The role of prognostic nutritional index in the management of pulmonary sarcomatoid carcinoma

Background Pulmonary sarcomatoid carcinoma is characterized by poor survival rates compared with other non-small cell lung cancer. Prognostic nutritional index has significant prognostic value in many malignant tumors. We conducted this retrospective study to investigate the role of prognostic nutritional index in patients with pulmonary sarcomatoid carcinoma and to determine prognostic factors. Methods Of 8176 patients with resected lung cancer in a single high-volume institution between 2008 and 2015, 91 patients with pathologically diagnosed sarcomatoid carcinoma were included in our study and evaluated. Kaplan–Meier analysis and Cox regression analysis were conducted to analyze clinicopathologic data. Subgroup analysis of overall survival (OS) and recurrence-free survival (RFS) among pulmonary sarcomatoid carcinoma patients were also conducted. Results Univariable analysis showed that tumor size (P = 0.018 in OS), and P = 0.021 in RFS), tumor stage(P < 0.001 in OS, and P = 0.002 in RFS), nodal metastasis (P < 0.001 in OS, and P < 0.001 in RFS), pathological stage (P < 0.001 in OS, and P < 0.001 in RFS), treatment modality (P = 0.032 in OS, and P = 0.059 in RFS) and PNI (P < 0.001 in OS, and P < 0.001 in RFS), were significant factors of both OS and RFS. In multivariable analysis, for OS, the pathological stage (Hazard ratio (HR) 1.432; 95% confidence interval (95% CI) 1.210–1.695; P < 0.001) and PNI (HR 0.812; 95% CI 0.761–0.865; P < 0.001) were independent prognostic factors. And for RFS, We found PNI as an independent prognostic factor (HR 0.792; 95% CI 0.739–0.848; P < 0.001), and the pathological stage (HR 1.373; 95% CI 1.160–1.625; P < 0.001). In the subgroup of patients with PNI ≥ 49.4, univariable analysis showed treatment modality was a significant factor of overall survival (P = 0.001); multivariable analysis showed patients received postoperative chemotherapy (HR 0.288; 95% CI 0.095–0.874; P = 0.028) or postoperative chemotherapy with targeted therapy (HR 0.148; 95% CI 0.030–0.726; P = 0.019) has better overall survival rates. Conclusion The PNI and the pathological TNM stage are independent prognostic factors for pulmonary sarcomatoid carcinoma. PNI is an important indicator for the selection of postoperative adjuvant therapy. Patients with PNI ≥ 49.4 may benefit from postoperative chemotherapy and targeted therapy. We still need further prospective studies to confirm these results.


Background
Pulmonary sarcomatoid carcinoma (PSC) is considered as a rare subtype of non-small cell lung cancers with very aggressive behavior [1]. The proportion of patients with pulmonary sarcomatoid carcinoma developed

Open Access
Clinical Sarcoma Research *Correspondence: 20504744@qq.com 1 Department of Thoracic Surgery, The Fourth Hospital of Hebei Medical University, No. 12, Jiankang Road, Shijiazhuang 050000, China Full list of author information is available at the end of the article recurrence, even after R0 surgery is appreciable [2]. Studies reported poor survival outcome in patients with early-stage PSCs [3]. Several case report showed that PSC is resistant to chemotherapy [1,[3][4][5]. Therefore, predicting the prognosis of PSC patients accurately is important to improve PSC patients' survival and to provide important information to the management of PSC patients.
The postoperative complications, and the long-term outcomes of patients with malignances have been considered to be associated significantly with preoperative nutritional condition and immunological status [6][7][8]. The prognostic nutritional index (PNI), calculated based on combining the serum albumin concentration with total peripheral lymphocyte count, was initially used to assess the immune-nutritional status before or after surgery and postoperational complications in patients underwent gastrointestinal surgery [7]. Recent study show that the PNI is a prognostic factor for various carcinomas [8][9][10]. PNI has not yet been investigated in PSC patients to our knowledge. Therefore, we studied the correlation between the PNI and clinical characteristics and the PNI's impact on the overall survival (OS) and recurrence-free survival (RFS) in PSC patients.

