Patient Characteristics: A total of 137 patients with advanced or recurrent NSCLC who received MONO or COMBO as their first-line therapy were analyzed. The NSCLC staging was based on the eighth edition of the TNM classification for lung cancer from the International Association for the Study of Lung Cancer. None of the patients was known to have any treatable driver oncogenes, such as epidermal growth factor receptor gene mutations or anaplastic lymphoma kinase gene rearrangements. Among the included patients, 84 were treated with MONO and 53 were treated with COMBO (Table 1). The median follow-up was 374 days (range; 6 to 1,220) for the MONO group and 394 days (range; 78 to 689) for the COMBO group. Patient characteristics, including smoking status, allergy history, histology, and brain metastases, were similar between the two groups. A history of allergy including asthma, pollinosis, food, drug, and contrast agents was observed in 44 patients (32.1%) in the whole population. Compared with the group treated with COMBO, patients in the MONO group were older (median: 73 vs. 69 years, p < 0.001) and had a poorer Eastern Cooperative Oncology Group (ECOG) performance status (p < 0.001). PD-L1 TPS, which was evaluated using the 22C3 pharmDx assay (Dako), was significantly higher in the MONO group compared with the COMBO group (p < 0.001). The patients with PD-L1 TPS ≧ 50% accounted for 84.7% in the MONO group and 17.0% in the COMBO group. However, because PD-L1 TPS was not always measured in the COMBO group, 12 patients (22.6%) had an unknown status. Compared to the COMBO group, prior anticancer therapies, such as surgery and radiation, were administered more frequently in the MONO group (p = 0.010). Of note, 23 patients (27.4%) receiving MONO had undergone surgery. Discontinuation of the first-line therapy because of disease progression, AEs, or patient death occurred in 55 of 84 (65.5%) patients in the MONO group and 34 of 53 (64.2%) in the COMBO group (p = 1.000). The second-line treatment after discontinuation of the first-line therapy was administered to 27 of 84 patients (32.1%) in the MONO group and 20 of 53 patients (37.7%) in the COMBO group, with no statistical difference. The frequency of adverse events (AEs), including immune-related AEs and interstitial lung pneumonitis, was not significantly different between the two groups. There was also no association between the frequency of AEs and PEC before the first or second administration of each therapy (data not shown).
Table 1
|
Overall
(n = 137)
|
Pembrolizumab monotherapy group
(n = 84)
|
Pembrolizumab plus chemotherapy group
(n = 53)
|
p-value
|
Age—median (range)
|
70 (41–88)
|
73 (46–88)
|
69 (41–76)
|
<0.001
|
Gender—male no. (%)
|
102 (74.5)
|
64 (76.2)
|
38 (71.7)
|
0.554
|
ECOG performance status score—no. (%)
|
|
|
|
|
0
|
23 (16.8)
|
11 (13.1)
|
12 (22.6)
|
<0.001
|
1
|
82 (59.9)
|
42 (50.0)
|
40 (75.5)
|
|
2
|
23 (16.8)
|
23 (27.4)
|
0 (0)
|
|
3 or 4
|
9 (6.5)
|
8 (9.5)
|
1 (1.9)
|
|
Smoking status—no. (%)
|
|
|
|
|
Current
|
42 (30.7)
|
23 (27.4)
|
19 (35.8)
|
0.749
|
Former
|
80 (58.4)
|
51 (60.7)
|
29 (54.7)
|
|
Never
|
14 (10.2)
|
9 (10.7)
|
5 (9.4)
|
|
Unknown
|
1 (0.7)
|
1 (1.2)
|
0 (0)
|
|
Allergic history—no. (%)
|
44 (32.1)
|
26 (31.0)
|
18 (34.0)
|
0.712
|
Histology—no. (%)
|
|
|
|
|
Squamous
|
45 (32.8)
|
27 (32.1)
|
18 (34.0)
|
0.853
|
Nonsquamous
|
92 (67.2)
|
57 (67.9)
|
35 (66.0)
|
|
Brain metastases—no. (%)
|
19 (13.9)
|
11 (13.1)
|
8 (15.1)
|
0.802
|
PD-L1 TPS—no. (%)
|
|
|
|
|
<1%
|
14 (10.2)
|
0 (0)
|
14 (26.4)
|
<0.001
|
1–49%
|
31 (22.6)
|
13 (15.3)
|
18 (34.0)
|
|
≧50%
|
80 (58.4)
|
71 (84.7)
|
9 (17.0)
|
|
Unknown
|
12 (8.8)
|
0 (0)
|
12 (22.6)
|
|
Prior anticancer therapy—no. (%)
|
|
|
|
|
None
|
96 (70.0)
|
51 (60.7)
|
45 (84.9)
|
0.010
|
Chemoradiation therapy
|
9 (6.6)
|
6 (7.1)
|
3 (5.7)
|
|
Surgery
|
28 (20.5)
|
23 (27.4)
|
5 (9.4)
|
|
Radiation
|
4 (2.9)
|
4 (4.8)
|
0 (0)
|
|
Adverse events—no. (%)
|
32 (23.4)
|
20 (23.8)
|
