Clinical and molecular characteristics, transcriptome, and drug efficacy of ERBB2 tyrosine kinase and non-tyrosine kinase mutations in non-small cell lung cancer

doi:10.21203/rs.3.rs-3149860/v1

Download PDF

Article

Clinical and molecular characteristics, transcriptome, and drug efficacy of ERBB2 tyrosine kinase and non-tyrosine kinase mutations in non-small cell lung cancer

https://doi.org/10.21203/rs.3.rs-3149860/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

ERBB2 mutations in the tyrosine kinase domain (TKD) have been widely reported in non-small cell lung cancer (NSCLC). More and more non-tyrosine kinase domain (non-TKD) mutations of ERBB2 have been detected. However, the clinical effects of non-TKD mutations are still unknown. Therefore, this study aims to study the molecular and clinical characteristics, transcriptome differences, and sensitive drugs of TKD and non-TKD mutations in NSCLC. Gene mutation, RNA sequencing, and clinical information of NSCLC with ERBB2 mutations were downloaded. Bioinformatics methods, such as gene mutation analysis, differentially expressed genes (DEGs), functional enrichment analysis, protein-protein interaction (PPI), hub gene identification, and drug sensitivity analysis, were adopted. Finally, four patients were included to reflect the treatment response. The somatic mutation rate of ERBB2 was 3.2%. TKD and non-TKD mutations mainly occurred in lung adenocarcinoma. Non-TKD mutations have a better prognosis. Up-regulated DEGs are primarily involved in immune and inflammatory pathways. We then proved that BTK, LYN, and PIK3CA mutations have a better prognosis than wild type in patients of NSCLC. The drug sensitivity study found that the TKD group was more sensitive to 5 drugs, and the non-TKD was 28.TKD and non-TKD mutations identify two independent subsets of ERBB2 mutations in NSCLC. Non-TKD mutations have a better prognosis and can also be used as targets for ERBB2. Our study can provide a foundation for further clinical research, highlighting the importance of individualized treatment for patients with different mutation domains.

Biological sciences/Cancer

Health sciences/Oncology

Health sciences/Pathogenesis

clinicopathologic feature

non-small cell lung cancer

non-tyrosine kinase mutation

prognosis

tyrosine kinase mutations

ERBB2, also known as HER2, is a member of the ErbB protein family, which includes epidermal growth factor receptors (EGFR or HER1/ErbB1), ErbB2, EGFR3 (or HER3/ErbB3), and EGFR4 (or HER4/ErbB4). ERBB2 receptor dimerization activates downstream cascades, mainly including phosphatidyl inositol kinase/protein kinase B (AKT) and mitogen-activated protein kinase (MAPK) signaling pathways, which are essential for cell proliferation, differentiation, and migration [1]. The ERBB2 gene is a proto-oncogene located at 17q11.2-q12 that encodes a transmembrane glycoprotein with intrinsic tyrosine kinase activity.

ERBB2 amplification or overexpression is a biomarker of anti-Erbb2 targeted therapy in breast cancer. Trastuzumab or trastuzumab plus pertuzumab lays the foundation for treating breast cancer patients with ERBB2 amplification or overexpression. With the discovery of oncogenic drivers and tyrosine kinase inhibitors targeting these drivers, advanced non-small cell lung cancer treatment strategy has shifted from a pathological to a molecular one. These advances have led to new decisions in the clinical management of non-small cell lung cancer, including detecting HER2 changes. HER2 changes mainly include HER2 mutation, HER2 amplification, and HER2 overexpression, and the incidence in NSCLC is 1%-6.7%, 2%-22%, and 7.7%-23%, respectively, all of which are associated with poor prognosis [2]. Although the proportion of HER2 amplification and HER2 overexpression in NSCLC is not low, clinical trials of trastuzumab combined with chemotherapy in NSCLC have shown disappointing results [3].

