Low baseline plasma PCSK9 level is associated with good clinical outcomes of immune checkpoint inhibitors in advanced non‐small cell lung cancer

Patient characteristics

A total of 55 patients with advanced NSCLC were enrolled in the study (Figure 1). The median age of all patients was 66 years (range, 48–77 years) and 85.5% were male. The median body mass index (BMI) was 23.1 (range, 16.1–36.1). The majority of patients were stage IV (35, 63.6%) and never smokers (29, 52.7%). A total of 40 patients underwent a test for PD-L1, 13 (32.5%) were positive and 27 (67.5%) were negative. A total of 15 patients (27.3%) had bone metastasis at baseline. Most patients (39,70.9%) received first-line ICI combination therapy and all patients (55,100%) had an ECOG PS of 0 or 1. The median follow-up period was 7.7 months.

image

Flow chart of the study. NSCLC, non-small cell lung cancer; ICI, immune checkpoint inhibitor

The median baseline plasma PCSK9 level was 279.3 ng/ml (range, 144.3–495.4 ng/ml) in the whole cohort. The optimal threshold level for plasma PCSK9 level was calculated by X-tile, and the plasma PCSK9 concentration of 232.2 ng/ml provided the best threshold. Thus, 14 patients were categorized to the low PCSK9 group (≤232.2 ng/ml) and 41 patients to the high group (>232.2 ng/ml) (Figure 2a). Among the low group, seven (50.0%) patients were tested for PD-L1 and four (57.1%) patients were positive. As for the high group, 33 (80.5%) patients had the PD-L1 test, and nine (27.3%) patients were positive. The baseline characteristics of patients in both groups were well-balanced. The baseline clinical characteristics of patients are shown in Table 1.

image

Association between baseline plasma PCSK9 level and efficacy to ICIs therapy. Low group: PCSK9 ≤ 232.2 ng/ml; High group: PCSK9 > 232.2 ng/ml. (a) The baseline plasma PCSK9 level between two groups. (b) PFS of total population enrolled in study. (c) PFS comparison between patients with low level of PCSK9 and high level of PCSK9. (d) Best percent change in the target tumor burden from baseline in total population. (e) Treatment response towards ICI therapy between patients with low and high levels of PCSK9. PCSK9, proprotein convertase subtilisin/kexin type 9; ICIs, immune checkpoint inhibitors; PFS, progress-free survival

