Abstract
Background/Aim: Osimertinib blood levels and their impact on treatment continuation in patients with EGFR mutation-positive lung cancer is not known. This study investigated the drug blood levels and risk factors affecting treatment continuation. Patients and Methods: Fifty-six patients with recurrent and inoperable epidermal growth factor receptor (EGFR) mutation-positive non-small cell lung cancer who received Osimertinib (80 mg once daily, daily dose) between October 1, 2016, and August 31, 2021, were included. Patients were classified into two groups using a cutoff blood level of 155 ng/ml. The primary endpoint was the relationship between Osimertinib exposure and efficacy, and secondary endpoints were the relationship between Osimertinib exposure and side effects, and the effect of covariates on efficacy and blood levels. Results: The median progression-free survival (PFS) for evaluable patients in the steady-state trough concentration (Cmin ss) ≥155 ng/ml and Cmin ss <155 ng/ml groups was 18.7 months and 31.2 months. Serum albumin (Alb) levels were 3.73±0.40 g/dl and 3.93±0.28 g/dl (p=0.030), respectively, and in multivariate analysis, Alb <3.7 g/dl was associated with a hazard ratio of 5.304 (95%CI=1.431-19.66; p=0.013), indicating that Alb <3.7 g/dl significantly shortened PFS. Conclusion: Free blood concentration of Osimertinib may have been increased by a combination of factors, including decreased hepatic metabolic function and decreased albumin production caused by systemic inflammation in patients with cancer. However, there was no effect of Osimertinib Cmin ss on PFS.
Lung cancer is one of the leading causes of cancer-related mortality, with approximately 235,760 new cases and 131,880 deaths reported in the United States in 2021 (1). Gefitinib, an epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitor (EGFR-TKI), is commonly used for the treatment of patients with EGFR-positive advanced non-small cell lung cancer. However, treatment of patients with resistant mutations in the EGFR gene has been limited to the use of EGFR-TKIs. However, resistance to EGFR-TKI is reported to develop in most patients within 14 months of starting therapy, leading to disease progression (2). Various mechanisms of resistance to EGFR TKI have been validated. The most common mechanism is EGFR T790M mutation, which accounts for approximately 50% of all causes of EGFR TKI resistance in NSCLC patients (3, 4). This led to the use of third-generation TKI Osimertinib, which is potent against both resistant (T790M) and active (Del19, L585R) mutations (5).
Although Osimertinib is administered in fixed doses, its pharmacokinetic parameters have been reported to vary widely between patients, with Osimertinib area under the plasma concentration-time curve from drug administration to just before the next dose under steady state conditions (AUCss) ranging between 20-78% (6). Brown et al. reported no relationship between the exposure dose (AUCss) of Osimertinib and duration of response or tumor size reduction (7). In contrast, patients with higher blood Osimertinib levels have been reported to show a trend towards shorter overall survival (OS). Moreover, poor prognostic factors for treatment have not been identified (8, 9).
Furthermore, in the treatment of non-small-cell lung cancer with Osimertinib, some patients had to discontinue the drug due to side effects. Early detection and response to adverse effects are important factors for continuation of treatment and improvement of therapeutic efficacy of anti-cancer drugs. Brown et al. reported a correlation between AUCss and adverse effects (rash, diarrhea, and QT interval) (7). However, the AUC is difficult to implement in clinical practice because it requires blood sampling at multiple time points. Because the Css,max(maximum concentration)/Css,min(minimal concentration) of Osimertinib at steady state has been reported to be approximately 1.5 (6), and shows a flat blood concentration-time trend profile throughout the dosing interval, trough values may be an alternative indicator for AUCss. Therefore, the measurement of trough values can be used for early detection and response to adverse drug reactions. Moreover, it can also be used as a parameter for determining drug discontinuation and other reductions in treatment efficiency. In contrast, neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), Geriatric Nutritional Risk Index (GNRI), and Prognostic Nutritional Index (PNI) have been reported to affect TKI survival and adverse effects (10-13). Therefore, appropriate nutritional assessment is important for therapy continuation. However, the relationship between the blood levels of Osimertinib and nutritional indices has not yet been reported. Therefore, the aim of this study was to investigate Osimertinib exposure, efficacy, and toxicity in patients with EGFR mutation-positive lung cancer. We also aimed to explore the risk factors affecting the blood levels of Osimertinib.