Therefore, in lung cancer, targeted therapy for ERBB2 mutation will dominate in the future. The ERBB2 mutant domain consists of an Extracellular domain (ECD), Transmembrane domain (TMD), Juxtamembrane domain (JMD), Tyrosine kinase domain (TKD), and intracellular domain of a carboxyl-terminal tail. ECD comprises the receptor-L domain, furin‐like cysteine-enrichment domain, and growth factor receptor domain IV. All other mutant domains except TKD are classified as non-TKD [4]. ERBB2 mutations in the TKD, especially exon 20 insertion mutations, have been widely reported in NSCLC. Due to the popularity of next-generation sequencing, more and more non-TKD mutations have been detected. However, the clinical effect of non-TKD mutations remains unknown [5]. In non-TKD preclinical studies, like V659E and G660D, mutations in TMD can also be used as driving mutations in NSCLC [6]. In case reports, patients carrying the ERBB2 V659E mutation showed sensitivity to afatinib [7]. Therefore, non-TKD mutations may also be therapeutic targets for ERBB2 therapies. Moreover, studies have found that patients with different ERBB2 mutation domains in NSCLC have different therapeutic effects. In a study assessing Ado-Trastuzumab Emtansine's efficacy for Patients With HER2-Mutant Lung Cancers, patients with the ERBB2 exon 20 insertion mutation had an ORR of 54.5%, but with the ERBB2 exon 19 mutation did not respond [8]. Therefore, this study aimed to investigate the molecular and clinical characteristics, transcriptome differences, and drug sensitivity analysis of TKD and non-TKD mutations.

There is increasing evidence that TKD and non-TKD mutations may be independent subgroups. Hence, exploring differences in molecular characteristics, clinical information, transcriptome, prognosis, and drug sensitivity between the two groups is essential. Here, the difference in gene mutation, clinical data, gene expression, and pathways between TKD and non-TKD were studied. Our study helps to understand the difference between TKD and non-TKD mutations in NSCLC and highlights the importance of targeted treatment for patients with different mutation domains.

Study cohort

Through the public database cBioPortal for Cancer Genomics (https://www.cbioportal.org/), downloading the detailed data of NSCLC samples with ERBB2 mutations from the Cancer genome atlas, identifying 8854 patients, among which 366 have ERBB2 mutation, with a somatic mutation rate of 3.2%. After removing duplication, two hundred forty-four patients with clinical information are adopted, and 116 samples are used for analyzing the mutation landscape. Maftools, an R Bioconductor package, are commonly used in genomic cancer studies. The dataset contained RNAseq data of 30 NSCLC patients with ERBB2 mutations, including 13 TKD mutations and 17 with non-TKD. Finally, four patients from Zhujiang Hospital of Southern Medical University were included to reflect the treatment response.

Mutation analyses

This study adopts next-generation sequencing, the mutation domain defined as the region where the ERBB2 mutation occurs. The mutant domains refer to the Pfam database (http://pfam.xfam.org/), including receptor-L domains (amino acid locations: 52–173 and 366–486), furin‐like cysteine-enrichment domains (183–343), growth factor receptor domains IV (510–643), transmembrane domains (654–675), and tyrosine kinase domains (720–976). The non-TKD is defined as the above ERBB2 domain in addition to TKD. The oncogenic function of ERBB2 mutations refers to the OncoKB (https://oncokb.org/) precise oncology database, which contains information on the biological effects and therapy. For those who can not be found in OncoKB, we use the polyphen2 database (http://genetics.bwh.harvard), including missense mutation and splice site mutation.

Clinical characteristics

We gather the patient with the ERBB2 mutation, downloading relevant clinical information from the cBioPortal for Cancer Genomics database, which includes pathological type, mutation domain, smoking status, smoking index, TMB, MSI score, PD-L1 score, tumor stage, tumor size, gender, age at diagnosis and related survival information, like Progression-free survival (PFS) and Overall survival (OS). PFS is the time between the start of a patient's treatment and the onset of disease progression or death from any cause. OS is the time from the initial diagnosis to death or last visit. Survival analysis was conducted for TKD and non-TKD patients in the ERBB2 mutation domain in NSCLC patients.

The identification of differentially expressed genes and Enrichment Analysis.

We were using R to normalize and screen DEGs. P < 0.01 in DEGs was considered significant. Six hundred thirty-eight up-regulated DEG genes were analyzed to enrich function and pathway. GO analysis divided DEGs into three categories: biological process (BP), molecular function (MF), and cell component (CC). GO analysis and KEGG enrichment analysis were performed using the clusterProfiler package with a p-value cutoff set to 0.05.

