Investigation of age and smoking in NSCLC patients with uncommon EGFR mutations (2025)

Abstract

INTRODuCTION

The epidermal growth factor receptor (EGFR) mutation was thefirst driver gene discovered for non-small cell lung cancer (NSCLC)(Attili) (John). Among EGFR mutations, Ex19deletion and Exon 21 L858R are two of the mostcommon mutations, and there are multiple uncommon mutations (1,2).Among the uncommon mutations, G719X,L861Q, and S768I mutations arerelatively frequent (1,2). Therefore, there have been many reportsthat treat these mutations collectively as major uncommon mutations(1,2). It is known that patients with these gene mutations respondto second-generation EGFR- tyrosine kinase inhibitors (TKIs), butpatients with Exon 20 insertions do not respond to EGFR-TKIs (3,4).At present, the existence of patients with many compound mutationswith common or uncommon EGFR mutations has been recognized (4). Notonly are they rare, but they are also genetically heterogeneouspopulations, and their responses to therapeutic drugs are not thesame. As such, there are not many studies investigating patientbackgrounds, such as age and smoking, in detail (5-19). Inparticular, only a few studies have shown information on more than100 patients with uncommon mutations (6-8,11,13,15,18).

In view of this, we conducted this study to clarify clinicalcharacteristics, with particular focus on age and smoking history,of NSCLC patients with uncom- mon EGFR mutations.

MATERIALS and METHODS

The medical records of all NSCLC patients diagnosed at 14 medicalinstitutions in our prefecture from July 2002 to December 2023 wereexamined. Based on the World Health Organization classification, thepathological diagnosis of each NSCLC patient was made (20). Beforestarting treatment, all patients underwent TNM classification usinghead computed tomography or magnetic resonance imaging, bone orpositron emission scan, and abdominal ultrasound and/or computedtomography (21). At the time of NSCLC diagnosis, the followingpatient background characteristics were investigated: Sex, age,Eastern Cooperative Oncology Group performance status (PS), clinicalstage, presence of EGFR mutation and EGFR mutation subtype. The‘number of cigarettes smoked per day’ and ‘years of smoking’ werealso investigated. The product of these indices was used as thesmoking index (22,23).

For statistical analyses, the Chi-squared test was used to testfor differences in proportions. The Mann- Whitney U test was used tocompare values between two unmatched groups, such as patient age andsmoking index. Correlations were examined using the Spearmancorrelation coefficient. A P-value less than 0.01 was considered toindicate a significant difference.

This study was approved by the Institutional Review Board ofUniversity of Tsukuba Mito Medical Center/ Mito Kyodo GeneralHospital (NO-23-53) and by each institute that participated in thisstudy.

RESuLTS

Characteristics of Patients

During the study period, clinical information on 158 patientswith uncommon EGFR mutations was col- lected from 14 institutions.Median age of these patients was 72 years (range, 35-92 years), andthere were 86 male and 72 female patients. There were 153 patientswith adenocarcinoma and five patients with other histological types.The clinical stage was IA-IIIC in 83 patients, and IVA-B in 75patients. With regard to PS, 137 patients had PS 0-1, and 21patients had PS 2-4. There were 98 patients with major uncommonmutations, G719X, L861Q, andS768I, 41 with com- pound mutations, and 19 withExon 20 insertions. Shows the age distribution of all patientsFigure 1. The highest number of patients were in their 70s. Therewere 25 patients aged up to 59 years, and 47 patients aged 75 yearsor older. There was a significant differ- ence between theproportion of patients aged up to 59 years and that aged 75 yearsand older (p= 0.0046).

Comparison Among uncommon EGFR Mutations

A comparison of patient background factors was per- formed withpatients with three major mutations, G719X,L861Q, and S768I, as Group 1,patients with compound mutations as Group 2, and patients with Exon20 insertions as Group 3. Patient background

factors among the three groups are shown in Table 1. There wereno significant differences in age, sex his- tology, clinical stage,or PS among the three groups. In addition, we focused on smoking andcompared the percentage of non-smokers, the percentage of lightsmokers (smoking index of 100 or less), and the smoking index, butthere were no significant differ- ences among the three groups.

