Search results
The bibliographic search yielded 5,319 unique citations. Following the initial titles and abstracts screening, 56 citations were retained for full-text screening. Of these, 11 studies met the inclusion criteria. Thus, we included 11 studies in the current systematic review (Fig. 1: PRISMA 2020 flow diagram). The patient population in each RCT is summarized in Table 1.
Table 1
Summary characteristics of the included studies
Trial | Design | Population | ICI regimen | CT Regimen | No. | Primary Endpoint | QoL Measure |
CHECKMATE 648 | Phase 3 single-blinded RCT | Previously untreated, unresectable advanced, recurrent, or metastatic esophageal squamous cell carcinoma | Nivolumab + Chemotherapy; Nivolumab + ipilimumab | Fluorouracil + cisplatin | 970 | PFS, OS | FACT-E |
ESCORT-1st | Phase 3 double-blinded RCT | Previously untreated advanced or metastatic esophageal squamous cell carcinoma | Camrelizumab + Chemotherapy | Paclitaxel + cisplatin | 596 | PFS, OS | QLQ-C30 and QLQ-OES18 |
KEYNOTE 062 | Phase 3 single-blinded RCT | Previously untreated advanced gastric/gastroesophageal junction adenocarcinoma | Pembrolizumab | 5-Fluorouracil or capecitabine + cisplatin | 495 | PFS, OS | QLQ-C30, QLQ-STO22 |
KEYNOTE 590 | Phase 3 double-blinded RCT | Previously untreated locally advanced/unresectable or metastatic adenocarcinoma or esophageal squamous cell carcinoma or Siewert type 1 esophagogastric junction adenocarcinoma | Pembrolizumab + Chemotherapy | 5-Fluorouracil + cisplatin | 711 | PFS, OS | QLQ-C30, QLQ-OES18, and EQ-5D-5L |
KEYNOTE 181 | Phase 3 open-label RCT | Previously treated advanced esophageal squamous cell carcinoma | Pembrolizumab | Paclitaxel, docetaxel, or irinotecan. | 404 | OS | QLQ-C30, OES18, and EQ-5D. |
KEYNOTE 061 | Phase 3 open-label RCT | Previously treated advanced gastric/gastroesophageal junction adenocarcinoma | Pembrolizumab | Paclitaxel | 371 | PFS. OS | QLQ-C30, QLQ-STO22, and EQ-5D-3L |
ORIENT-15 | Phase 3 double-blinded RCT | Previously untreated advanced or metastatic esophageal squamous cell carcinoma | Sintilimab + Chemotherapy | 5-Fluorouracil or capecitabine + cisplatin | 659 | OS | EQ-5D-5L |
ATTRACTION-3 | Phase 3 open-label RCT | Previously treated advanced esophageal squamous cell carcinoma | Nivolumab | Paclitaxel or docetaxel | 419 | OS | EQ-5D-5L |
ORIENT-16 | Phase 3 double-blinded RCT | Previously untreated advanced gastric/gastroesophageal junction adenocarcinoma | Sintilimab + Chemotherapy | Oxaliplatin + capecitabine | 650 | OS | EQ-5D-5L |
CheckMate 649 | Phase 3 open-label RCT | Previously untreated, unresectable, non-HER2-positive gastric, gastro-esophageal junction, or esophageal adenocarcinoma, | Nivolumab + Chemotherapy | Capecitabine and oxaliplatin or leucovorin, fluorouracil, and oxaliplatin | 1581 | PFS, OS | FACT-G, EQ-5D U I UK set [Not extracted], EQ-5D VAS |
ATTRACTION 4 | Phase 3 double-blinded RCT | Previously untreated, unresectable, non-HER2-positive gastric or gastro-esophageal junction | Nivolumab + Chemotherapy | Oxaliplatin + capecitabine | 724 | PFS, OS | FACT-G, EQ-5D 3L |