12 (22.6)
|
1.000
|
Discontinuation due to disease progression, adverse events, or patient death—no. (%)
|
89 (65.0)
|
55 (65.5)
|
34 (64.2)
|
1.000
|
Administration of second-line treatment after discontinuation of first-line therapy—no. (%)
|
47 (34.3)
|
27 (32.1)
|
20 (37.7)
|
0.580
|
Peripheral eosinophil count—median (range) (×106/L)
|
|
|
|
|
Eosinophil count before first administration of the treatment
|
160 (0–2,840)
|
155 (0–2,840)
|
169 (20–690)
|
0.086
|
Eosinophil count before second administration of the treatment
|
150 (0–1,590)
|
185 (0–1,590)
|
90 (0–670)
|
0.002
|
Absolute PECs were assessed on the nearest day before the administration of each anticancer therapy. In the whole population, the median PEC was 160 × 106/L (range 0–2, 840 × 106/L) before the first administration of MONO or COMBO and 150 × 106/L (range 0–1,590 × 106/L) before the second administration. Therefore, we set 150 × 106/L as the cut-off to distinguish high and low PEC for further analyses. PEC before the first administration of therapy (PEC1) was 155 × 106/L (range 0–2,840 × 106/L) in the MONO group and 169 × 106/L (20–690 × 106/L) in the COMBO group (p = 0.086). Interestingly, PEC before the second administration (PEC2) was significantly higher in the MONO group (median: 185 × 106/L [0–1,590]), compared with the COMBO group (90 × 106/L [0–670]) (p = 0.002). The differences between PEC1 and PEC2 were determined in the two groups. Elevated, stable, or decreased PEC was observed in 41, 4, and 39 patients in the MONO group (p = 0.307) and in 12, 4, and 37 patients (p = 0.006) in the COMBO group, respectively. In other words, PEC2 in the COMBO group was significantly decreased compared with PEC1 (p = 0.002).
Analysis of Survival and Treatment Response in the Whole Population: To test the efficacy of PEC as a biomarker for selecting MONO or COMBO treatment, we checked the survival status and treatment response in this dataset. Evaluation of the overall response rate (ORR) was based on the Response Evaluation Criteria in Solid Tumors, version 1.1: complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD), or not evaluated (NE).
The Kaplan–Meier analysis revealed no significant differences in PFS (median: 10.8 months [95% confidence interval (CI): 6.44–16.1] vs. 8.41 months [5.95–19.6], Fig. 1a) or OS (median: 19.6 months [95% CI: 14.2-NR] vs. NR [15.7–NR], Fig. 1b) between the MONO and COMBO groups. However, the ORR was significantly higher in the COMBO group compared with the MONO group (62.3% vs. 46.5%, p = 0.041) (Table 2).
Table 2
Response for each regimen.
no. (%)
|
Pembrolizumab monotherapy group (n = 84)
|
Pembrolizumab plus chemotherapy group (n = 53)
|
p-value
|
CR+PR
|
39 (46.5)
|
33 (62.3)
|
0.041
|
CR+PR+SD
|
55 (65.5)
|
48 (90.6)
|
|
CR
|
3 (3.60)
|
0 (0)
|
|
PR
|
36 (42.9)
|
33 (62.3)
|
|
SD
|
16 (19.0)
|
15 (28.3)
|
|
PD
|
22 (26.2)
|
5 (9.4)
|
|
NE
|
7 (8.30)
|
0 (0)
|
|
Is PEC Useful to Decide Whether to Choose Pembrolizumab Monotherapy or Combination Therapy Plus Chemotherapy? We further investigated whether PEC can guide the selection of MONO or COMBO therapy for the first-line treatment of metastatic NSCLC. In patients with high or low PEC1, the Kaplan–Meier analysis revealed no statistical differences between the MONO therapy and the COMBO therapy (Supplementary Fig. 1). In the high PEC1 subgroup, the median PFS was 13.6 months (95% CI: 8.87–NR, n = 46) in the MONO group vs. 10.6 months (5.29–NR, n = 27) in the COMBO group (p = 0.872) (Supplementary Fig. 1a). The median OS was 21.1 months (95% CI: 12.1–NR) in the MONO group vs. NR (15.7–NR) in the COMBO group (p = 0.239, Supplementary Fig. 1b). In the low PEC1 subgroup, the median PFS was 8.51 (95% CI: 1.71–11.6, n = 38) in the MONO group vs. 7.03 months (4.01–19.6, n = 26) in the COMBO group (p = 0.721, Supplementary Fig. 1c). The median OS was 16.8 months (95% CI: 10.5–25.4) in the MONO group vs. 16.3 months (14.9–NR) in the COMBO group (p = 0.506, Supplementary Fig. 1d).