Correlation and Comparison between Age and Smoking Index in theThree Groups Due to uncommon EGFR mutations

Figure 2-A shows the correlation between smoking index and age inall 158 patients. There was no sig- nificant correlation betweensmoking index and age in these patients (Spearman’s rank correlationco- efficient p= 0.9059, p= 0.009). Next, we investigated thecorrelation between smoking index and age among the 88 smokers. Theresults are shown in Figure 2-B. For smokers only, there was asignificant correlation between smoking index and age (Spearman’srank correlation co-efficient p= 0.0002, p= 0.397).

Patients were divided into non-smokers and smokers, and their agedistributions are shown in Figures 3-A and B. In both groups, 70non-smokers and 88 smokers, the most modal value for age was in the70s. For non-smokers, the peak was at ages 70-74,

Figure 1. Age distribution of all 158 NSCLC patientswith uncommon EGFR mutation (86 male patients and 72 femalepatients).

Figure 2. Correlation between smoking index and agein all 158 NSCLC patients (A). There was no significantcorrelation between smoking index and age in these patients(Spearman’s rank correlation co-efficient p= 0.9059, p= 0.009).Correlation between smok- ing index and age in 88 NSCLC patientswith smoking habit (B). There was a significantcorrelation between smoking index and age (Spearman’s rankcorrelation co-efficient p= 0.0002, p= 0.397).

and for smokers, it was at ages 65-69. There was a significantdifference in the age distribution of the non-smoker and smokergroups (p= 0.0192, Chi- squared test).

DISCuSSION

This study confirmed the following results: The median age of the158 patients with EGFR uncommon mutations was 72 years. Regardingthe age distribution

of all patients, the most modal value for age was in the 70s.There was a significant difference between the proportion of thepatients aged up to 59 years and the proportion of those aged 75years or older. In 88 patients with a smoking habit, a significantcorrelation was found between the smoking index and age. Amongnon-smokers, there was a peak between ages 70 and 74, which wasolder than the peak among smokers.

AB

Figure 3. Age distribution of 70 NSCLC patientswithout smoking habit (A) and 88 NSCLC with smokinghabit (B). Significant differ- ence in the agedistribution between these two groups of patients (p= 0.0192,Chi-squared test).

Most studies to date on patients with uncommon EGFR mutationshave focused on stage III-IV patients (5-7,10,12-17), and very fewreports have included data on patients at all stages (9,19).Furthermore, the number of patients evaluated in these previousstud- ies has been very small; 26 and 40 patients, respec- tively(9,19). In past surveys of patients with stage III-IV disease anduncommon EGFR mutations, the proportion of female patients was37%-75% (5-7,9- 19), and the proportion of patients with PS 0-1 was60%-100% (6,9,10,12-17). The proportion of patients withadenocarcinoma was 16.7% in a study of 291 patients by Evans et al.(11), but other studies have generally reported proportions over 90%(5,7,10-17). In previous studies involving more than 100 patients,uncommon mutations have been classified into three groups: Majoruncommon mutations, G719X, L861Q,and S768I; compound mutations; and Exon 20 inser-tions (6,7,11,13,15,18). Our study involved a rela- tively largenumber of patients, including patients at all stages of NSCLC fromstage IA to stage IVB. In this survey, 42.4% were women, 86.7% werepatients with PS 0-1, and 96.8% were adenocarcinoma patients.Focusing on uncommon mutation subtypes, 98 patients (62%) had major,41 (25.9%) had com- pound, and 19 (12%) had Exon 20 insertions. Itis

known that the positive rate of EGFR mutations in NSCLC differsbetween Asians and Caucasians (24), and it was necessary to confirmthese background factors. However, these results were notsignificantly different from previous studies (5-19).

