Background
Lung cancer is a prevalent disease that seriously endangers global public health [
1‐
4]. According to statistics, there were about 2.20 million newly-diagnosed lung cancer cases and 1.79 million mortalities worldwide every year [
4,
5]. Lung cancer accounts for more than 20% of cancer-related deaths worldwide, surpassing the combined mortality rates of prostate, breast, and colon cancers [
1,
6‐
8]. Despite the progress made in targeted therapy and immunotherapy in the recent decades, platinum-based chemotherapy remains the most widely used treatment option in clinical practice [
9‐
12]. However, due to individual variations in sensitivity, only a subset of patients benefits from this treatment [
13]. Given the potential toxic reactions, it is urgent to discover reliable predictive biomarkers to predict the prognosis, therapeutic efficacy and toxicity of lung cancer patients, which is crucial for promoting personalized medicine and enhancing therapeutic outcomes [
14‐
16].
Cyclins D1 (
CCND1) plays a vital role in cell cycle regulation which mediates the G1 to S phase transition [
17‐
19]. It also has a fundamental involvement in human cancer progression, including cell proliferation, transcription, chromosome duplication and stability, DNA damage response, metabolism, tumor migration and invasion [
17,
20,
21]. Multiple clinical studies demonstrated that dysregulation of
CCND1 is associated with poor prognosis and platinum-based chemotherapy response in various human cancers, highlighting its potential as a tumor predictive biomarker [
22‐
32].
Single nucleotide polymorphisms (SNPs) refer to DNA sequence polymorphisms caused by single nucleotide variation at the genomic level, accounting for over 90% of all known polymorphisms [
33‐
35]. Cyclins D1 is the second most frequently amplified locus in human solid tumors [
36,
37]. The association between
CCND1 A870G (rs9344) polymorphism and cancer risk has been previously investigated in lung cancer [
38‐
43]. However, due to the limited number of studies and sample size, the exact role of
CCND1 polymorphism in predicting lung cancer risk remains unclear. Only few studies have been conducted to investigate the correlation between
CCND1 rs9344 and platinum-based chemotherapy response in lung cancer.
This study aimed to investigate the association of CCND1 rs9344 with cancer susceptibility, platinum-based chemotherapy, toxicity and overall survival of patients with lung cancer by performing hospital-based case-control study. Additionally, a meta-analysis was conducted using 5432 cases and 6452 control samples to evaluate the association between CCND1 rs9344 polymorphism and lung cancer risk. The results may provide evidence in support of the potential utilization of CCND1 rs9344 as a predictive biomarker for prognosis and chemotherapy sensitivity in Chinese patients with lung cancer in certain conditions.
Methods
Study design
Setting
During November 2011 to May 2013, 498 patients with primary lung cancer (diagnosed by cytology or histology) were consecutively recruited at Xiangya Hospital and the Affiliated Cancer Hospital of Central South University in Changsha, Hunan Province, China. During the same period, 213 healthy controls were collected from the physical examination center of Xiangya Hospital of Central South University. This study was approved by the Ethics Committee of Xiangya School of Medicine, Central South University (registration number: CTXY-110008-2), and all subjects enrolled have signed the informed consent.
Participants
All patients had been histologically or cytologically confirmed to have primary lung cancer. Subjects who were pregnant, lactating, had active infections, symptomatic brain or leptomeningeal metastases, or other previous or concurrent malignancies were excluded from the study. Among them, 467 patients were enrolled in the platinum-based chemotherapy response study. The inclusion criteria were as follows: (1) They were not administered radiotherapy and/or biological therapy prior to or during chemotherapy; (2) they received at least two cycles of platinum-based chemotherapy; (3) they underwent full follow-up (to March 2017); (4) tumors were assessed before and during treatment using the same imaging methods (Supplementary Table
1). Platinum-based chemotherapy regimens include pemetrexed + platinum (PP), gemcitabine + platinum (GP), paclitaxel + platinum (TP), docetaxel + platinum (DP), etoposide + platinum (EP), and other platinum-based chemotherapy regimens (irinotecan + platinum, navibine + platinum). In the case of the healthy controls, individuals with a smoking history, a history of lung ailments, or those engaged in high-risk occupations such as chemical, construction, asbestos, and coal mining work were excluded.