Construction of the PPI Network and Identification of Hub Genes

PPI network construction and Hub gene identification: using gene interactions online retrieval tools (STRING, https://www.string-db.org/) constructing the PPI network of DEG, whose PPI score threshold is 0.7 or higher. We identified the hub gene (top 10 genes) using the Cytohub plugin in Cytoscape (version 3.8.2, https://cytoscape.org/).

Drug sensitivity analysis

OncoPredict is an R package that predicts drug response [9]. The oncoPredict in R is used to predict drug sensitivity. This package is used to study the sensitivity of commonly used chemotherapy and molecule-targeted drugs in TKD and non-TKD. Tumor drug sensitivity genomics (GDSC https://www.cancerrxgene.org/) and the Broad Institute Cancer Cell Line Encyclopedia (CCLE: https://portals.broadinstitute.org/ccle_legacy/home) cancer gene expression profile will organize the gene expression patterns and cancer cells to drugs most half inhibitory concentration (IC50) fitting, calculating of 198 kinds of medicines. The sensitivity of the drug (between TKD and non-TKD) was analyzed by rank sum test.

Statistical Analysis

R version 4.1.3 was used for bioinformatics analysis, and p-value or adjusted p-value < 0.05 was considered statistically significant. The Kaplan-Meier was used to estimate the event time distribution, and the log-rank test was used to compare the PFS and OS of the patients.

3.1 Mutational landscape and characteristics of ERBB2 in non-small cell lung cancer

3.1.1 Basic mutational characteristics

55% (135/244) of ERBB2 mutations occurred in TKD, and 45% (109/244) occurred in non-TKD. In TKD, In_Frame_Insertion mutation (In_Frame_Ins) (45.08%, 110/135) was the most common type of mutation, followed by missense mutation (8.2%, 20/135). Among non-TKD, missense mutations (34.02%, 83/109) were the most common type (Table 1).

Table 1

Clinical characteristics of patients included in the study
Characteristics	TKD(N = 135)	non-TKD(N = 109)	Total(N = 244)	p-value
Cancer Type(%)				< 0.001
LUAD	129(52.87)	77(31.56)	206(84.43)
LUSC	0(0.0)	20(8.20)	20(8.20)
NSCLC^a	6(2.46)	12(4.92)	18(7.38)
Annotation(%)				< 0.001
Oncogenic	130(53.28)	51(20.90)	181(74.18)
Benign	0(0.0)	33(13.52)	33(13.52)
Unknow	5(2.05)	25(10.25)	30(12.30)
Mutation Type(%)				< 0.001
In_Frame_Ins	110(45.08)	0(0.0)	110(45.08)
Missense_Mutation	20(8.20)	83(34.02)	103(42.21)
Other^b	5(2.05)	26(10.66)	31(12.70)
Smoker(%)				< 0.001
Yes	65(26.64)	61(25.00)	126(51.64)
No	61(25.00)	25(10.25)	86(35.25)
Unknow	9(3.69)	23(9.43)	32(13.11)
Person Cigarette Smoking History Pack Year Value
Mean ± SD	11.36 ± 13.15	33.74 ± 28.79	22.38 ± 24.81
TMB
Mean ± SD	4.99 ± 5.99	15.27 ± 16.03	9.57 ± 12.65
PD-L1 score
Mean ± SD	42.92 ± 39.22	16.00 ± 35.78	35.00 ± 39.21
MSI Score
Mean ± SD	0.31 ± 1.75	1.36 ± 5.07	0.82 ± 3.75
Stage(%)				0.07
I	25(12.82)	35(17.95)	60(30.77)
II	6(3.08)	6(3.08)	12(6.15)
III	8(4.10)	9(4.62)	17(8.72)
IV	66(33.85)	40(20.51)	106(54.36)
T Stage(%)				0.23
T1	3(7.32)	12(29.27)	15(36.59)
T2	2(4.88)	14(34.15)	16(39.02)
T3	1(2.44)	2(4.88)	3(7.32)
T4	1(2.44)	1(2.44)	2(4.88)
T_X	3(7.32)	2(4.88)	5(12.20)
N Stage(%)				0.74
N0	15(25.42)	30(50.85)	45(76.27)
N1	2(3.39)	3(5.08)	5(8.47)
N2	2(3.39)	3(5.08)	5(8.47)
N3	1(1.69)	0(0.0)	1(1.69)
N_X	1(1.69)	2(3.39)	3(5.08)
Characteristics	TKD(N = 135)	non-TKD(N = 109)	Total(N = 244)	p-value
M Stage(%) M0	7(17.50)	25(62.50)	32(80.00)	0.07
M1	3(7.50)	4(10.00)	7(17.50)
M_X	0(0.0)	1(2.50)	1(2.50)
CT Size
Mean ± SD	2.40 ± 1.23	1.95 ± 1.13	2.23 ± 1.20
Sex(%)				0.12
Female	81(34.18)	51(21.52)	132(55.70)
Male	53(22.36)	52(21.94)	105(44.30)
Diagnosis Age
Mean ± SD	62.84 ± 10.68	64.26 ± 9.98	63.52 ± 10.35
Median[min-max]	64.00[37.00,86.00]	66.00[45.00,88.00]	65.00[37.00,88.00]
DFS
Mean ± SD	14.46 ± 15.87	28.66 ± 21.80	26.29 ± 21.23
a Specific pathological type was unknown. b Other mutations include Nonsense_Mutation, Frame_Shift_Del, Frame_Shift_Ins, Splice_Region, Splice_Site, and fusion. LAUD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; NSCLC, non-small cell lung cancer; TMB, Tumor Mutation Burden; DFS, Disease-Free Survival.