In previously conducted EGFR-TKI clinical trials, median age ofthe patients with uncommon EGFR mutations was 58-64 years (6,12,14).In a recent TKI clinical trial of over 40 patients with uncommonEGFR mutations, median age has been found as 72 years (19). On theother hand, in most studies in clinical practice except one (11),median age has been found as 59-68 years (5,7-11,13,15-18). Theexception is a study of 291 patients in the United Kingdom by Evanset al., in which the average age is

70.1 years (11). The results from clinical practice from Evans etal. and our study suggest that even patients older than 70 yearsmight harbor uncommon EGFR mutations (11). Not conducting a searchfor driver genes in NSCLC due to advanced age should be avoided, asthis might limit treatment options.

In studies conducted so far, the proportion of smokers amongpatients with uncommon EGFR mutations has been found as 44%-59% inclinical trials (12,14) and as 48%-69% in clinical practice, exceptfor one study

from India involving 40 patients, which has shown a non-smokingrate of 83% (7,9,10,13,15-18). In the present study, the proportionof non-smokers was 44.3%. Although it has been reported that thepro- portion of smokers is higher in patients with uncom- mon EGFRmutations than in patients with common mutations, to the best of ourknowledge, there have been no reports that have considered bothsmoking history and age (13). The results of this study show that inboth groups, 70 smokers and 88 non-smokers, the most modal value forage was in the 70s. On the other hand, there was a significantdifference in age distribution between the non-smoker and smokergroups. In other words, the number of patients increased up to theage of 74 in both non-smoking and smoking groups, and after that,the distribution showed a difference between the two groups. A sig-nificant correlation was found between age and smoking index inpatients with smoking history. It has been speculated that this isnot simply due to an increase in the number of years of smoking, butthat there may be some other cause that remains unknown.

Although the above novel findings were obtained, this study hassome limitations. We used several testing methods for EGFRmutations, but comparisons could not be made because the testingmethods were not integrated due to the multicenter nature of thestudy, and there was no information on EGFR gene- negative patientswho were treated around the same time. In addition, the study periodwas long because it was intended to collect a large number ofpatients. However, there are few reports that have investigated morethan 150 patients, and we do believe that the information obtainedmight be useful for the future medical treatment of patients withuncommon EGFR mutations.

CONCLuSION

The implementation of driver gene testing for NSCLC is expectedto provide important information for selecting treatment optionstailored to the patient. Therefore, even in NSCLC patients who areelderly or who have a history of smoking, although it is unclearwhether EGFR mutations are common or uncommon, EGFR gene testingshould be performed in case an EGFR mutation is present.

Acknowledgment

We would like to thank the researchers at the followingfacilities for their cooperation in this study:

Hitachi General Hospital, Hitachinaka General Hospital,Ibarakihigashi Hospital, Tsukuba Medica Center Hospital, TsuchiuraKyodo General Hospital, Ibaraki Medical Center-Tokyo MedicalUniversity.

Ethical Committee Approval: This study was obtainedfrom General Hospital Mito Kyodo Hospital Director of HospitalEthics Committee (Decision no: 23-53, Date: 13.12.2023).

CONFLICT of INTEREST

The authors declare that they have no conflict of interest.

AuTHORSHIP CONTRIBuTIONS

Concept/Design: YM, SO, KM, HS Analysis/Interpretation: YM, SO,KM, HS

Data acqusition: YM, MT, TK, TL, TN, TS, KM, RN, KI, TS

Writing: YM, KM, HS Clinical Revision: HS Final Approval: HS,NH

REFERENCES

• 1.

  Attili I, Passaro A, Pisapia P, Malapelle U, deMarinis F. Uncommon EGFR compound mutations in non-small celllung cancer (NSCLC): A systematic review of availableevidence. Curr Oncol 2022; 29: 255-66. https://doi.org/10.3390/curroncol29010024

• 2.

  John T, Taylor A, Wang H, Eichinger C, Freeman C,Ahn MJ. Uncommon EGFR mutations in non-small-cell lung can-cer: A systematic literature review of prevalence and clinicaloutcomes. Cancer Epidemiol 2022; 76: 102080.https://doi.org/10.1016/j.canep.2021.102080

• 3.

