The Pain Prevalence, Characteristics and Related Factors of Non-Small Cell Lung Cancer (NSCLC) Referral to Palliative Care Service


 Background：This study is to analyze the occurrence and clinical characteristics of cancer pain in advanced non-small cell lung cancer (NSCLC) patients who referral to palliative care treatment, as well as correlation between pain performance and clinical characteristics, laboratory indicators, and previous treatment.Methods ：Advanced NSCLC patients referred to the Palliative Care Unit (PCU) from January of 2016 to December of 2018 were included in our study. The Brief Pain Inventory (BPI) was used to evaluate cancer pain (CP). Results: Of 54 (56.7%) patients reported with pain, there were 23 patients (42.6%) who suffered from severe pain (score >=7) at the admission. There was a significant difference in the frequency of pain between patients with and without bone metastasis (p= 0.006), between patients with more than two sites and non-more than two sites metastasis (p=0.032). Patients received epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI) had lower pain prevalence (p=0.033). Patients with bone metastases (p=0.005), with more than two metastases (p=0.022) and those with higer C-reactive protein (CRP) index (p= 0.012) had higher mean scores of pain, with significant differences. Conclusions:Our study showed the general characteristics, pain performance and related factors of advanced NSCLC patients who referral to palliative care service. This study confirmed that these patents’ clinical manifestations of pain were correlated with bone metastasis, metastasis with more than two sites, elevated CRP, and the EGFR-TKI usage. Whether there is a certain potential connection between these factors deserves further study.

For advanced lung cancer patients, especially non-small cell lung cancer (NSCLC) patients, gene detection is very important. For patients with gene mutation, targeted therapy is the rst-line treatment recommended by the current guidelines [6]. Since epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI) has been rst-line recommended in NSCLC patients with epidermal growth factor receptor (EGFR) mutation and has been widely used, patients' baseline situation had been affected who referal to palliative care. Clinical literatures indicated that EGFR-TKI may have analgesic properties [7].
EGFR-TKI were testi ed to attenuate neuropathic pain in humans in previous research [8]. The effect of EGFR mutation and the EGFR -TKI on cancer pain in NSCLC are worth study.
A large number of literatures reported the current situation of CP in advanced cancer patients, but few studies have focused CP characteristics of speci c single disease patients, including common cancers such as NSCLC [9]. The objective of this study is to determine the ouucrance and clinical characteristics of CP in NSCLC patients who referral to palliative care treatment, as well as laboratory indicators, previous treatments and other related factors.

Samples and procedures
Advanced NSCLC patients referred to the PCU from January of 2016 to December of 2018 were included in our study. Patients with incomplete general information were excluded from our study. Patients who were survived until the data collection were analyzed as censor. The data of 95 patients with de nite pathology were nally included in the analysis. We followed the ethical guidelines of the FUSCC review board. As all the data assessed in the study were obtained as part of routine clinical assessments from patients' medical records, written consent was not obtained, in accordance with the guidelines of the Chinese Ministry of Health.

Data Collection
Four experienced palliative care attending physicians who received training participated in the retrospective collection of patients' medical information through electronic chart review. Electronic medical chart is recorded in the Union Medical System (UMS) of FUSCC, which is an electronic medical record system that contains all medical information of patients in FUSCC, including demographic and clinical data. The following variables were extracted into SPSS software: (1) general baseline demographic characteristics-including age, sex, indigenous (Shanghainese) or not, medical insurance status, and area of residence (rural or urban according to the address of patients' permanent residence); (2) disease-related characteristics-including cancer diagnosis, pathology, gene detection report, metastatic sites and previous treatment.
Pain and symptom assessment CP was de ned as pain directly because of the tumor presence or secondary to anti-enoplastic treatment.
The Brief Pain Inventory (BPI) was used to evaluate CP [10].. The BPI scale is an assessment tool for cancer related pain which has been translated into Chinese. Pain intensity was measured with the Numeric Scoring Scale (NRS), which ranged from 0 to 10 (0 = no pain, 10 = worst possible pain). For analysis, pain intensity was categorized into no pain (NRS 0) mild pain (NRS 1-3) and moderate(NRS 4-6)to severe pain (NRS 7-10). Pain medication included both classic analgesics, such as opioids, paracetamol and NSAIDs; and co-analgesics, such as steroids, tri-cyclic antidepressants and antiepileptics. The BPI survey was assessed at the admission of patients.

Statistical Analysis
Patients characteristics and disease features were analyzed by descriptive statistics such as proportions,means,standard deviations, medians, and mode. 95% con dence intervals (95% CI) were calculated for median length. SPSS software (v.16.0) was employed for statistical analysis. p value less than 0.05 was considered to be signi cant.
62.1% (n = 59) patients received no anti-cancer treatments since their disease diagnoses. Details were shown in Table 1.