Patients
Medical records between January 2008 and December 2015 were reviewed for 8176 consecutive patients with resected lung cancer and lymph node dissection at the 4th hospital of Hebei Medical University. Patients diagnosed as pulmonary sarcomatoid carcinoma with R0 resection and complete clinic-pathological data were included for analysis. The laboratory results were obtained within 1 week before operation. Owing to the retrospective design, patient consent was waived. The study protocol was approved by the Ethics Committee of the 4th hospital of Hebei Medical University. The followup was conducted at clinic and by telephone call from designated personale in follow-up center in our hospital until October 31, 2019, or patient's death. We obtained the clinical characteristics of patients retrospectively from medical records and evaluated these characteristics as prognostic factors. These factors included the patient's age, sex, smoking and drinking habits, tumor size, tumor stage, tumor location, lymph node metastasis, therapeutic methods, and pathological stage. Preoperative data were obtained from the patients' medical records, including serum albumin and total lymphocyte count from peripheral blood tests. Then, the following formula was used to calculate the PNI: 10* serum albumin (g/ dl) + 0.005* total lymphocyte count (cells per mm 3 ) [7].

Statistical analysis
The categorical variables between groups were compared using the X 2 test. Continuous variables with normal distribution were compared using the t test. The median value of the follow-up was 51 months. The OS was considered as the time from the operation to death or last follow-up. The survival curves were generated by the Kaplan-Meier method. Differences among the curves were evaluated by log-rank test. The mean of PNI was used as the cutoff value since PNI value confirm to the normal distribution in the population. According to the cutoff value of the PNI, all patients were divided into a PNI-high group or a PNI-low group. Variables found significant in the univariable analysis were entered into a multivariable analysis (COX proportional hazards models). We also try to consider PNI as a continuous variable in the time-to-event models when studying the independent association of PNI with OS and RFS. All P values of < 0.05 were considered to be significant, and confidence intervals (CI) were calculated at the 95% level. The statistical analyses were performed using SPSS software (version 24.0; SPSS, Chicago, IL).

Data
None of these patients received preoperative chemotherapy. Postoperative adjuvant chemotherapy with an platinum-based regimen (pemetrexed + cisplatin) was accepted in a total of 30 patients. And 30 patients received postoperative adjuvant chemotherapy with targeted therapy. We used the 8th edition of the American Joint Committee on Cancer TNM classification system to classified the stage of PSC [11].  4; n = 52). We didn't find the distributions of TNM categories differ significantly between these two groups.

PNI and survival
To evaluate the importance of PNI in survival, we try to consider PNI as a continuous variable in the timeto-event models. We compared OS and RFS in patients

PNI and treatment therapy
The difference in OS rates was significant between the group of patients with surgery and the group of patients with postoperative chemotherapy (P = 0.004) but not with postoperative chemotherapy and targeted therapy (P = 0.185), neither between the group of patients with postoperative chemotherapy and the group of patients with postoperative chemotherapy and targeted therapy (P = 0.331) (Fig. 4).
The difference in RFS rates was significant between the group of patients with surgery and the group of patients with postoperative chemotherapy (P = 0.009) but not with postoperative chemotherapy and targeted therapy (P = 0.229), neither between the group of patients with postoperative chemotherapy and the group of patients with postoperative chemotherapy and targeted therapy (P = 0.295) (Fig. 5).
In the subgroup of patients with PNI ≥ 49.4, univariable analysis showed treatment modality was a significant factor of overall survival (P = 0.001) and recurrencefree survival (P = 0.005); multivariable analysis showed patients received postoperative chemotherapy (HR 0.288; 95% CI 0.095-0.874; P = 0.028) or postoperative chemotherapy with targeted therapy (HR 0.148; 95% CI 0.030-0.726; P = 0.019) has better overall survival rates; In the subgroup of patients with PNI < 49.4, we didn't find the same results. In the subgroup of patients with PNI ≥ 49.4, patients received postoperative chemotherapy with targeted therapy has better OS rates than those received postoperative chemotherapy (P = 0.048, X 2 = 3.924); In the subgroup of patients with PNI < 49.4, we didn't find the same results (P = 0.143, X 2 = 2.145) (Tables 3, 4) (Fig. 6a-d).