3.1.2 Co-mutation in tumor genes and tumor suppressor genes

In patients with ERBB2 mutations, there were 14 cancer genes or tumor suppressor genes with at least 5% co-mutation rates, including TP53, EGFR, KRAS, PIK3CA, STK11, BRAF, and KEAP1. TP53 was the most common co-mutated gene in ERBB2 mutations (62.0%) and co-mutated with other genes more frequently in non-TKD patients than in TKD (Fig. 1).

3.2. Clinical characteristics of TKD and non-TKD in NSCLC

3.2.1 Basic clinical characteristics (Table 1)

3.2.1.1 Both TKD and non-TKD mutations mainly occur in lung adenocarcinoma. In lung squamous cell carcinoma, all mutations occur in the non-TKD mutant domain (p < 0.001). Six mutations were oncogenic in 20 patients with lung squamous cell carcinoma; eight had unknown mutation significance.

3.2.1.2 Most TKD mutations were oncogenic, while 39.2% were benign mutations in the non-TKD mutation (p < 0.001).

3.2.1.3 All In_Frame_Ins occurred in the TKD domain, while most of the Missense_Mutation occurred in the non-TKD domain (p < 0.001).

3.2.1.4 Non-TKD mutations occur more frequently in patients who smoke, and the smoking index (33.74 ± 28.79) is much higher than that in patients with TKD.

3.2.1.5 Patients with the non-TKD mutation had higher tumor mutation burden (TMB) and MSI score than those with TKD mutation (Fig. 2B-C).

3.2.1.6 TKD mutation had a higher PD-L1 score than those with the non-TKD mutation.

In addition, there were no significant differences in tumor stage, size, gender, and age between non-TKD and TKD patients. The mean age at diagnosis of TKD was 62.84 ± 10.68; 60% were female, and 48% were smokers. Among all TKD mutations, Y772_A775dupYVMA was the most dominant mutation (51.1%, 69/135), followed by G778_P780dupGSP (8.1%, 11/135) and G776delinsVC (10.4%, 14/135). Among all non-TKD mutations, S310F was the most dominant mutation (11% 12/109), followed by R103Q (6.4% 7/109). The most common mutations above were oncogenic mutations.

3.2.2 Survival prognosis

In the database of cbioportal, the PFS of non-TKD is longer than TKD and has statistical difference(p < 0.001) (Fig. 2A).

3.3 The identification of differentially expressed genes and Enrichment Analysis.

3.3.1 Differential genes: 788 DEGs were screened, and p < 0.01 was the threshold after adjustment. Compared with non-TKD samples, 638 up-regulated and 149 down-regulated DEGs were identified in TKD samples.