Next, we evaluated PEC2 with respect to survival analysis. The Kaplan–Meier results indicated that the high PEC2 subgroup showed no statistical differences between the MONO (n = 53) and COMBO (n = 17) groups (Fig. 1c and 1d). The median PFS was 16.1 months (95% CI: 10.6–NR) in the MONO group vs. 10.1 months (3.68–NR) in the COMBO group (p = 0.462, Fig. 1c). The median OS was not reached in either the MONO (95% CI: 18.2–NR) or COMBO (8.84–NR) group (p = 0.885, Fig. 1d). However, survival analysis using low PEC2 revealed that patients with low PEC2 receiving the MONO therapy (n = 31) had an unfavorable tendency of PFS (Fig. 1e) and a statistically shorter OS (Fig. 1f), compared with those receiving the COMBO therapy (n = 36). In the low PEC2 subgroup, the median PFS was 5.75 (95% CI: 1.64–10.3) in the MONO group vs. 7.59 months (5.95–NR) in the COMBO group (p = 0.054, Fig. 1e). The median OS was 12.0 (95% CI: 5.52–19.6) in the MONO group vs. NR (15.1–NR) in the COMBO group (p = 0.019, Fig. 1f). These results indicate that PEC2 is a potential prognostic biomarker that warrants further analysis.
Can PEC Predict the Subgroup with a Survival Benefit from the Pembrolizumab Monotherapy? We investigated whether PEC2 can predict the subgroup that has a favorable survival benefit from the MONO therapy. The Kaplan–Meier analysis demonstrated that patients with high PEC2 had significantly better survival rates compared with those with low PEC2. The median PFS was 16.1 months (95% CI: 10.6–NR) in the high PEC2 subgroup (n = 53) vs. 5.75 months (1.64–10.3) in the low PEC2 subgroup (n = 31, p = 0.001, Fig. 2a). The median OS was NR (95% CI: 18.2–NR) in the high PEC2 subgroup vs. 12.0 months (5.52–19.6) in the low PEC2 subgroup (p = 0.005, Fig. 2b). The response rate for the MONO group was significantly higher in the high PEC2 subgroup (56.6%) compared with the low PEC2 subgroup (29.0%) (p = 0.024, Table 3).
Table 3
Responses stratified according to PEC2 for each regimen
Pembrolizumab monotherapy
no. (%)
|
High PEC2 subgroup (n = 53)
|
Low PEC2 subgroup (n = 31)
|
p-value
|
CR+PR
|
30 (56.6)
|
9 (29.0)
|
0.024
|
CR+PR+SD
|
38 (71.7)
|
17 (54.8)
|
|
CR
|
2 (3.8)
|
1 (3.2)
|
|
PR
|
28 (52.8)
|
8 (25.8)
|
|
SD
|
8 (15.1)
|
8 (25.8)
|
|
PD
|
10 (18.9)
|
12 (38.7)
|
|
NE
|
5 (9.4)
|
2 (6.5)
|
|
Pembrolizumab plus chemotherapy
no. (%)
|
High PEC2 subgroup (n = 17)
|
Low PEC2 subgroup (n = 36)
|
p-value
|
CR+PR
|
9 (52.9)
|
24 (66.7)
|
0.375
|
CR+PR+SD
|
15 (88.2)
|
33 (91.7)
|
|
CR
|
0 (0)
|
0 (0)
|
|
PR
|
9 (52.9)
|
24 (66.7)
|
|
SD
|
6 (35.3)
|
9 (25.0)
|
|
PD
|
2 (11.8)
|
3 (8.3)
|
|
NE
|
0 (0)
|
0 (0)
|
|
A univariable Cox proportional hazard regression analysis for PFS revealed that the histology of squamous NSCLC (p = 0.048), discontinuation of treatment (p < 0.001), administration of the second-line treatment after discontinuation of the first-line therapy (p = 0.013), and low PEC2 (p = 0.002) were significantly associated with poor PFS (Table 4). The univariable analysis for OS revealed that the histology of squamous NSCLC (p = 0.031), discontinuation because of disease progression, AEs, death (p = 0.002), and low PEC2 (p = 0.006) were significantly associated with shorter OS.