Discussion
This study offers the first evidence of the prognostic value of PNI in patients with completely resected pulmonary sarcomatoid carcinoma. Pulmonary sarcomatoid carcinoma (PSC) comprises less than 1% of lung cancers, and they respond poorly to systemic therapy [12]. Immunonutritional status plays an important role in postoperative outcomes. Therefore, we attempted to explore the PNI which was designed to represent immune-nutritional status as a new prognostic factor [8][9][10][11]. In our study, multivariable analysis showed that PNI and TNM In 2017, Lococo et al. in a multicenter study found that among their cohort, spindle cell carcinoma represented (29%), giant cell carcinoma (8%), pleomorphic carcinoma (62%), carcinosarcoma (1%), pulmonary blastoma (0%) [13]. In 2017, Rahouma et al. went through the the SEER database between 1973 and 2013, and revealed predominance of less aggressive histological subtypes in the 1993-2013 time period. They found spindle and giant cell carcinoma represented 72.9%, pleomorphic carcinoma 13.4%, carcinosarcoma 11.5% and pulmonary blastoma 2.2% [14]. Our findings in histologic subtype of PSCs seem to be inconsistent with these results of previous studies. In the present study, our results indicated that giant and spindle cell carcinoma represented the majority of the entire cohort. It might be reasonable that this rate was higher than those reported previously considering the race and region differences.
In our study, low PNI significantly correlated with advanced TNM stage, backing up the hypothesis that patients with a low PNI have a worse tumor progression. The advanced pathological stage may cause impaired immune-nutritional status. In this study, we didn't find low PNI was associated with smoking which indicated that in the low PNI group, smoking-related  inflammation's contribution was limited. Similar results can be found in previous studies regarding resectable NSCLC [15]. We also found that low PNI significantly correlated with treatment modality. It might be reasonable that the patients with better nutrition status and less systemic inflammation may be recommended to receive adjuvant therapy at the pre-gene therapy time.
The most commonly used PNI, was first reported in prediction of postoperative complications in gastrointestinal operation [6]. Many studies have reported that the PNI has prognostic value for various malignancies [7][8][9][10]. Qiu et al. reported that the PNI was an independent prognostic factor for patients who were received radical surgery with NSCLC [16]. Hong et al. proved that PNI could assist to identify small cell lung cancer patients with poor prognosis [17]. Shoji et al. indicated the PNI's value in predicting postoperative recurrence in patients with stage I NSCLC [18]. Similarly in this study, the hazard of the high-PNI group was lower than the one of the low-PNI group, which suggested that patients with lower PNI had reduced survival as a group, the PNI was an independent predicting factor in patients with resectable PSC. We also revealed that the incidences of recurrence and recurrence-related death were significantly higher in the low-PNI group. These results indicated that a low PNI may be strongly associated with disease-specific death, and lead to a worse outcome in patients with PSC. Vieira [19][20][21][22]. In 2017, Schrock et al. reported clinical outcomes for 10 PSC patients received targeted or immunotherapy, three had partial responses and three had stable disease [22]. These reports suggest that targeted therapies and immunotherapy might have encouraging outcomes for patients with PSCs. However, in this study, we found that the overall survival of patients who received adjuvant chemotherapy combined with targeted therapy was not significantly better than that of patients who simply received adjuvant chemotherapy. In this study, the adjuvant chemotherapy regimen was pemetrexed plus cisplatin for 4 cycles. After the emergence of targeted therapy, we prescribed gefitinib or icotinib orally for 2 years or until disease progression in patients with 19Del or L858R mutations along with adjuvant chemotherapy. According to the results of this study, patients with pulmonary sarcomatoid carcinoma who received adjuvant chemotherapy combined with targeted therapy do not have an improved prognosis compared with those who simply received adjuvant chemotherapy. Further stratified analysis revealed that the subgroup of patients with higher PNI (PNI ≥ 49.4) has better overall survival in the targeted therapy combined with adjuvant chemotherapy group than in the adjuvant chemotherapy group. The same advantage was not found in the subgroup of patients with PNI < 49.4. Therefore, we infer that the level of PNI in patients with pulmonary sarcomatoid carcinoma may determine whether they can benefit from targeted therapy after surgery. This study's result suggested that targeted therapy combined with adjuvant chemotherapy could be a suitable option for patients with lung sarcomatoid carcinoma with higher PNI after surgery. For patients with pulmonary sarcomatoid carcinoma with PNI < 49.4, postoperative adjuvant chemotherapy could be a suitable choice.
The limitations of this study are: it is an observational study. Although it has standardized entry criteria, and predefined endpoints for assessing survival, it is difficult to draw a very strong conclusion on the basis of our retrospective study, because of the heterogeneity of real-life patient populations and the small sample size from one center. And that might be the reason that our study is at variance with the results from other observational surveys. Hence, a large-sample, double-blinded, randomized prospective study with multicenter-participated is warranted to validate our results.

Conclusions
This study indicated that the PNI and the pathological stage system are strong predictors of OS and RFS for patients with PSC. Patients with low PNI have even worse prognosis in this population. PNI is an important indicator for the selection of postoperative adjuvant therapy. Patients with PNI ≥ 49.4 may benefit from postoperative chemotherapy and targeted therapy. We still need further prospective studies to confirm these results.