QoL (Quality of Life), RCT (Randomized controlled trial), PFS (Progression free survival), OS (overall survival), ICI (immune checkpoint inhibitor), CT (chemotherapy), QLQ-STO22 (Quality of Life Questionnaire-Stomach), EQ-5D (EuroQol 5 Dimension 5 Level), VAS (Visual analogue scale), QLQ-OES18 (Quality of Life Questionnaire – Oesophageal cancer-specific module), QLQ-C30 (Core Quality of Life Questionnaire), FACT-E (Functional Assessment of Cancer Therapy – Esophageal), FACT-G (Functional assessment of cancer therapy - Stomach) |
Characteristics of the included studies
Among the 11 phase-3 RCTs included, five were double-blinded, two were single-blinded, and the remaining were open-label trials. Seven trials, including CHECKMATE 648, KEYNOTE 061 ESCORT-1st, KEYNOTE 181, ATTRACTION-3, KEYNOTE 590, and ORIENT-15 [10,15,21–25], enrolled patients with previously untreated, advanced, or metastatic esophageal squamous cell carcinoma or adenocarcinoma. Three trials, KEYNOTE 062, ORIENT-16, and ATTRACTION 4 [26–28], focused on patients with previously untreated, advanced gastric/gastroesophageal junction adenocarcinoma. In the CheckMate 649 trial [29], patients with previously untreated, unresectable, non-HER2-positive gastroesophageal adenocarcinoma were recruited and received either Nivolumab monotherapy, single-agent chemotherapy, or Nivolumab plus chemotherapy. Most studies evaluated pembrolizumab (36%) or nivolumab (36%), and studies reported on 37 HRQoL tools (Table 1). The baseline characteristics of the included patients are summarized in Table 2.
Table 2
Baseline characteristics and survival outcomes of the included studies
Trial | Arms | No. | Age, median (range) | Male sex, no. (%) | ECOG ≥ 1, no. (%) | Histologic type, no. (%) | PD-L1 expression ≥ 1%, no. (%) | Metastatic, no. (%) | Number with metastases ≥ 2, no. (%) | Esophagectomy/gastrectomy | mPFS | mOS |
SSC | Adeno |
CHECKMATE 648 | Nivolumab + Chemotherapy | 321 | 64 (40–90) | 253 (79) | 171 (53) | 311 (97) | - | 158 (49) | 184 (57) | 163 (51) | 0 | 5.8 | 13.2 |
Nivolumab + ipilimumab | 325 | 64 (40–90) | 269 (83) | 174 (54) | 322 (99) | - | 158 (49) | 196 (60) | 165 (51) | 0 | 2.9 | 12.7 |
Chemotherapy | 324 | 64 (26–81) | 275 (85) | 170 (52) | 318 (98) | - | 157 (48) | 187 (58) | 166 (51) | 0 | 5.6 | 10.7 |
ESCORT-1st | Camrelizumab + Chemotherapy | 298 | 62 (56–66) | 260 (87.2) | 227 (76.2) | 298 (100) | - | 166 (55.7) | 298 (100) | 160 (53.7) | 116 (38.9) | 6.9 | 15.3 |
Chemotherapy | 298 | 62 (56–67) | 263 (88.3) | 232 (77.9) | 298 (100) | - | 163 (54.7) | 298 (100) | 141 (47.3) | 95 (31.9) | 5.6 | 12 |
KEYNOTE 062 | Pembrolizumab | 239 | - | - | - | - | - | - | - | - | - | - | - |
Chemotherapy | 234 | - | - | - | - | - | - | - | - | - | - | - |
KEYNOTE 590 | Pembrolizumab + Chemotherapy | 356 | - | - | - | - | - | - | - | - | - | - | - |
Chemotherapy | 355 | - | - | - | - | - | - | - | - | - | - | - |
KEYNOTE 181 | Pembrolizumab | 217 | - | - | - | - | - | - | - | - | - | - | - |
Chemotherapy | 188 | - | - | - | - | - | - | - | - | - | - | - |
KEYNOTE 061 | Pembrolizumab | 188 | - | - | - | - | - | - | - | - | - | - | - |
Chemotherapy | 183 | - | - | - | - | - | - | - | - | - | - | - |