  Dorta-Suárez M, de Miguel M, Amor-Carro O, CalderónJM, González-Ortega M, Rodríguez-Abreu D. The state of the artof EGFR exon 20 insertions in non-small cell lung cancer:Diagnosis and future perspectives. Cancer Treat Rev 2023; 124:102671. https://doi.org/10.1016/j.ctrv.2023.102671

• 4.

  Pretelli G, Spagnolo CC, Ciappina G, Santarpia M,Pasello G. Overview on therapeutic options in uncommon EGFRmutant non-small cell lung cancer (NSCLC): New lights for an unmetmedical need. Int J Mol Sci 2023; 24: 8878.https://doi.org/10.3390/ijms24108878

• 5.

  Heigener DF, Schumann C, Sebastian M, Sadjadian P,Stehle I, Märten A, et al. Afatinib Compassionate UseConsortium (ACUC). Afatinib in non-small cell lung can- cerharboring uncommon EGFR Mutations pretreated with reversibleEGFR inhibitors. Oncologist 2015; 20: 1167-74.https://doi.org/10.1634/theoncolo-gist.2015-0073

• 6.

  Yang JC, Sequist LV, Geater SL, Tsai CM, Mok TS,Schuler M, et al. Clinical activity of afatinib in patients withadvanced non-small-cell lung cancer harbouring uncom- mon EGFRmutations: A combined post-hoc analysis of LUX-Lung 2, LUX-Lung 3,and LUX-Lung 6. Lancet Oncol 2015; 16: 830-8.https://doi.org/10.1016/S1470-2045(15)00026-1

• 7.

  Tu HY, Ke EE, Yang JJ, Sun YL, Yan HH, Zheng MY, et al.A comprehensive review of uncommon EGFR mutations in patients withnon-small cell lung cancer. Lung Cancer 2017; 114: 96-102.https://doi.org/10.1016/j.lung-can.2017.11.005

• 8.

  Wu SG, Chang YL, Yu CJ, Yang PC, Shih JY. Lungadenocar- cinoma patients of young age have lower EGFR mutationrate and poorer efficacy of EGFR tyrosine kinase inhibi- tors. ERJOpen Res 2017; 3: 00092-2016. https://doi.org/10.1183/23120541.00092-2016

• 9.

  Frega S, Lorenzi M, Fassan M, Indraccolo S, CalabreseF, Favaretto A, et al. Clinical features and treatment outcome ofnon-small cell lung cancer (NSCLC) patients with uncommon orcomplex epidermal growth factor receptor (EGFR) mutations.Oncotarget 2017; 8: 32626-38.https://doi.org/10.18632/oncotarget.15945

• 10.

  Chantharasamee J, Poungvarin N, Danchaivijitr P,Techawatanawanna S. Clinical outcome of treatment of metastaticnon-small cell lung cancer in patients harboring uncommon EGFRmutation. BMC Cancer 2019; 19: 701.https://doi.org/10.1186/s12885-019-5913-9

• 11.

  Evans M, O’Sullivan B, Smith M, Hughes F, Mullis T,Trim N, et al. Large-scale EGFR mutation testing in clinical prac-tice: Analysis of a series of 18,920 non-small cell lung cancercases. Pathol Oncol Res 2019; 25: 1401-9. https://doi.org/10.1007/s12253-018-0460-2

• 12.

  Cho JH, Lim SH, An HJ, Kim KH, Park KU, Kang EJ, et al.Osimertinib for patients with non-small-cell lung cancer harboringuncommon EGFR mutations: A multicenter, open-label, phase II trial(KCSG-LU15-09). J Clin Oncol 2020; 38: 488-95.https://doi.org/10.1200/JCO.19.00931

• 13.

  Ko HW, Shie SS, Wang CW, Chiu CT, Wang CL, Yang TY, etal. Association of smoking status with non-small cell lung cancerpatients harboring uncommon epidermal growth factor receptormutation. Front Immunol 2022; 13: 1011092.https://doi.org/10.3389/fimmu.2022.1011092

• 14.