3.3.2 Enrichment Analysis of functional pathway

The function and pathway enrichment of 638 up-regulated DEG genes were analyzed. CC-related DEGs were significantly enriched in the collagen-containing extracellular matrix, endolysosomes, and granular secretory membrane (Fig. 3A). BP-related DEGs were significantly enriched in leukocyte activation, migration, and activation involved in the immune response (Fig. 3B). DEGs linked to MF are significantly enriched in immune receptor activity, cytokine receptor activity, sialic acid binding, and phosphoric diester hydrolase activity (Fig. 3C).

3.4 Construction of the PPI Network and Identification of Hub Genes

LYN, BTK, PIK3CA, STAT3, KITLG, STAT5A, EPOR, INPP5D, LCP2, and MS4A2, the top ten genes obtained through PPI and Cytoscape analysis, were considered as hub genes. LYN, BTK, and PIK3CA genes were in the most critical position and became the focus of subsequent research (Fig. 3D-F).

3.5 LYN, BTK, and PIK3CA mutation and prognosis.

The above study identified LYN, BTK, and PIK3CA genes as hub genes. In the cbioportal database, the predictive effect of the above hub gene mutation on the prognosis of patients with NSCLC-ERBB2 mutation was further investigated. Due to the small number of commutant BTK or LYN in ERBB2, the prognostic effect cannot be further predicted. The overall survival of PIK3CA mutation was compared in patients with TKD and non-TKD mutations, and there was no significant difference between the two groups. (Fig S1) The LYN, BTK, and PIK3CA mutations were compared in all NSCLC patients. Compared with wild-type patients, patients with LYN or BTK mutations had no significant difference in PFS(Fig. 4A-B). However, patients with PIK3CA mutations had longer PFS than wild-type patients. The combination of BTK, PIK3CA, and LYN mutation was superior to single prediction (p 5.95e-10) (Fig. 4D-G), indicating that viewing BTK, LYN, and PIK3CA mutation as a whole could better predict the prognosis of NSCLC patients.

3.6 Selecting sensitive drugs for patients with TKD and non-TKD mutations by oncoPredict

To find drugs suitable for patients with TKD and non-TKD mutations, we used the oncoPredict algorithm to convert the gene expression of TKD and non-TKD mutant samples into a drug susceptibility matrix (Table S1). Compared with the TKD group, the non-TKD group had a better response to Gefitinib_1010 (targeted drug, EGFR inhibitor), Rapamycin_1084 (targeted drug, PI3K/ mTOR inhibitor), Sorafenib_1085 (targeted drug, Receptor tyrosine kinase inhibitor), Linsitinib_1510 (targeted drug, IGF1R inhibitor), Uprosertib_1553 (targeted drug, AKT inhibitor), OSI-027_1594 (targeted drug, IGF1R inhibitor) mTOR inhibitor), VE-822_1613 (targeted drug, ATM inhibitor), WnT-C59_1622 (targeted drug, Wnt inhibitor), I-BET-762_1624 (targeted drug, BET inhibitor), OTX015_1626 (targeted drug, ATM inhibitor) BRD2/3/4 inhibitors), AZD5153_1706 (targeted drugs, BRD4 inhibitors), Ibrutinib_1799 (targeted drugs, BTK inhibitors), Oxaliplatin_1806 (chemotherapy drugs), Carmustine_1807 (Chemotherapy drug), Dihydrorotenone_1827 (mitochondrial inhibitor), OF-1_1853 (chemotherapy drug), AZD3759_1915 (targeted drug, EGFR inhibitor), GDC0810_1925 (Estrogen receptor depressant), GSK2578215A_1927 (targeted drug, LRRK2 inhibitor), I-BRD9_1928(targeted drug, BRD9 inhibitor), NVP-ADW742_193 (Targeted drug, IGF-1R inhibitor), Savolitinib_1936 (targeted drug, MET inhibitor), BIBR-1532_2043 (apoptosis inducer), MK-8776_2046 (targeted drug, Chk1 inhibitor), Ulixertinib_2047 (targeted drug, ERK1/2 kinase inhibitor), VX-11e_2096 (targeted drug, ERK inhibitor), Uprosertib_2106 (targeted drug, Akt inhibitor), LJI308_2107 (targeted drug, RSK inhibitor). AZD6482_2169(targeted drug, PI3K inhibitor), Doramapimod_1042(targeted drug, MAPK inhibitor), SB505124_1194(targeted drug, TGFβR inhibitor), Daporinad_1248(target drug, TGFβr inhibitor), Daporinad_1248(Targeted drugs, NAMPT inhibitors), Lapatinib_1558(targeted drugs, EGFR inhibitors) are more sensitive in TKD. Based on the boxplot results, the significantly differentially sensitive drugs in the two groups were visualized (Fig. 5).

