Clinical verification of the relationship between serum lipid metabolism and immune activity in breast cancer patients treated with neoadjuvant chemotherapy

Data from the Osaka City University Graduate School of Medicine (Osaka, Japan) between April 2007 and March 2018 were analyzed. A total of 351 patients were diagnosed with early stage breast cancer (stage IIA, IIB, IIIA) and underwent with primary systemic treatment and curative surgery. We excluded 24 patients treated with neoadjuvant endocrine therapy and included 327 patients treated with NAC in this study. T and N factors and tumor stage were stratified based on the TNM Classification, UICC Seventh Edition [27]. Tumors were classified into intrinsic subtypes according to the immunohistochemical expression of the estrogen receptor (ER), progesterone receptor (PgR), and HER2. We defined ER + and/or PgR + and HER2-breast cancer as luminal, ER + and/or PgR + and HER2 + breast cancer as luminal-HER2, ER- and PgR-, HER2 + breast cancer as HER2-enriched, and ER-, PgR-, and HER2-breast cancer as TNBC. The antitumor effect was assessed according to the Response Evaluation Criteria in Solid Tumors [28]. The objective response (OR) was calculated as the sum of the clinical partial response and complete response (CR). All the patients underwent mastectomy or breast-conserving surgery after NAC. The pCR was defined as the complete disappearance of the invasive compartment of the lesion with or without intraductal components, including the lymph nodes” [29]. Postoperative adjuvant therapy suitable for each intrinsic breast cancer subtype was performed, and standard postoperative radiotherapy to the remnant breast was administered if necessary. All patients underwent physical examinations, blood tests, ultrasonography, computed tomography, and bone scintigraphy scans. Overall survival (OS) was defined as the time from curative surgery to death from any cause, and recurrence-free survival (RFS) was defined as freedom from all locoregional and distant recurrences. The median follow-up time for the assessment of OS was 5.5 years (range 0.2–12.4 years) and for RFS was 4.9 years (range 0.1–12.0 years).

Peripheral blood samples were obtained at the time of diagnosis, and preoperative blood samples were obtained within a week before surgery. We evaluated the serum lipid levels, including total cholesterol (TC) levels [categorized as low (≤ 149 mg/dl), normal (150–219 mg/dl), and high (≥ 220 mg/dl)] and triglyceride (TG) levels [categorized as low (≤ 49 mg/dl), normal (50–149 mg/dl), and high (≥ 150 mg/dl)]. The differential white blood cell counts were analyzed using a Coulter LH 750 Hematology Analyzer (Beckman Coulter, Brea, CA, USA). The neutrophil-to-lymphocyte ratio (NLR) was calculated from the blood samples by dividing the absolute neutrophil count by the absolute lymphocyte count (ALC).

Statistical analyses were performed using the JMP13 software (SAS Institute, Cary, NC, USA). Associations among the variables were analyzed using the χ² or Fisher’s exact test, as appropriate. OS and RFS were estimated using the Kaplan–Meier method and log-rank test. Statistical significance was set at P < 0.05.

The differences in clinicopathological features due to intrinsic breast cancer subtypes are presented in Table 1. A total of 327 patients were included in this study. Among these, 108 (33.0%), 42 (12.9%), 72 (22.0%), and 105 (32.1%) had luminal, luminal-HER2, HER2-enriched, and TNBC, respectively. NAC-related pCR was observed in 121 patients (37.0%). The evaluation based on the clinicopathological features revealed that the pCR rate was significantly higher in HER2-enriched (59.7%, 43/72) and TNBC patients (44.8%, 47/105) (P < 0.001). OR was observed in 295 patients (90.2%). The OR rate was high in all the breast cancer subtypes, and no significant differences were observed (P = 0.070). Patients with high TC levels increased after NAC in each breast cancer subtypes other than HER2-enriched. Furthermore, patients with TG levels increased after NAC in all subtypes, and the rate of change was highest especially in TNBC (21.0% → 48.1%).

In all breast cancer patients, RFS and OS were significantly longer in patients who achieved pCR than in those who did not (P < 0.001 and P = 0.006, log-rank, respectively; Additional file 1: Fig. S1a, Additional file 2: Fig. S2a). Furthermore, these outcomes were also significantly better in patients who achieved OR than in those who did not (P < 0.001 and P = 0.001, log-rank, respectively; Additional file 3: Figs. S3a, Additional file 4: Fig. S4a). In addition, we investigated the prognostic factors for RFS and OS for each breast cancer subtype. Among patients with luminal cancer, no significant differences were observed in RFS (P = 0.882, P = 0.399, log-rank, respectively) and OS (P = 0.861, P = 0.202, log-rank, respectively) according to the clinicopathological responses, pCR, and OR (Additional file 1: Fig. S1b, Additional file 2: Fig. S2b, Additional file 3: Fig. S3b, Additional file 4: Fig. S4b). In contrast, among the patients with TNBC, RFS (P = 0.005, P < 0.001, log-rank, respectively) and OS (P = 0.003, P < 0.001, log-rank, respectively) were significantly longer in patients who achieved pCR or OR than in those who did not (Additional file 1: Fig. S1e, Additional file 2: Fig. S2e, Additional file 3: Fig. S3e, Additional file 4: Fig. S4e).

The relationship between lipid metabolism and chemosensitivity was examined (Table 2). There were no significant correlations between lipid metabolism and the pCR in any breast cancer subtype. In contrast, only TNBC patients with OR had significantly higher TG levels after NAC than patients without OR (P = 0.049).

We also investigated the prognostic value of serum lipid levels before and after NAC for each intrinsic breast cancer subtype. In patients with HER2-enriched breast cancer, those with normal TC levels before NAC had a significantly better OS than those with high TC levels (P = 0.013, log-rank test) (Fig. 1d), and in patients with TNBC, the group with high TC levels after NAC had significantly better OS than those with normal TC levels (P = 0.014, log-rank test) (Fig. 2e). There was no association between recurrence and TC levels. Also, there was no relationship between the prognosis and triglyceride levels before and after NAC (Additional files 5–10: Figs. S5–S10).

The pre-NAC ALC ranged from 712.8 to 4446.2 (mean, 1811.0; median, 1749; standard deviation, 613.9), and the pre-NAC NLR ranged from 0.5 to 10.6 (mean, 2.3; median, 2.0; standard deviation, 1.2). The post-NAC ALC ranged from 285.6 to 3697.7 (mean, 1122.4; median, 1005.4; standard deviation, 517.0), and the post-NAC NLR ranged from 0.3 to 15.9 (mean, 2.9; median, 2.4; standard deviation, 1.9). We defined the pre-NAC median as the cutoff value for the ALC and NLR. There were no significant correlations between the systemic immune activity and the effect of NAC in all the breast cancer patients or each of the breast cancer subtypes (Additional file 11: Table S1).

The relationship between lipid metabolism and systemic immune activity is shown in Table 3. Patients with a high ALC before NAC had significantly higher TG levels after NAC in all the breast cancers (P = 0.001). In addition, among the patients with HER2-enriched breast cancer, high TG levels after NAC were associated significantly with a high ALC before NAC (P = 0.021), and high TG levels before NAC were associated significantly with a high ALC after NAC (P = 0.046). Furthermore, among patients with TNBC, high TG levels after NAC were associated significantly with a high ALC (P = 0.008) and a low NLR (P = 0.025) before NAC, while high TG levels before NAC were associated significantly with a low NLR after NAC (P = 0.034).