ORIENT-15 | Sintilimab + Chemotherapy | 327 | 63 (57–67) | 279 (85) | 250 (76) | 327 (100) | - | 174 (53) | 285 (87) | - | 96 (29) | 7.2 | 16.7 |
Chemotherapy | 332 | 63 (56–67) | 288 (87) | 251 (76) | 332 (100) | - | 188 (57) | 287 (86) | - | 113 (34) | 5.7 | 12.5 |
ATTRACTION-3 | Nivolumab | 210 | 64 (57–69) | 179 (85) | 109 (52) | 210 (100) | - | 101 (48) | 94 (88) | 121 (58) | 111 (53) | 1.7 | 10.9 |
Chemotherapy | 209 | 67 (57–72) | 185 (89) | 102 (49) | 209 (100) | - | 102 (49) | 100 (83) | 118 (56) | 94 (45) | 3.4 | 8.4 |
ORIENT-16 | Sintilimab + Chemotherapy | 327 | - | - | - | - | - | - | - | - | - | - | - |
Chemotherapy | 323 | - | - | - | - | - | - | - | - | - | - | - |
CheckMate 649 | Nivolumab + Chemotherapy | 789 | 62 (54–69) | 540 (68) | 463 (59) | - | - | 126 (16) | 757 (96) | 602 (76) | - | 7.7 | 13.8 |
Chemotherapy | 792 | 61 (53–68) | 560 (71) | 455 (57) | - | - | 127 (16) | 756 (95) | 583 (74) | - | 6.9 | 11.6 |
ATTRACTION 4 | Nivolumab + Chemotherapy | 362 | 64 (25–86) | 253 (70) | 167 (46) | - | - | 58 (16) | - | 254 (70) | 105 (29) | 10.5 | 17.5 |
Chemotherapy | 362 | 65 (27–89) | 270 (75) | 168 (46) | - | - | 56 (15) | - | 257 (71) | 104 (29) | 8.3 | 17.2 |
PD-L1 (Programmed Cell Death Ligand 1), mPFS (mean progression free survival), mOS (mean overall survival), ECOG (Eastern Cooperative Oncology Group Performance Status), |
The pooled analysis favored the ICI arm (both monotherapy and combination regimens) over the chemotherapy arm in terms of Functional Assessment of Cancer Therapy-Esophageal (FACT-E; n = 2, Mean Difference [MD] 2.7 [95% 0.1, 5.3], p = 0.041), European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-Oesophageal Module (QLQ-OES18) pain scale (n = 3, MD -1.8 [95% CI -3.7, -0.0], p = 0.048), EuroQol-5D (EQ-5D) pain scale (n = 2, MD 5.5 [95% CI 2.6, 8.4], p < .001), and Quality of Life Questionnaire-Stomach Cancer Module (QLQ-STO22) dysphagia subscale (n = 2, MD 5.4 [95% CI 2.0, 8.7], p = 0.002). The pooled analyses were homogenous (Q p > 0.1, I2 = 0%). On the contrary, the pooled effect estimates showed comparable EORTC QLQ-C30 Global Health (p = 0.617) and EQ-5D index (p = 0.367) between study arms (Fig. 2).
In the subgroup analysis comparing ICIs plus chemotherapy versus chemotherapy alone, the effect estimates favored ICIs plus chemotherapy arm over chemotherapy arm regarding FACT-E (n = 2, MD 2.7 [95% CI 0.1, 5.3], p = 0.041) and EORTC QLQ-OES18 pain scale (n = 2 MD -2.2 [95% CI -4.3, -0.2], p = 0.030); the pooled analyses were homogenous (Q p > 0.1, I2 = 0%) (Fig. 3).
Likewise, the effect estimates favored the ICI only arms over chemotherapy arms regarding QLQ-STO22 hair loss subscale [MD -23.2 (95% CI -29.7 to -16.7), p < 0.001], QLQ-STO22 dysphagia subscale [MD 6.7 (95% CI 1.7 to 11.7), p = 0.009], EQ-5D pain scale [MD 6.9 (95% CI 2.9 to 10.9), p < 0.001], and QLQ-OES18 saliva subscale [MD 5.8 (95% CI 0.1 to 11.6), p = 0.046) (Fig. 4). There were no significant differences between ICI monotherapy and the chemotherapy arms in other HRQoL scales and subscales (p > 0.05), which can be attributed to the small number of trials in this subgroup.
There may be some degree of bias in QLQ-C30 Physical, Pain and Insomnia as shown in the supplementary Figs. 5,6, and 7.