3.7 Therapeutic response in patients with TKD and non-TKD mutation

Two patients with TKD mutation and two with non-TKD mutation were collected on the clinical characteristic and the outcome of antitumor therapy. One patient with TKD mutation and one with non-TKD were treated with pyrotinib. PFS of the TKD mutation patient was as prolonged as 5.8 months, but no therapeutic effect was observed in the non-TKD one due to loss of follow-up. One TKD mutation and one non-TKD mutation were treated with disitamab vedotin, with PFS of 1.2 months and 1.5 months, respectively. One patient with common TKD mutation was treated with DS-8201 and had PFS for 6.5 months (Fig. 6). During the treatment of DS-8201, tumor markers decreased significantly. CEA decreased from 515ug/L to 112ug/ L; CA125 decreased from 1003KU/L to 110KU/L; NSE decreased from 38.1ug/L to 15.8ug/L. When drug resistance occurred, CEA, CA125, and NSE increased to 407ug/L, 1995KU/L, and 34.1ug/L, respectively (Fig. 7A). The other non-TKD mutation patient was treated with DS-8201, with PFS of 3.7 months. During therapy, CEA decreased from 11522ug/L to 7187ug/ L; CA125 decreased from 452KU/L to 318KU/L; NSE decreased from 112ug/L to 35.2ug/L. When drug resistance occurred, CEA, CA125, and NSE increased to 8619ug/L, 418KU/L, and 30.8ug/L, respectively (Fig. 7B).

In this study, the somatic mutation rate of ERBB2 was 3.2%, similar to previous studies [10]. Our results indicate that many patients with TKD and non-TKD may have the significant oncogenic ability of ERBB2 mutations. Most ERBB2 mutations occurring in the TKD domain are oncogenic mutations, while the benign mutations occurring in the non-TKD domain account for 39.2%. In TKD, Y772_A775dupYVMA was the most dominant mutation, followed by G778_P780dupGSP and G776delinsVC. Among all non-TKD mutations, S310F was the most dominant mutation, followed by R103Q. The most common mutations above are oncogenic, consistent with previous reports [11]. In colorectal cancer, the most common ERBB2 mutations are I655V, V842I, and R678Q [12], which indicates the heterogeneity of mutations of the same oncogene in different types of cancer. There was no significant difference in tumor stage, size, gender, or age between non-TKD and TKD patients. The only factor associated with ERBB2 mutation was smoking status and smoking index. Non-TKD mutations occur more frequently in smoking patients, and the smoking index of non-TKD (33.74 ± 28.79) is much higher than that of TKD patients, consistent with previous studies [5].

Previous studies mainly focused on TKD mutation, especially exon 20 insertion mutation, and rarely included non-TKD mutation. However, various preclinical studies and case reports have found that non-TKD mutation may be a therapeutic target. With the wide application of ADC drugs and the approval of relevant indications, the difference between TKD and non-TKD mutations may be the reference targets for different ADC drugs. This study also found that all TKD and most non-TKD mutations occurred in adenocarcinoma. In contrast, all mutations in squamous cell carcinoma were non-TKD, which provided more comprehensive information about ERBB2 mutations in different domains of TKD and non-TKD. It was also observed that the co-mutation frequency of EGFR, KRAS, and other genes was significantly higher in non-TKD mutations. In the meantime, a higher frequency of KEAP1 and STK11 mutations was observed in non-TKD. KEAP1 mutation was a poor prognostic factor, and STK11 mutation was a negative predictor of immune checkpoint inhibitors. [13]

After finding the differences in mutational landscape and clinical characteristics between TKD and non-TKD mutation, we analyzed the survival prognosis of the two groups. We discovered that PFS of patients with the non-TKD mutation was longer than those with the TKD. Although the OS in the non-TKD group was longer than that in the TKD group, there was no statistical difference, possibly due to the insufficient sample. More cases could be included for further study.