Table 4
Univariable analysis of PFS and OS for pembrolizumab monotherapy
|
|
PFS
|
OS
|
Variables
|
Category
|
Hazard ratio
|
95% CI
|
p-value
|
Hazard ratio
|
95% CI
|
p-value
|
Age
|
≧75
|
1.48
|
0.79–2.77
|
0.222
|
1.80
|
0.97–3.33
|
0.061
|
Gender
|
male
|
0.96
|
0.45–2.01
|
0.904
|
1.03
|
0.49–2.15
|
0.941
|
ECOG-PS score
|
0–1
|
1.40
|
0.74–2.61
|
0.302
|
1.58
|
0.85–2.94
|
0.151
|
Smoking status
|
Yes
|
1.33
|
0.47–3.75
|
0.585
|
1.51
|
0.54–4.22
|
0.435
|
Allergic history
|
Yes
|
0.91
|
0.47–1.77
|
0.778
|
0.80
|
0.42–1.54
|
0.513
|
Histology
|
Squamous
|
1.88
|
1.01–3.51
|
0.048
|
1.94
|
1.06–3.53
|
0.031
|
Brain metastases
|
Yes
|
0.37
|
0.09–1.53
|
0.168
|
1.03
|
0.40–2.61
|
0.958
|
PD-L1 TPS
|
≧50
|
0.66
|
0.27–1.59
|
0.354
|
0.54
|
0.25–1.19
|
0.125
|
Prior anticancer therapy
|
Yes
|
1.26
|
0.68–2.35
|
0.461
|
1.32
|
0.72–2.40
|
0.367
|
Adverse events
|
Yes
|
0.59
|
0.28–1.27
|
0.180
|
0.98
|
0.48–1.99
|
0.955
|
Discontinuation due to disease progression, adverse events, or patient death
|
Yes
|
11.2
|
3.90–31.7
|
< 0.001
|
3.32
|
1.52–7.24
|
0.002
|
Administration of second-line treatment after discontinuation of the first-line therapy
|
Yes
|
2.12
|
1.17–3.85
|
0.013
|
1.18
|
0.64–2.18
|
0.600
|
Eosinophil count before first administration of the treatment
|
<150 ×106/L
|
1.70
|
0.91–3.18
|
0.094
|
1.35
|
0.74–2.46
|
0.326
|
Eosinophil count before second administration of the treatment (×106/L)
|
< 150 ×106/L
|
2.64
|
1.41–4.96
|
0.002
|
2.31
|
1.27–4.22
|
0.006
|
A multivariable analysis for PFS and OS was performed with parameters that had p values < 0.1 in the univariable analysis to identify independent parameters (Table 5). The multivariable analysis for PFS revealed that low PEC2 (p = 0.043), discontinuation because of disease progression, AEs, or death (p < 0.001) were significantly associated with unfavorable PFS. The multivariable analysis for OS revealed that discontinuation of treatment (p = 0.002) was significantly associated with unfavorable OS; however, PEC2 did not reach statistical significance (p = 0.061).
Table 5
Multivariable analysis of PFS and OS for pembrolizumab monotherapy
|
|
PFS
|
OS
|
Variables
|
Category
|
Hazard ratio
|
95% CI
|
p-value
|
Hazard Ratio
|
95% CI
|
p-value
|
Age
|
≧75
|
|
|
|
1.84
|
0.98–3.46
|
0.058
|
Histology
|
Squamous
|
1.22
|
0.67–2.21
|
0.523
|
1.31
|
0.70–2.45
|
0.398
|
Discontinuation due to disease progression, adverse events, or patient death
|
Yes
|
10.0
|
3.48–28.8
|
<0.001
|
3.32
|
1.53–7.24
|
0.002
|
Administration of second-line treatment after discontinuation of the first-line therapy
|
Yes
|
0.89
|
0.48–1.65
|
0.714
|
|
|
|
Eosinophil count before the first administration of the treatment
|
<150 ×106/L
|
1.10
|
0.47–2.58
|
0.822
|
|
|
|
Eosinophil count before the second administration of the treatment
|
<150 ×106/L
|
1.84
|
1.02–3.32
|
0.043
|
1.82
|
0.97–3.40
|
0.061
|
Can PEC Identify the Subgroup with a Survival Benefit from Pembrolizumab Plus Chemotherapy? We determined whether PEC2 could identify the subgroup with improved survival by the COMBO therapy. The Kaplan–Meier analysis revealed no statistical differences between the high and low PEC2 subgroups. The median PFS was 10.1 months (95% CI: 3.68–NR) in the high PEC2 subgroup (n = 17) vs. 7.59 months (5.95–19.6) in the low PEC2 subgroup (n = 36, p = 0.916, Fig. 2c). The median OS was not reached in either subgroup (95% CI: 8.84–NR in the high PEC2 subgroup vs. 15.1–NR in the low PEC2 subgroup, p = 0.892, Fig. 2d). No relationship was observed between PEC2 and the ORR when comparing high and low PEC2 subgroups. The ORRs were 52.9% and 66.7%, respectively (p = 0.375, Table 3).