Patients and Methods
Study population. Patients with recurrent and inoperable EGFR mutation-positive non-small cell lung cancer who visited Sapporo Minami-Sanjo Hospital between October 1, 2016, and August 31, 2021, and received Osimertinib (80 mg daily sequential doses) formed our study population. Patients from whom blood samples could not be obtained during the period when Osimertinib was being administered were excluded from this study. In addition, the analysis was based on the cut-off value of 155 ng/ml of blood Osimertinib concentration, which was calculated using the receiver operating characteristic (ROC) curve with reference to progression-free survival (PFS) in the international phase III study (AURA3 study). The patients were classified into two groups: those with blood trough concentration (Cmin ss) of 155 ng/ml or higher (Cmin ss≥155) and those with blood trough concentration of <155 ng/ml (Cmin ss<155).
Data collection. The primary endpoint of this study was the relationship between Osimertinib exposure and efficacy. The secondary endpoints were the relationship between Osimertinib exposure and side effects, and the effect of covariates on efficacy and blood levels. The efficacy endpoint was defined as PFS in patients treated with Osimertinib.
Data regarding the following parameters were collected from their medical records: age, sex, smoking status, EGFR mutation status, tumor stage, central nervous system (CNS) metastasis, treatment continuation status, weight loss, and previous drug therapy. PFS (cut-off date: 31-August-2021) was also calculated to evaluate the efficacy of Osimertinib. Nutritional and inflammatory statuses were assessed using albumin (Alb), NLR, LMR, GNRI, PNI, Body Mass Index (BMI), and C-reactive protein (CRP) at the time of blood sampling. Safety was evaluated for the following adverse events: acne rash, pruritic skin, skin dryness, nail ring infection, pneumonia, diarrhea, stomatitis, decreased appetite, elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels, thrombocytopenia, neutropenia, leukopenia, hemoglobin. The Common Terminology Criteria for Adverse Events (CTCAE) ver 5.0 was used to determine the severity of adverse events.
Collection of blood samples and measurement of Osimertinib serum concentrations. Blood samples were obtained from the residual portion of blood collected for clinical tests. Samples were centrifuged and serum was collected after centrifugation and stored at −20°C until assayed. Serum concentrations of Osimertinib were determined using a validated liquid chromatography method and tandem mass spectrometry (14). The calibration range was 10-1,000 ng/ml, with a lower limit of quantitation (LLOQ) of 10 ng/ml. Trough concentrations were estimated from an algorithm using population-pharmacokinetic parameters based on the last dose and blood collection times (7).
Statistical analysis. JMP® Pro ver. 16 (SAS Institute Inc., Cary, NC, USA) was used for statistical analyses. The t-test or Fisher’s direct probability test was used for the comparison of patients sociodemographic data, and the Kaplan-Meier method was used to compare PFS. Level of significance was set at p<0.05. Pearson’s correlation matrix was used to evaluate the correlation of Cmin ss with PFS and nutritional assessment index. For univariate and multivariate analyses, the Cox proportional hazards model was used to assess the impact of individual covariates on PFS. Explanatory variables used in the multivariate analysis included factors with p<0.05 in the univariate analysis. Multiple regression analysis was performed with Cmin as the objective variable to evaluate the covariates of blood levels.
This study was conducted in accordance with the Declaration of Helsinki and its amendments and the ethical guidelines for medical and health research involving human participants. Ethical approval was obtained from the Hokkaido Sapporo Minami-Sanjo Hospital (approval no. 19-06-021).
Since this was a prospective observational study, written or verbal consent was obtained from the study participants. All the study participants were informed about the study and assured of the opportunity to opt-in. All the patients provided written informed consent to participate in the study prior to the commencement of the study. We ensured that the confidential patient information was protected. Data were anonymized prior to processing.
