Lung cancer is the most common malignancy worldwide, with non-small cell lung cancer (NSCLC) accounting for 85% of all cases [1, 2]. Despite the steady decline in mortality and rapid increase in survival in recent years, specifically for NSCLC, lung cancer remains the most deadly cancer in both men and women, with a 5-year overall survival of 21% for all stages combined [1].

Oncogenic driver mutations are responsible for the initiation and development of NSCLC. Clinical efficacy of targeting oncogenic driver genes such as epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), c-ros oncogene 1 (ROS1), and B-Raf proto-oncogene (BRAF) has been confirmed in NSCLC treatment [3, 4, 5, 6]. So far, several treatment guidelines have been proposed for NSCLC patients with targetable driver mutations [7, 8]. However, the prognostic value of those mutations remains controversial. Therefore, a better understanding of the mutational landscape and the prognostic value of the mutations in NSCLC may provide valuable information for risk stratification and refining of the treatment strategies.

Next-generation sequencing (NGS) has been widely used to detect germline and somatic mutations in human cancers, showing advantages compared with conventional genomic sequencing methods. Previous studies have shown that some somatic mutations are strongly associated with the survival of NSCLC patients, providing valuable information for clinical management of NSCLC patients. For example, it was found that the combination of 6 hotspot mutations of EGFR, PIK3CA, and tumor suppressor protein p53 (TP53) predicts poor survival of NSCLC patients [9]. The presence of specific DDR gene alterations was associated with worse prognosis of advanced NSCLC [10]. In addition, studies found that the tumor mutation burden, i.e., the approximate amount of gene mutation that occurs in the genome of a cancer cell, can predict the response of NSCLC to immunotherapeutic agents [11]. Therefore, using NGS to detect somatic mutations and identify prognostic indicators may provide important information to guide treatment selection and improve outcomes for NSCLC management.

TP53 is the most frequently mutated gene in human cancer, with a frequency of 50–60% [12]. Although many studies have shown that TP53 mutations are associated with worse prognoses in cancer carriers, the prognostic value of TP53 mutation in NSCLC remains somewhat controversial [13, 14] due to different mutated exons and different cancer stages [15, 16, 17]. TP53 contains 11 exons, and most TP53 mutations occur in exons 5–8 spanning 540 nucleotides that encode the DNA-binding domain of p53 protein [13]. Previously, all p53 mutants have been considered functionally equal; however, increasing evidence has revealed that mutations occurring in different positions of a single gene, e.g., TP53, could have a different effect on the prognosis of cancer [18, 19, 20], thus emphasizing the importance of stratifying NSCLC patients according to the classification of detailed gene mutations.

In this study, we analyzed the effect of different firstline treatment strategies for advanced NSCLC patients. To factors that significantly associated with PFS of patients we carried out a detailed stratification of patients accoding to their TNM stages and genomic alterations. We then investigated the associations of high-frequency mutations with PFS of unresectable stage IV NSCLC patients receiving different first-line treatment regimens. Our results suggested that mutations in exon 5 of TP53 and low-density lipoprotein receptor-related protein 1B (LRP1B) were potential adverse prognostic indicators for stage IV NSCLC.

In the present study, we sought to identify somatic mutations associated with PFS in patients with advanced NSCLC after first-line treatment. Analysis of tumor tissue samples from 101 NSCLC patients in stages III and IV using NGS revealed that TP53 was the most frequently mutated gene, with exon 5 representing the most common mutated site. In patients with stage IV NSCLC, TP53 mutation in exon 5 was an independent predictor of poor PFS after the first-line treatment. Furthermore, mutations in TP53 exon 5 and LRP1B were associated with worse PFS after first-line chemotherapy and targeted therapy, respectively. These results suggested that TP53 exon 5 and LRP1B mutations are negative prognostic indicators for advanced NSCLC.

The NGS data may guide the treatment options to improve clinical outcomes. Cao et al. [24] have demonstrated that after identifying driver mutations by NGS, patients receiving matched targeted therapy achieve significantly longer PFS than those receiving non-targeted therapy. A recent meta-analysis, which included 11 studies and 2874 participants, has shown that ALK inhibitor targeted therapy causes a significant increase in PFS of patients with ALK-rearranged NSCLC compared with chemotherapy [25]. Similarly, in the present study, the first-line targeted therapy remarkably prolonged PFS of patients compared with first-line chemotherapy, regardless of the mutation status. In patients who harbored driver gene EGFR/ALK/ROS/BRAF mutations matched targeted therapy substantially prolonged the PFS of patients compared with non-targeted therapy. Thus, NGS data can indeed inform therapy decisions in patients with advanced NSCLC.

TP53 is the most frequently altered gene in NSCLC, occurring in 30–65% of cases [15, 26, 27]. The prevalence rate of TP53 mutation in this study was 68%. About 80–90% of TP53 mutations encode missense proteins with a reduced capacity to bind to a specific DNA sequence that regulates gene transcription [12, 28]. This study found that missense mutation was the most common type of TP53 mutation, found in 46% of patients. TP53 mutations frequently occur in the ‘hot-spot’ region of exons 5–8 encoding the DNA-binding domain [20, 29, 30, 31]. In the present study, mutations in exons 5–8 accounted for 65% of all TP53 mutations, and exon 5 was the most common mutated location. Vega et al. [32] explored the clinical significance of TP53 exon 5 mutation, demonstrating that patients with squamous cell lung tumors carrying TP53 mutations in exon 5 exhibit worse prognoses than those carrying mutations in other locations. Nonetheless, since then, many studies have focused on the prognostic value of TP53 exon 8 mutations, revealing that TP53 mutations affecting exon 8 but not exon 5 indicate poor prognosis in NSCLC patients [20, 26, 33]. In the current study, multivariate Cox regression analysis showed that TP53 exon 5 mutation was an independent predictor for worse PFS in patients with stage IV NSCLC after the first-line treatment. Moreover, TP53 mutation in exon 5 was correlated with poor PFS after first-line treatment regardless of the treatment option, suggesting that mutation in exon 5 might identify a subgroup of NSCLC patients with unfavorable prognoses.