Relationship between pain characteristics and main baseline characteristics
There was a signi cant difference in the frequency of pain between patients with and without bone metastasis (p = 0.006), between patients with more than two sites and non-more than two sites metastasis (p = 0.032). Patients received EGFR-Tyrosine kinase inhibitors (EGFR-TKI) had lower pain prevalence (p = 0.033). For pain intensity, Patients with bone metastases (p = 0.005) and those with more than two metastases (p = 0.022) had higher mean scores of pain, with signi cant differences. For CRP index with 11 as the boundary, there was a signi cant difference in the mean score of pain intensity (p = 0.012). Although there was no statistical difference in the results, patients who had used EGFR-TKI had a lower mean pain score. Details were shown in Table 3.  Our study showed the general characteristics, pain and related factors of patients who referraled to Palliative care service of advanced NSCLC patients. It is a relatively rare description on the characteristics of pain caused by a certain cancer. In addition to re ecting patient's pain features, we try to nd some indicators and factors that can re ect patient's pain frequency and its' severity in clinical data such as in ammatory markers, related symptom severity. Our study con rmed that the clinical manifestations of pain in NSCLC patients were correlated with bone metastasis, metastasis with more than two sites, elevated CRP, and the EGFR-TKI usage. Whether there is a certain potential connection between these factors deserves to be further studied.
Of all 95 included patients, 54 patients (56.7%) reported cancer pain suffering. Of 54 patients with pain, there were 23 patients (42.6%) who suffered from severe pain (score > = 7) at the admission of PC. The incidence of pain in lung cancer patients was reported quite differently. It may be due to different baseline conditions of patients, different research background, etc. Prevalence of pain in the early stages of lung cancer has been reported to be as high as 56% [11]. Over 90% of patients with advanced lung cancer experienced pain and fatigue in another research [12]. In another study of patients with NSCLC, pain was reported to be only 37% [13]. These studies were quite heterogeneous with differences in sample sizes, cancer stage and clinical variables, making it di cult to ascertain the exact prevalence of pain among advanced lung cancer patients.
In our patients' baseline, at the time of admission to PC, 60% of patients had been con rmed combined with bone metastasis. Approximately 40% of NSCLC patients developed bone metastases in their disease course [14]. Previous studies have testi ed that bone metastases strongly affected presence of cancer pain [15]. Bone pain was thought to be related with stimulation of sensory bers distributed in the periosteum. Osteodynia was also known as related with prostaglandins, endothelins, nerve growth factors or in ammatory cells, and maybe the activation of several acid sensing receptors by local tissue acidosis [16,17]. The presence of bone metastasis also affects the pain distribution of NSCLC patients. Bone metastases sites of lung cancer are most common in the lumbar spine, pelvis, thoracic spine and ribs. In our research, the most frequent pain location was chest (n = 22), followed by dorsum (n = 11). The incidence of dorsum is high in patients, which may be related to the proportion of bone metastasis of our patients.
With the development of targeted therapy, more patients have done the corresponding gene detection before receiving systematic treatment. There was no signi cant correlation between EGFR mutation status and pain characteristics in our research but patients used EGFR-TKI reported lower pain intensity with statistical differences. Previous research also showed that EGFR-TKI strongly reduced nocifensive behavior in mouse models of in ammatory and chronic pain. The PI3K/AKT/mTOR pathway and upregulated mitogen-activated protein kinase (MAPK) expression maybe involved in enhancement of nociception by activation of EGFR [18]. No EGFR-TKI usage, together with smoking, histological subtype and performance status were shown as associated with bone metastasis' skeletal-related events (SRS) [19]. The systemic effect and mechanism of EGFR mutation and EGFR-TKI usage, their relationship with bone metastasis and pain characteristics are worth of further study.
C reactive protein (CRP), together with in ammation markers were testi ed to be related with pain manifestaion in previous research [20,21]. Among breast cancer patients, women with pain detected higher CRP levels before anticancer treatment than women without pain [22]. It was reported that systemic in ammation maybe linked CP, but the relationship between pain and systemic in ammation in cancer was less well understood [23,24 ]. The current ndings suggest a potential association between pain and in ammatory processes in NSCLC patients, with potential attention for future research and treatment strategies.

Limitation
Due to the limited number of cases, the speci c EGFR mutation sites of patients, the diversity of different generation of EGFR-TKI drugs used, and other gene mutations of NSCLC patients were not included in analysis. The further relationship between speci c drugs, cycles, and pain performance in patients using EGFR-TKI will be worth further study.