 months for all of the enrolled patients. PD-1 inhibitor combined with chemotherapy achieved better PFS than PD-1 inhibitor monotherapy (12.0 v 6.0 months, HR: 0.19, 95% CI: 0.05–0.75, p = 0.018) and provided a tendency of prolonged OS (17.0 v 10.0 months, HR: 0.32, 95% CI: 0.09–1.13, p = 0.077). However, additional research with multicenter randomized controlled trials and long-term follow-up is needed to further investigate the efficacy of immunotherapy with or without chemotherapy.

While PD-1 inhibitor in combination with chemotherapy is associated with favorable efficacy outcomes, the safety profile should be considered in daily clinical practice. In fact, there was no statistically significant difference of AEs between ICIs plus chemotherapy and chemotherapy alone [22]. In the KEYNOTE-361 trial, there was no new or unexpected safety signal reported for pembrolizumab plus platinum-based chemotherapy, and this treatment had a similar safety profile to chemotherapy alone [11]. Treatment-related AEs of PD-1 and PD-L1 inhibitors in clinical trials were evaluated in a systemic review and meta-analysis [26]. The most common all-grade AEs were fatigue (18.26%), pruritus (10.61%), and diarrhea (9.47%). The most common grade 3 or higher AEs were fatigue (0.89%), anemia (0.78%), and aspartate aminotransferase increase (0.75%) [26]. In the present study, treatment-related AEs of any grade occurred in 71.4% and 85.7% of patients who received PD-1 inhibitor or PD-1 plus chemotherapy, respectively, and only 7 (20.0%) patients had grade 3 or 4 AEs. The most common AEs were fatigue, gastrointestinal symptoms, and blood system and skin disorders. Most importantly, immune-related myocarditis is a potentially fatal toxicity of ICIs therapy. Myocarditis after ICIs therapy may be more common than appreciated (prevalence of 1.14%), occurs early after starting treatment, has a malignant course, and responds to higher steroid doses [27]. Unfortunately, one patient (2.9%) in our cohort suffered from immune-related myocarditis, while required admission to the ICU and treatment with high dose steroid for recovery.

Prognostic factors help stratify patients for treatment by identifying patients with risks and potential benefits, and are important tools in the management of cancer [28]. Several studies have evaluated prognostic factors of survival in patients with advanced or metastatic UC treated with platinum-based chemotherapy or ICIs, and identified ECOG PS, hemoglobin level, liver metastasis, platelet count, NLR, and LDH as prognostic factors [13, 14]. The impact of ECOG PS on treatment outcomes of advanced UC treated with first line ICIs was evaluated in a real-world study, and patients with a PS of 0 to 1 had longer OS than patients with a PS ≥ 2 [13]. NLR is a simple index of systemic inflammation and a biomarker associated with tumor aggressiveness, and the majority of the studies reported a positive role of a high NLR (≥ 3) in the prediction of worse OS [29]. Ogihara et al. [30] found that elevated pretreatment NLR may be a novel biomarker for identifying poor responders to pembrolizumab among platinum-resistant metastatic UC patients. In our study, three pretreatment variables (ECOG PS, visceral metastases, and NLR) with univariate significance were selected to be included in the multivariate analysis using the Cox proportional hazards model, and ECOG PS and NLR were identified as independent prognostic factors, which are consistent with the previous studies [13, 30]. However, visceral metastases (including liver metastasis) did not show its prognostic value in our study. There were only 2 patients with liver metastasis in our entire cohort, and one of them achieved PR after treatment with PD-1 inhibitor. Therefore, the prognostic value of visceral metastases should be further investigated.

There are some limitations that must be considered in interpreting the results of our study. First of all, it is a retrospective, observational study with limited sample size. As the treatment outcomes of PD-1/PD-L1 inhibitors alone or combined with chemotherapy as first-line therapy for advanced UC are controversial in the well-designed clinical trials [10, 11], the results of our study likely reflect the ‘real-world’ clinical practice and patient care. Secondly, although many important factors that can influence long-term survival were investigated in the present study, our study does not rule out a potential role for other confounding variables. Thirdly, patients were treated with PD-1 inhibitor alone or combined with chemotherapy irrespective of PD-L1 status. PD-L1 expression is associated with improved response rate with PD-1 and PD-L1 inhibitors [31]. Thus, further validation is needed including data based on PD-L1 status and other biomarkers.