This study identified Lyn, BTK, and PIK3CA as hub genes. The Lyn gene belongs to the Src kinase family and is on chromosome 8q13. Lyn gene is a primary regulator of various signal transduction pathways and is involved in the occurrence of a variety of malignant tumors [14]. BTK is a potent oncogene that plays a role in the downstream RAS/MAPK pathway and is essential for RAS-mediated transformation [15]. PIK3CA gene is a candidate driver gene for lung squamous cell carcinoma, which may be involved in tumor cell growth and the development of non-small cell lung cancer [16]. However, few studies have directly reported the relationship between these three genes and ERBB2 mutation in NSCLC. We can not further predict the prognostic effect due to the few commutant BTK or LYN. However, we found that patients with PIK3CA mutations had longer PFS than wild-type patients. And the combination of BTK, PIK3CA, and LYN mutation was superior to a single prediction, which indicates that viewing BTK, LYN, and PIK3CA mutation as a whole could better predict the prognosis of NSCLC patients.

Then, we analyzed the anti-tumor drug sensitivity of the samples with TKD and non-TKD mutation and found that the samples in the TKD group were sensitive to 5 drugs, while non-TKD were sensitive to 28 medicines. This finding may guide future treatment in TKD and non-TKD patients. Finally, four patients from an internal cohort were included to observe the treatment response. In one patient with TKD mutation treated with pyrotinib, PFS lasted up to 5.8 months, consistent with previous treatment results [17]. In another patient with TKD mutation, DS-8201 resulted in a PFS of 6.5 months, compatible with the median PFS of 8.2 months (95% CI, 6.0 to 11.9) in previous clinical studies [3]. In contrast, the non-TKD patient treated with DS-8201 had a PFS of 3.7 months, which was significantly shorter than the patient with the TKD mutation. This may be because the TKD mutation is more sensitive to DS-8201. In two patients with the non-TKD mutation, one patient was first treated with anlotinib, which got PR, then switched to pyrotinib due to the discovery of non-TKD mutation in the biopsy tissue samples. However, no non-TKD data on pyrotinib was obtained due to the loss of follow-up. Another patient with non-TKD mutation was treated with disitamab vedotin, PFS was 1.5 months, and then treated with bevacizumab, with overall survival of 8.8 months.

Acknowledgments

none

Disclosure

Data Availability

The authors claim that all data can be found in two databases: cbioportal (https://www.cbioportal.org/) and TCGA(https://xenabrowser.net/datapages/).

Funding Information

The authors state that the study was conducted without any business or financial relationships, which could be interpreted as a potential conflict of interest.

Conflict of Interest

The authors have no conflict of interest.

Ethics Statement

- Approval of the research protocol by an Institutional Reviewer Board: N/A

- Informed Consent: All informed consent was obtained from the subject(s) and/or guardian(s).

- Registry and the Registration No. of the study/trial: N/A

- Animal Studies: If not applicable, please write N/A

Author Contributions

Ting Wei and Xiaowen Wu designed the study and provided the patients of the study. Xiaopan Zheng and Haoxuan Ying completed the manuscript writing. Xiaopan Zheng and Haitao Li collected data and analyzed data, and Ting Wei, Weiliang Zhu, and Manming Cao approved the final manuscript. All authors contributed to the article and approved the submitted version.

