Advances in molecular biology have led to the discovery of numerous driver genes, resulting in great advances in the treatment of advanced non-small cell lung cancer (NSCLC). The epidermal growth factor receptor (EGFR) gene was the first driver gene to be discovered, and the introduction of the therapeutic tyrosine kinase inhibitors (TKIs) into clinical practice has led to a significant improvement in patient prognosis (1-4). For conventional cytotoxic antitumor drugs, the antitumor effects are expected to be dose-dependent, thus the most common method has been to determine the maximum tolerated dose by gradually increasing the dose. However, because such a conventional method was unable to determine the maximum tolerated dose of EGFR-TKIs, the recommended dose for each TKI was determined using different methods (5-9). Importantly, the phase I trials that determined the recommended dose of EGFR-TKIs included many patients who were younger than those seen in actual clinical practice, had good general condition, and did not have any comorbidities (5-9). Furthermore, the number of patients involved in these determinations was not necessarily large (5-9).

It was initially reported that EGFR gene-positive lung cancer patients tended to be female, younger, had adenocarcinoma, and were non-smokers (10). Subsequent clinical studies later revealed that EGFR gene-positive patients included a fair proportion of elderly patients and patients with smaller body sizes (11). Looking back at the EGFR-TKI clinical trials with a focus on dosage, the proportion of patients who received a reduced dose below the recommended dose was not so low as to be negligible (4). More recently, surveys of actual clinical practice have shown that EGFR-TKI dose reduction is being carried out in a not-insignificant proportion of patients (12-15).

In light of this, the present study was conducted to clarify the current clinical situation regarding the following two issues: the characteristics of patients undergoing EGFR-TKI dose reduction; and whether reducing the dose of EGFR-TKI has a detrimental effect on patient survival. Because this was a retrospective study that did not involve a large number of patients, the study does not directly lead to definitive conclusions. However, we believe that the results obtained in this study might provide some suggestions for similar research in the future, and we therefore report them herein.

Patients. This was a retrospective study that examined the medical records of all EGFR mutation-positive NSCLC patients who received EGFR-TKI treatment between August 2008 and April 2024 at two hospitals: Mito Kyodo Hospital, Mito Medical Center, University of Tsukuba, and Ryugasaki Saiseikai Hospital. The pathological diagnosis of NSCLC was based on the World Health Organization classification (16). Before the start of anticancer treatment, all patients with EGFR-positive NSCLC were clinically staged according to the TNM classification (17) using head computed tomography (CT) or magnetic resonance imaging, bone scan, 2-deoxy-2-[18F]-fluoro-D-glucose positron emission tomography/CT, ultrasound, and/or abdominal CT. EGFR mutation positivity in each patient was confirmed using either the cobas® EGFR Mutation Test (Roche Diagnostics, Tokyo, Japan), Oncomine Dx Target Test (Life Technologies Japan, Tokyo, Japan), or Foundation One (Foundation Medicine, Boston, MA, USA). The following clinical information was extracted from medical records: patient age, sex, Eastern Cooperative Oncology Group performance status (PS), histology, and clinical stage at the time of NSCLC diagnosis. Overall survival (OS) of each patient was defined as the time from the start of EGFR-TKI treatment to death or last follow-up, and progression-free survival (PFS) was defined as the time from the start of TKI treatment to disease progression or death. In this study, EGFR-TKI patients were defined as those receiving any of five TKIs that could be dose-reduced: erlotinib, afatinib, osimertinib, and dacomitinib. Dose reduction of EGFR-TKI was defined as any reduction from the recommended dose at least once during the course of treatment. No stratification was performed in this study regarding the amount or duration of dose reduction.

Statistical analysis. In this study, the chi-squared test, Mann-Whitney U-test, and Fisher’s exact test were used for statistical comparison between groups. Survival curves were created using the Kaplan-Meier method, and comparisons between groups were performed using the log-rank test. In addition, the effect of dose reduction on OS and PFS was adjusted using the Cox proportional hazards model. The association between patient characteristics [age, body surface area (BSA), weight, PS, and stage] and survival was also evaluated using multivariate analysis. R version 4.4.1 (2024-06-14, R Foundation, Vienna, Austria) was used for all statistical analyses.