Variables
The endpoints of the study were as follows: chemotherapy response was evaluated based on the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines and categorized as responders (complete response: CR, partial response: PR) or non-responders (stable disease: SD and progressive disease: PD). Two professional radiologists independently evaluated the CT scans of lung cancer patients before and after chemotherapy to assess the treatment effectiveness after two cycles of therapy. In case of disagreement, a third radiologist was consulted. Toxicity was assessed according to the National Cancer Institute Common Toxicity Criteria 3.0 during the first two cycles of chemotherapy regimen. Grade 3 or 4 toxicity was defined as severe toxicity. Severe gastrointestinal toxicity was grade 3 or 4 nausea and vomiting. Severe hematological toxicity included grade 3 or 4 hypochromia, leukopenia, neutropenia and thrombocytopenia. Patients who experienced any type of the grade 3 or 4 toxicities described above were defined as suffering severe overall toxicity.
For the lung caner patients, age, sex, smoking status, stage, histological type, and chemotherapy regimens were collected. For the healthy controls, age, sex and smoking status were collected. The above factors age, sex, smoking status, stage, histological type, and chemotherapy regimens were considered as covaraites in this study.
DNA extraction and genotyping analysis
Venous blood DNA was extracted using the Genomic DNA Purification Kit (Promega, Madison, WI, USA). CCND1 rs9344 was genotyped using the Sequenom MassARRAY System (Sequenom, San Diego, CA, USA).
Study selection and data extraction criteria of meta-analysis
The Pubmed, Embase and Cochrane databases were utilized to identify original studies examing the association between CCND1 rs9344 and lung cancer susceptibility (up to March 29, 2023). The search formula was: “CCND1 or Cyclin D1” and “genetic polymorphism or polymorphisms or variant or rs9344” and “lung cancer”. Included studies had to be original case-control studies with detailed CCND1 rs9344 genotype frequencies or available data. The qualities of selected studies were independently assessed and identified by two researchers. The following information was extracted from the included studies: the last name of the first author, year of publication, country, ethnicity, cancer type, source of cases and controls, number of cases and controls, genotyping method, genotype or allele frequency, and HWE p values for controls.
Statistical analysis
The study size was estimated using Power Analysis and Sample Size (PASS) 2021 (NCSS, LLC. Kaysville, Utah, USA) at a power value of 0.80. The chi-square test was used to assess differences in proportions between groups for the categorical variables. The median age of lung cancer patients, 57 years old, was used as cut-off value. The Hardy-Weinberg equilibrium was calculated using the chi-square test. Associations between CCND1 rs9344 and cancer susceptibility, therapeutic response and toxicity were estimated by unconditional logistic regression. Factors including age, sex, smoking status, stage, histological type, and chemotherapy regimens were considered as covaraites in this study. Survival curves were calculated using the Kaplan-Meier method, and survival analyses were conducted using Cox proportional hazards regression analysis. All significance tests were two-sided, and P < 0.05 was defined as statistically significant. The above analyses were performed using PLINK 1.9 and PASW statistics v18.0 (IBM Co., Armonk, NY, USA).
In the meta-analysis, the association between cancer risk and CCND1 rs9344 was assessed by calculating pooled OR and 95% CI. The heterogeneity of the effect size across studies was estimated and quantified by Cochrane’s Q test and I2 test. The random effect model is selected if P < 0.1 or I2 > 50%, otherwise, the fixed effect model is adopted. The stability of the results was assessed by sensitivity analysis. The inverted funnel plot was used to estimate the publication bias. All statistical analysis was performed in R4.2.3. P < 0.05 was considered statistically significant.
Discussion
Lung cancer remains one of the leading disease burdens. While the last two decades have witnessed the emergence of novel therapeutic approaches such as targeted therapy and immunotherapy, platinum-based chemotherapy remains the most widely employed treatment for lung cancer patients. However, only a subset of patients could benefit from platinum-based chemotherapy, while the others, who prove insensitive to platinum drugs, endure the burdens of toxic side effects without any associated improvement in survival outcomes. Deeper insight into the pathogenesis, discovery of predictive biomarkers and optimization in therapeutic methods may efficiently improve the treatment outcome [
48‐
50]. Based on this, one of the issues that urgently need to be addressed now discovering reliable biomarkers to identify individuals with a higher sensitivity to platinum-based chemotherapy. This expansion may provide promising possibilities for lung cancer diagnosis, treatment and prevention.