With comprehensive genomic profiling, concurrent genetic alterations have been discovered in human cancers. The presence of TP53 co-mutations can identify high-risk ALK${}^{\mathrm{+}}$-NSCLC cases with earlier treatment failure and shorter survival, regardless of the therapy [34]. EGFR mutations are considered mutually exclusive with other oncogenic drivers [35, 36]. In the present study, we found that KRAS/EGFR were mutually exclusive mutations, whereas the other 12 gene pairs were co-mutations in stage IV NSCLC, including TP53/CDKN2A. CDKN2A (cyclin-dependent kinase inhibitor 2A) encodes tumor suppressors p16${}^{\text{INK4a}}$ and p19${}^{\text{ARF}}$. Mutations in TP53 and inactivation of CDKN2A are commonly observed and highly involved in lung squamous cell carcinoma that accounts for 20–30% of NSCLCs [30]. A recent meta-analysis has demonstrated that concurrent TP53 mutation is a negative prognostic factor that is correlated with poor outcomes in patients with EGFR- or ALK-targeted therapy in advanced NSCLC [37]. TP53 mutation is also involved in primary resistance to EGFR-targeted therapy in patients with advanced NSCLC [37]. These reports suggest that TP53 may play an important role in resistance mechanisms against targeted therapy. In this study, we observed co-mutations of TP53 with EGFR and ALK in some cases, despite the lack of statistical significance, which could be due to the small sample size.

LRP1B encodes low-density lipoprotein receptor-related protein 1b, which is frequently mutated in NSCLC and acts as a putative tumor suppressor [38, 39]. In this study, we observed LRP1B mutation in 9% of patients with advanced NSCLC, which was remarkably lower than that observed in other NSCLC cohorts, possibly because the samples being investigated here only include stage III&IV NSCLC patients [40, 41]. Accumulating evidence has suggested that LRP1B alterations are correlated with high herterogeineity of tumors which may lead to poor prognosis [40]. Multiple studies have shown that LRP1B mutation improved outcomes of immunotherapy in patients with different types of cancer, including NSCLC [40, 42, 43, 44, 45, 46, 47]. One possible mechanism is that NSCLC patients with somatic mutation of LRP1B usually have higer tumor mutation burden which postitively associated with immunotherapy [40]. We found that patients with LRP1B mutation had significantly shorter PFS compared to those with wild-type LRP1B after first-line targeted therapy, which suggests that NSCLC patients with LRP1B mutation may benefit from immunotherapy rather than targeted therapy.

The present study has some limitations that need to be addressed in the future. First, 7 patients who received first-line immunotherapy were excluded from the correlation analysis between TP53 mutation and PFS due to the small sample size. To investigate the prognostic values of immunotherapy-related genes such as LRP1B, larger sample size is needed. Second, the predictive potential of TP53 exon 5 mutation in the prognosis of NSCLC patients should be tested in large validation cohorts. Long-term follow-up data are also needed to investigate whether TP53 exon 5 mutations correlate with the overall survival of the patients. Third, although many patients have received second and third-line therapies, numerous confounding factors such as loss of follow-up prevented us from data collection and analysis.

We used NGS to characterize the mutational landscape in tumor samples from patients with stages III or IV NSCLC. Mutations in TP53 exon 5 and LRP1B resulted in being negative prognostic factors for patients with stage IV NSCLC after first-line chemotherapy and targeted therapy, respectively. According to these findings, immunotherapy or immunochemotherapy are more recommended treatment stratigies for advanced NSCLC patients with somatic mutations in TP53 exon 5 and LRP1B. These findings reinforce the importance of genetic testing to guide first-line therapy for NSCLC management.

ALK, anaplastic lymphoma kinase; BRAF, B-Raf proto-oncogene; EGFR, epidermal growth factor receptor; LRP1B, low-density lipoprotein receptor-related protein 1B; NGS, next-generation sequencing; NSCLC, non-small cell lung cancer; PFS, progression-free survival; ROS1, c-ros oncogene 1; TP53, tumor suppressor protein p53.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

XR, JZ and FL contributed to the conception, design, Supervision, and guidance of the study; HF and HX, performed the data collection and analyses and wrote the manuscript; GD, CY and LL assisted in project design, guided the statistical analysis, and reviewed the manuscript; ZJ assisted in bioinformatic analysis; XY and HG contributed to data collection and collation; XS helped perform the analysis with constructive discussions. XR, JZ and FL critically reviewed the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

This study was approved by the Ethics Committee of Shanxi Bethune Hospital (No. YXLL-2022-045) and conducted according to the Declaration of Helsinki. As this was a retrospective study, the requirement for informed consent was waived.

This work was supported by Cancer Biobank of Tianjin Medical University Cancer Institute and Hospital.

This work was supported by grants from the Health Commission of Shanxi Provincial [grant number 2019014] and the CAPTRA-Lung Scientific research fund of Beijing Cancer Prevention and treatment Society [grant number CAPTRALung2022010].

LL and ZJ are both Medical Advisors of Genetron Health Inc and they declare no conflicts of interest. Other authors declare no potential conflicts of interest.