Ethics. This study complied with the “Ethical Guidelines for Clinical Research” of the Ministry of Health, Labor and Welfare of Japan. Comprehensive informed consent was obtained from each patient at the time of consultation. This study was approved by the Ethics Committee of the Mito Medical Center, University of Tsukuba (NO 20-57).

Comparison between the recommended dose group and the dose reduction group. The total number of patients administered EGFR-TKI during the study period was 165. Of these, 54 patients never underwent TKI dose reduction during the course of treatment (recommended dose group). On the other hand, 111 of the 165 patients (67.3%) experienced TKI dose reduction (dose reduction group).

Characteristics of the patients in these two groups are shown in Table I. The median BSA was 1.57 (range=1.2-1.95) in the recommended dose group but was significantly lower in the dose reduction group (1.47, range=1.18-1.96; p=0.029). The proportion of patients with a PS of 0-1 was 74.1% in the recommended dose group and in the dose reduction group it was significantly higher at 88.3% (p=0.026). The proportion of patients with a PS of 2-3 was 25.9% in the recommended dose group and 11.7% in the dose reduction group. The proportion of stage IV patients was significantly higher in the recommended dose group (85.2% of patients) than in the dose reduction group (64.9% of patients, p=0.010). The characteristics of the two groups of patients treated with osimertinib are shown in Table II.

Reason for dose reduction of TKI. The most common reason for dose reduction of TKI was side effects, including diarrhea in 22 patients, skin rash in 21 patients, loss of appetite in nine patients, and paronychia in six patients. Other reasons for dose reduction included advanced age in 16 patients, decreased PS in nine patients, and other reasons in 28 patients.

Survival in the two groups. There was no significant difference observed in PFS between the dose reduction and recommended dose groups (median PFS=9 months in the dose reduction group vs. median PFS=7 months in the recommended dose group, p=0.0834). On the other hand, OS was significantly longer in the dose reduction group than in the recommended dose group (median OS=20 months in the dose reduction group vs. median OS=17 months in the recommended dose group, p=0.011; Figure 1). In the analysis of patients aged 75 years or older with BSA <1.5 m², there was no significant difference in PFS or OS between the dose reduction group and the recommended dose group (median PFS, 9 vs. 10 months, respectively, p=0.878; median OS, 23 vs. 20 months, respectively, p=0.105; Figure 2). PFS was significantly longer in the osimertinib dose reduction group than in the recommended dose group (median PFS, 15 vs. 11 months, respectively, p=0.0482). On the other hand, there was no significant difference in OS (median OS, 23 vs. 17 months, respectively, p=0.158; Figure 3). A multivariate logistic regression analysis was performed with factors that had values of p<0.05 from the univariate analysis as explanatory variables and 3-year survival as the objective variable. As a result, TKI dose reduction was extracted as a favorable factor, with an odds ratio of 3.570 (p=0.028; Table III). Cox proportional hazards regression analysis also showed that being Stage IV was a risk factor with a hazard ratio of 1.56 (p=0.037). Whether the TKI dose was reduced or not was a favorable factor with a hazard ratio of 0.68 (p=0.046; Table IV).

This study revealed the following four points: 1) In the five types of EGFR-TKIs currently available, dose reduction was performed in 67.3% of the 165 TKI-administered patients researched; 2) In particular, dose reduction was performed in 40% of patients with the third-generation TKI, osimertinib; 3) The most common reasons for reducing the dose of TKI were side effects, low body weight, and old age. However, poor PS was not the reason for dose reduction; and 4) When comparing PFS and OS between patients who underwent dose reduction and those who did not, reducing the dose of TKI did not result in a disadvantage to survival. Furthermore, the OS was significantly longer in patients who underwent dose reduction than in those who did not.