Vermeulen, Z., V.F. Segers and G.W. De Keulenaer, ErbB2 signaling at the crossing between heart failure and cancer. Basic Res Cardiol, 2016. 111(6): p. 60.
Ren, S., et al., Consensus for HER2 alterations testing in non-small-cell lung cancer. ESMO Open, 2022. 7(1): p. 100395.
Li, B.T., et al., Trastuzumab Deruxtecan inHER2-Mutant Non–Small-Cell Lung Cancer. New England Journal of Medicine, 2022. 386(3): p. 241-251.
Gaibar, M., et al., Somatic Mutations inHER2 and Implications for Current Treatment Paradigms inHER2-Positive Breast Cancer. Journal of Oncology, 2020. 2020: p. 1-13.
Wei, X.W., et al., Mutational landscape and characteristics of ERBB2 in non-small cell lung cancer. Thorac Cancer, 2020. 11(6): p. 1512-1521.
Ou, S.I., et al., HER2 Transmembrane Domain (TMD) Mutations (V659/G660) That Stabilize Homo- and Heterodimerization Are Rare Oncogenic Drivers in Lung Adenocarcinoma That Respond to Afatinib. Journal of Thoracic Oncology, 2017. 12(3): p. 446-457.
Yamamoto, H., et al., Therapeutic Potential of Afatinib for Cancers withERBB2 (HER2) Transmembrane Domain Mutations G660D and V659E. The Oncologist, 2018. 23(2): p. 150-154.
Bob T. Li, R.S.D.B., et al., Ado-Trastuzumab Emtansine for Patients With HER2- Mutant Lung Cancers: Results From a Phase II Basket Trial. 2018. p. 36(24):2532-2537.
Shen, W., et al., Sangerbox: A comprehensive, interaction‐friendly clinical bioinformatics analysis platform. iMeta, 2022. 1(3).
Maeser, D., R.F. Gruener and R.S. Huang, oncoPredict: an R package for predictingin vivo or cancer patient drug response and biomarkers from cell line screening data. Briefings in Bioinformatics, 2021. 22(6).
Robichaux, J.P., et al., Pan-Cancer Landscape and Analysis of ERBB2 Mutations Identifies Poziotinib as a Clinically Active Inhibitor and Enhancer of T-DM1 Activity. Cancer Cell, 2019. 36(4): p. 444-457.e7.
Arcila, M.E., et al., Prevalence, Clinicopathologic Associations, and Molecular Spectrum of ERBB2 (HER2) Tyrosine Kinase Mutations in Lung Adenocarcinomas. Clinical Cancer Research, 2012. 18(18): p. 4910-4918.
Loree, J.M., et al., Molecular Landscape ofERBB2/ERBB3 Mutated Colorectal Cancer. JNCI: Journal of the National Cancer Institute, 2018. 110(12): p. 1409-1417.
Nisar, A. and M. A. Kayani, et al. (2022). "Fyn and Lyn gene polymorphisms impact the risk of thyroid cancer." Molecular Genetics and Genomics 297(6): 1649-1659.
Grassilli, E., et al., p65BTK Is a Novel Biomarker and Therapeutic Target in Solid Tumors. Frontiers in Cell and Developmental Biology, 2021. 9.
Wang, Y., et al., Clinical Significance of PIK3CA Gene in Non-Small-Cell Lung Cancer: A Systematic Review and Meta-Analysis. BioMed Research International, 2020. 2020: p. 1-9.
Song, Z., et al., Efficacy and safety of pyrotinib in advanced lung adenocarcinoma with HER2 mutations: a multicenter, single-arm, phase II trial. BMC Medicine, 2022. 20(1).

No competing interests reported.

FigS1.tif
Fig S1 The overall survival of PIK3CA mutation in TKD and non-TKD mutations.
TableS1.xlsx

Download PDF

Version 1

posted

You are reading this latest preprint version

Clinical and molecular characteristics, transcriptome, and drug efficacy of ERBB2 tyrosine kinase and non-tyrosine kinase mutations in non-small cell lung cancer

Status:

Version 1

Abstract

Figures

Introduction

Material and Methods

Study cohort

Mutation analyses

Clinical characteristics

Construction of the PPI Network and Identification of Hub Genes

Drug sensitivity analysis

Statistical Analysis

Results

3.1 Mutational landscape and characteristics of ERBB2 in non-small cell lung cancer

3.1.1 Basic mutational characteristics

3.1.2 Co-mutation in tumor genes and tumor suppressor genes

3.2. Clinical characteristics of TKD and non-TKD in NSCLC

3.2.1 Basic clinical characteristics (Table 1)

3.2.2 Survival prognosis

3.4 Construction of the PPI Network and Identification of Hub Genes

3.6 Selecting sensitive drugs for patients with TKD and non-TKD mutations by oncoPredict

3.7 Therapeutic response in patients with TKD and non-TKD mutation

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1