TABLE 1. Baseline demographic and clinical characteristics of patients All (%) (n = 55) PCSK9 ≤ 232.2 ng/ml (n = 14) PCSK9>232.2 ng/ml (n = 41) p-value Age (years) 0.316 ≤66 29 (52.7) 9 (64.3) 20 (48.8) >66 26 (47.3) 5 (35.7) 21 (51.2) Gender 0.638 Male 47 (85.5) 13 (92.9) 34 (82.9) Female 8 (14.5) 1 (7.1) 7 (17.1) BMI 0.279 <25 36 (65.5) 7 (50.0) 29 (70.7) ≥25 19 (34.5) 7 (50.0) 12 (29.3) Smoking history 0.392 Never 29 (52.7) 6 (42.9) 23 (56.1) Ever or current 26 (47.3) 8 (57.1) 18 (43.9) Histology 0.925 Adenocarcinoma 25 (45.5) 6 (42.9) 19 (46.3) Squamous 17 (30.9) 5 (35.7) 12 (29.3) NOS 13 (23.6) 3 (21.4) 10 (24.4) Stage 0.953 IIIb/IIIc 20 (36.4) 5 (35.7) 15 (36.6) IV 35 (63.6) 9 (64.3) 26 (63.4) Metastasis sitea Brain 8 (14.5) 1 (7.1) 7 (17.1) 0.638 Bone 15 (27.3) 5 (35.7) 10 (24.4) 0.636 Liver 3 (5.5) 1 (7.1) 2 (4.9) 1.000 No. of treatment lines 0.696 1 39 (70.9) 11 (78.6) 28 (68.3) ≥ 2 16 (29.1) 3 (21.4) 13 (31.7) Treatment regime 1.000 Monotherapy 6 (10.9) 1 (7.1) 5 (12.2) Combination therapy 49 (89.1) 13 (92.9) 36 (87.8) PD-L1 status 0.187 Positive 13 (23.6) 4 (28.6) 9 (22.0) Negative 27 (49.1) 3 (21.4) 24 (58.5) Not clear 15 (27.3) 7 (50.0) 8 (19.5) Cholesterol metabolismb Cholesterol 4.4 (2.9–7.6) 4.8 (3.9–6.3) 4.4 (2.9–7.6) 0.162 Triglyceride 1.4 (0.5–4.3) 1.6 (0.7–2.8) 1.3 (0.5–4.3) 0.388 HDL 1.1 (0.7–2.4) 1.1 (0.8–1.9) 1.1 (0.7–2.4) 0.570 LDL 2.8 (1.4–5.1) 3.2 (2.3–5.1) 2.7 (1.4–4.9) 0.364 Apo A1 1.2 (0.7–2.1) 1.2 (1.0–2.1) 1.3 (0.7–2.0) 0.829 Apo B 1.0 (0.6–1.4) 1.0 (0.6–1.4) 1.0 (0.6–1.3) 0.361 Apo E 38.1 (19.0–80.6) 38.6 (26.2–54.4) 38.5 (19.0–80.6) 0.683 Lp(a) 176.2 (10.4–1083.4) 152.2 (20.2–884.6) 246.5 (10.4–1083.4) 0.173 SdLDL 0.9 (0.3–1.8) 1.1 (0.5–1.3) 0.9 (0.3–1.8) 0.054 Lipase 31.2 (21.5–63.8) 32.3 (23.2–45.9) 31.1 (21.5–63.8) 0.743 Association between baseline plasma PCSK9 level and clinical outcomes of ICIs

The median PFS of the entire cohort was 9.9 months (Figure 2b). The median PFS was longer in the low PCSK9 group than that in the high group (NR vs.7.37 months, p = 0.017, HR = 0.207, 95% CI: 0.086–0.498) (Figure 2c). The low group had a better response to ICI therapy, but the difference was not significant (ORR 71.4% vs. 43.9%, p = 0.075, DCR 100% vs. 80.5%, p = 0.098) (Figure 2d, e).

Due to the vital role of PD-L1 in ICIs therapy, we divided patients into PD-L1 positive and negative groups. Low baseline PCSK9 level was associated with longer median PFS in both groups, although the difference was not significant (NR vs. 6.30 months, p = 0.067, HR = 0.184, 95% CI: 0.030–1.124; NR vs. 9.90 months, p = 0.145, HR = 0.267, 95% CI: 0.045–1.578) (Figure 3a, b). In addition, in the high PCSK9 group, patients with PD-L1 ≥ 1% showed a shorter median PFS compared with those <1%, but with no significant difference (6.30 vs. 9.90 months, p = 0.919, HR = 1.058, 95% CI: 0.361–3.096) (Figure 3c).

image

PFS of different PD-L1 expression. (a) PFS for patients PD-L1 ≥ 1%. (b) PFS for patients PD-L1 <1%. (c) PFS for patients with high level of baseline PCSK9. PD-L1, programmed cell death-ligand 1