Dose-finding studies of many antitumor drugs have not necessarily investigated large numbers of patients (18). The same was true for the clinical trials used to set the recommended doses of EGFR-TKIs, which were determined based on the results of 50-60 patients (5,8). Unlike conventional antitumor drugs, side effects of the most EGFR-TKIs are not dose-dependent (19). Therefore, except for the maximum tolerated dose determined for skin toxicity caused by erlotinib, the maximum tolerated dose has not been determined, and the recommended dose has been determined taking into account efficacy and side effects (7-9). EGFR-positive patients are evaluated to be predominantly younger women, and the median age of the EGFR-positive patients in most clinical trials was in the 50s (5-7). There have been attempts to improve the response rate of gefitinib, a first-generation TKI, by combining it with other antitumor drugs and taking into account gene polymorphisms (20,21). However, since gefitinib could only be administered in one dosage form, 250 mg tablet, it was not possible to reduce the dose during treatment. For second- and third-generation TKIs, a dose finding study have been attempted taking into account the site of metastasis (22). Subsequent studies in actual clinical practice have reported that EGFR-positive patients included elderly and underweight patients (11). Patients with these characteristics have rarely been included in clinical trials, and reducing the dose of TKIs in these patients seemed not to be anticipated. However, a closer look at the clinical trials of EGFR-TKIs reveal that dose reduction of TKIs occurs in 20%-30% of patients (4), and furthermore, in clinical practice, dose reduction occurs in 30%-40% of patients. In addition, this latter study found that appropriate dose reduction was not associated with poorer outcomes. Considering these results, there seems little need to hesitate to reduce the dose of TKIs when side effects occur, out of concern that it may shorten survival. Adherence to the recommended doses of TKIs might not be necessary, especially in elderly and underweight patients. Actually, since it is currently impossible to measure TKI blood concentrations in clinical practice, the appearance of a facial rash might also be used as a reference in clinical practice, as it is considered evidence that the TKI blood concentration is within the appropriate range. Also, in analyzing the difference in survival rates between all TKI treatments and osimertinib specifically, two factors stand out: osimertinib has been shown to provide a longer PFS than second-generation TKIs, and immune checkpoint inhibitor treatment has become available since osimertinib was introduced. Based on this, our considerations were as follows: Unlike second-generation TKIs, dose reduction of osimertinib due to side effects did not impair the duration of response, and if the dose reduction was appropriate, it might contribute to extending PFS. On the other hand, regarding OS, many EGFR-positive patients respond to immune checkpoint inhibitors after TKIs (23), and we speculated that this effect outweighed that of reducing the dose of TKIs.

Although this study was able to provide new findings as described above, it had several limitations: The first was that this was a retrospective study. It is desirable to investigate as many patients as possible, but even in this regard, caution is required. The next limitation to note is that the comparison of survival times was not performed using propensity matching to match patient backgrounds. It was determined that if this comparison had been performed, the study would have been able to come closer to more convincing evidence. However, we were able to show that there was no worsening of prognosis in the group that included a greater number of patients with poorer clinical conditions, such as older age or low body weight, and we evaluated that our initial objective was achieved. Although phase II trials have been conducted using TKI doses lower than the recommended levels for elderly patients (24), it is unlikely that phase I trials to establish the recommended dose of EGFR-TKIs, which have already been in clinical use for some time, will be conducted in the future. In that sense, it is important to conduct real-world clinical investigations including as many patients as possible. We believe that our results provide considerable insight into the future clinical application of EGFR-TKIs.

When side effects occur in patients with EGFR gene-mutated NSCLC in clinical practice, particularly in elderly or underweight patients, reducing the dose of TKI is a suitable option, with no cause for hesitation.

No funding was received.

The Authors declare that they have no competing interests in relation to this study.

AM, YM, KM, and HS designed this study. AM, HM, YM, SO, GO, SS, KM, TK, and HS collected data. AM, YM, KM, and HS analyzed the data. AM and HS prepared the manuscript. TS, YY, HS, and NH supervised this study. All Authors have approved the final version of the manuscript for submission.