Unbalanced cycle regulation is one of the hallmarks of carcinogenesis. Cyclin D1 plays a crucial role in the transition from the G1 to the S phase of the cell cycle, thus being widely recognized as a pivotal element during the malignant transformation process [
51]. The rs9344 (A870G), located in exon 4 of
CCND1 gene, is a frequent gene polymorphism that regulates alternative splicing and enables the expression of the transcribed Cyclin D1b. The prediction value of
CCND1 rs9344 in the prognosis of lung cancer patients has been investigated in several previous studies. However, few of them concentrated on platinum-based chemotherapy response. Hsia, et al. reported that among the lung cancer patients and cancer-free healthy controls, genotype distribution (
P = 0.0003) and allelic frequency (
P = 0.0007) of
CCND1 rs9344 were significantly different. Individuals who carried the AG and GG genotypes had a 0.59- and 0.52-fold risk of lung cancer compared to the AA genotype, respectively (95% CI, 0.44–0.78 and 0.35–0.79) [
40]. Sobti et al. also indicated that the AG genotype was correlated with a higher risk of lung cancer (OR = 1.7, 95% CI = 0.92–3.14) [
46]. Gautschi, et al. found that
CCND1 GG genotype was significantly correlated with platinum-based chemotherapy response (
P = 0.04), while no significant difference was identified in patients’ prognosis among different genotypes [
41]. However, Cakina, et al. indicated that no correlation was found in
CCND1 A870G polymorphism between lung cancer patients and controls [
43].
This study conducted a hospital-based case-control investigation focusing on lung cancer, and systematically investigated the association between CCND1 rs9344 and lung cancer susceptibility, platinum-based chemotherapy sensitivity, toxicity, and overall survival. While no significant differences were observed in the general population, the predictive potential of CCND1 rs9344 was established within specific patient subgroups. For cancer susceptibility, patients with the AA genotype exhibited a significantly higher risk than those with the GG + GA genotype (recessive model, adjusted OR = 1.755, 95%CI = 1.057–2.912, P = 0.030). In the context of platinum-based chemotherapy, CCND1 rs9344 showed significant correlations with therapy response in patients receiving the PP regimen (additive model: adjusted OR = 1.926, 95%CI = 1.029–3.605, P = 0.040; recessive model: adjusted OR = 11.340, 95%CI = 1.428–90.100, P = 0.022). This significant association was also observed among ADC patients (recessive model: adjusted OR = 3.345, 95%CI = 1.276–8.765, P = 0.014). Furthermore, an increased risk of overall toxicity was found in both NSCLC (additive model: adjusted OR = 1.395, 95%CI = 1.025–1.897, P = 0.034; recessive model: adjusted OR = 1.852, 95%CI = 1.088–3.152, P = 0.023) and ADC patients (additive model: adjusted OR = 1.547, 95%CI = 1.015–2.359, P = 0.043; recessive model: adjusted OR = 2.030, 95%CI = 1.017–4.052, P = 0.045). Notably, in non-smokers, CCND1 rs9344 was significantly associated with a higher risk of gastrointestinal toxicity (adjusted OR = 2.620, 95%CI = 1.083–6.336, P = 0.035).
In addition to the case-control study, a comprehensive meta-analysis for previous research on CCND1 rs9344 and lung cancer susceptibility was conducted. In line with our findings, no significant correlation was observed on a overall scale. This may arise from various factors such as variations in sample selection and distribution, disparities in research quality, substantial heterogeneity in environmental factors, or gene-environment interactions. The results of our study and meta-analysis consistently suggest that the predictive role of CCND1 rs9344 in therapeutic efficacy and prognosis of lung cancer patients may not be effective for all individuals, but rather requires more precise subgroup analysis. Besides, the lack of statistical significance at the overall level may also be caused by various factors in different studies, including differences in sample selection and distribution, variations in study quality, substantial heterogeneity of environmental factors, or gene-environment interactions. The predictive value of CCND1 rs9344 remains to be further validated in large samples through stratified analysis.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.