  Li HS, Yang GJ, Cai Y, Li JL, Xu HY, Zhang T, et al.Dacomitinib for advanced non-small cell lung cancer patientsharboring major uncommon EGFR alterations: A dual-center,single-arm, ambispective cohort study in China. Front Pharmacol2022; 13: 919652. https://doi.org/10.3389/fphar.2022.919652

• 15.

  Popat S, Hsia TC, Hung JY, Jung HA, Shih JY, Park CK,et al. Tyrosine kinase inhibitor activity in patients with NSCLCharboring uncommon EGFR mutations: A retrospective internationalcohort study (UpSwinG). Oncologist 2022; 27: 255-65.https://doi.org/10.1093/oncolo/oyac022

• 16.

  Ullas B, Shrinidhi N, Mansi S, Narayan S, Parveen J,Surender D, et al. All EGFR mutations are (not) created equal:Focus on uncommon EGFR mutations. J Cancer Res Clin Oncol 2023;149: 1541-9. https://doi.org/10.1007/s00432-022-04033-x

• 17.

  Bar J, Peled N, Schokrpur S, Wolner M, Rotem O, GirardN, et al. Uncommon EGFR mutations: International case series onefficacy of osimertinib in real-life practice in first- linesetting (UNICORN). J Thorac Oncol 2023; 18: 169-80.https://doi.org/10.1016/j.jtho.2022.10.004

• 18.

  Wang C, Zhao K, Hu S, Dong W, Gong Y, Xie C. Clinicaloutcomes of afatinib versus osimertinib in patients with non-smallcell lung cancer with uncommon EGFR muta- tions: A pooledanalysis. Oncologist 2023; 28: e397-e405.https://doi.org/10.1093/oncolo/oyad111

• 19.

  Okuma Y, Kubota K, Shimokawa M, Hashimoto K, KawashimaY, Sakamoto T, et al. Tokyo Cooperative Oncology Group (TCOG).First-line osimertinib for previ- ously untreated patients withNSCLC and uncommon EGFR mutations: The UNICORN phase 2nonrandomized clinical trial. JAMA Oncol 2024; 10: 43-51.https://doi.org/10.1001/jamaoncol.2023.5013

• 20.

  WHO Classification of Tumours Editorial Board. WorldHealth Organization Classification of Tumours, Thoracic Tumours5th ed. IARC Press, 2021.

• 21.

  Goldstraw P, Chansky K, Crowley J, Rami-Porta R,Asamura H, Eberhardt WE, et al. International Association for theStudy of Lung Cancer Staging and Prognostic Factors Committee,Advisory Boards, and Participating Institutions, InternationalAssociation for the Study of Lung Cancer Staging and PrognosticFactors Committee Advisory Boards and Participating InstitutionsThe IASLC lung can- cer staging project: Proposals for revision ofthe TNM stage groupings in the forthcoming (eighth) edition of theTNM classification for lung cancer. J Thorac Oncol 2016; 11:39-51. https://doi.org/10.1016/j.jtho.2015.09.009

• 22.

  Maeshima AM, Tochigi N, Tsuta K, Asamura H, Matsuno Y.Histological evaluation of the effect of smoking on periph- eralsmall adenocarcinomas of the lung. J Thorac Oncol 2008; 3:698-703. https://doi.org/10.1097/JTO.0b013e31817c60ae

• 23.

  Nakamura R, Inage Y, Tobita R, Yoneyama S, Numata T,Ota K, et al. Sarcopenia in resected NSCLC: Effect onpostoperative outcomes. J Thorac Oncol 2018; 13: 895-903.https://doi.org/10.1016/j.jtho.2018.04.035

• 24.

  Batra U, Biswas B, Prabhash K, Krishna MV. Differentialclinicopathological features, treatments and outcomes in patientswith Exon 19 deletion and Exon 21 L858R EGFR mutation-positiveadenocarcinoma non-small-cell lung cancer. BMJ Open Respir Res2023; 10: e001492. https://doi.org/10.1136/bmjresp-2022-001492