Younger patients (≤66 years) with a low level of PCSK9 had a longer median PFS and a higher treatment response than those with a high level of PCSK9 (NR vs. 5.83 months, p = 0.021, HR = 0.134, 95% CI: 0.044–0.409; ORR 66.7% vs. 30.0%, p = 0.106, DCR 100% vs. 75%, p = 0.153) (Figure 4a, b), but the difference was not observed in elder patients (>66 years) (NR vs. 8.97 months, p = 0.499, HR = 0.509, 95% CI: 0.101–2.564; ORR 80.0% vs. 57.1%, p = 0.617, DCR 100% vs. 85.7%, p = 1.000) (Figure 4c, d). In addition, for patients receiving first-line treatment, the low PCSK9 group had a longer PFS and a higher treatment response than the high group (NR vs. 8.97 months, p = 0.022, HR = 0.138, 95% CI: 0.047–0.400; ORR 63.6% vs. 46.4%, p = 0.480, DCR 100% vs. 89.3%, p = 0.545) (Figure 4e, f), while there was no difference in patients who received subsequent lines of ICIs (NR vs. 7.37 months, p = 0.488, HR = 0.495, 95% CI: 0.094–2.611; ORR 100% vs. 38.5%, p = 0.200, DCR 100% vs. 61.5%, p = 0.509) (Figure 4g, h).

image

PFS and treatment response of subgroups in total population. Younger patients: ≤66; Elder patients >66. (a) PFS for younger patients. (b) Treatment response for younger patients. (c) PFS for elder patients. (d) Treatment response for elder patients. (e) PFS for patients with first-line treatment. (f) Treatment response for patients with first-line treatment. (f) PFS for patients with subsequent treatment. (h) Treatment response for patients with subsequent treatment

In addition, we performed correlation analysis between plasma PCSK9 level and peripheral biochemical indicators due to the role of PCSK9 in cholesterol metabolism reported in previous studies. As shown in Table 2, baseline PCSK9 level was significantly correlated with baseline lipoprotein (a) (Lp[a]) level (r = 0.409, p = 0.003). Kaplan–Meier analysis was also conducted but no significant differences were found between the two groups with different levels of cholesterol metabolism.

TABLE 2. Correlations between baseline PCSK9 level and cholesterol metabolism PCSK9 r p Cholesterol (mmol/l) −0.057 0.696 Triglyceride (mmol/l) −0.189 0.189 HDL (mmol/l) 0.051 0.726 LDL (mmol/l) 0.065 0.655 Apo A1 (g/l) −0.054 0.708 Apo B (g/l) 0.016 0.911 Apo E (g/l) 0.025 0.865 Lp(a) (mg/l) 0.409 0.003 SdLDL (mmol/l) −0.217 0.129 Lipase (u/l) 0.003 0.982 Univariate and multivariate analysis of clinical characteristics for clinical outcomes

The univariate analysis for PFS suggested that combination therapy (p = 0.056, HR = 0.344) and low level of PCSK9 (≤232.2 ng/ml) (p = 0.03, HR = 0.197) were associated with longer median PFS. The multivariate analysis showed that only PCSK9 concentration was independently associated with PFS (p = 0.032, HR = 0.201). The detailed information of univariate and multivariate analysis is shown in Table 3.

TABLE 3. Univariate and multivariate analyses of clinical parameters of PFS in patients Factors Univariate analyses Multivariate analyses HR 95% CI p-value HR 95% CI p-value Age (years) ≤66/>66 0.995 0.426–2.325 0.991 Gender Male/female 0.978 0.330–2.902 0.968 BMI <25/≥25 1.274 0.498–3.257 0.614 Smoking history Ever or current/never 0.855 0.367–1.990 0.716 History Adeno/NOS 2.133 0.593–7.673 0.246 Squa/NOS 2.607 0.695–9.780 0.155 PD-L1 Positive/negative 0.667 0.233–1.907 0.449 Stage IIIb or IIIc/IV 0.703 0.288–1.713 0.438 Metastasis sitea Bone (yes/no) 1.175 0.482–2.865 0.722 No. of treatment lines 1/≥2 0.582 0.240–1.410 0.231 Treatment regime Combination/monotherapy 0.344 0.115–1.028 0.056 0.360 0.120–1.076 0.067 PCSK9, ug/ml ≤232.2/>232.2 0.197 0.045–0.855 0.030 0.201 0.046–0.872 0.032

留言 (0)

沒有登入
gif