Results
Study population. A total of 99 patients were treated with Osimertinib during the study period. Blood samples were not available for 43 patients, thereby leaving 56 patients for analysis. Twenty patients (35.7%) were categorized in the Osimertinib group with Cmin ss≥155 ng/ml and 36 (64.3%) belonged to the group with Cmin ss<155 ng/ml [Cmin ss<155] (Figure 1).
Flow chart of patient’s selection.
Patient characteristics. The mean age of the study participants was 73 years (range=67-77 years). The study population comprised 16 (29%) males and 40 (71%) females. Different types of EGFR mutations in the study population included: EX19 (19Del) in 24 (43%) patients, EX21 (L858R) in 28 (50%), and others in 4 (7%). Thirty three (59%) patients had stage IV cancer and recurrence was observed in 23 (41%) patients. CNS metastases occurred in 18 (32%) patients, weight loss was observed in 9 (16%) patients. Thirty five (63%) and 21 (38%) patients were in the 1st and 2nd lines, respectively. There were no differences in the patient characteristics between the two groups (Cmin ss≥155 and Cmin ss<155) before treatment initiation (Table I).
Background of analysis data.
Comparison of PFS in eligible patients. Thirteen (23.2%) and 18 (32.1%) patients with [Cmin ss≥155] and [Cmin ss<155], respectively, developed progressive disease while receiving Osimertinib. The Kaplan-Meier curves of PFS for both groups are shown in Figure 2. The median PFS was 18.7 months [95% confidence interval (CI)=15.5-21.9] and 31.2 months (95%CI=26.3-36.2) for patients with [Cmin ss≥155] and [Cmin ss<155], respectively. The PFS was significantly prolonged in the group with [Cmin ss<155] (Figure 2).
Progression-free survival (PFS) comparison by Cmin ss. Kaplan-Meier curve of PFS in patients treated with Osimertinib, in patients with median Cmin,pred below the population median of 155 ng/ml (solid line) and in patients with a median osimertinib Cmin ss pred ≥155 ng/ml (dotted line). Cmin ss pred: predicted trough plasma concentration.
Comparison of nutritional assessment parameters in patients. Serum Alb was 3.73±0.40 g/dl and 3.93±0.28 g/dl (p=0.030) in patients with [Cmin ss≥155] and [Cmin ss<155], respectively. The serum Alb levels were significantly higher in the [Cmin ss<155] group. No differences were observed in terms of NLR, LMR, GNRI, and PNI between the two groups (Table II). The correlation coefficient between Cmin ss and Alb was r=−0.408 (p=0.002) and that between Cmin ss and NLR was r=0.343 (p=0.010), indicating a correlation (Figure 3A and B).
Comparison of various parameters among patients taking osimertinib.
Correlations between Cmin ss and serum albumin and neutrophil-to-lymphocyte ratio (NLR). A) Correlation between Cmin ss and serum albumin levels. The r- and p-values provided above each graph represent Pearson’s correlation coefficient and the p-value, respectively. The serum albumin levels were strongly correlated with the Osimertinib Cmin (r=−0.408, p=0.009). B) Correlation between Osimertinib Cmin ss and NLR. The r- and p-values provided above each graph represent multiple regression analysis and the p-value, respectively. The NLR levels were correlated with the Osimertinib Cmin ss (r=−0.343, p=0.010).
Univariate and multivariate analysis. Covariates that may affect the efficacy of Osimertinib were sex, age, BMI, Cmin ss (≥155), Alb (<3.7 g/dl), NLR (≥3.9), LMR (<2.8), GNRI (<92), Stage IV, CNS metastases, CRP, and no prior therapy. Table III shows the results of univariate and multivariate Cox regression analyses of PFS. In univariate analysis, the factors that showed p<0.05 were Cmin ss≥155, Alb (<3.7 g/dl), CNS metastasis, and CRP. In multivariate analysis, CNS metastasis and Alb <3.7 g/dl had hazard ratios (HRs) of 3.59 (95%CI=1.35-9.57; p=0.01) and 5.30 (95%CI=1.43-19.7; p=0.01), respectively, and these factors were shown to result in shorter PFS. In contrast, Cmin ss and CRP levels did not differ significantly in relation to PFS. Multiple regression analysis with Cmin as the objective variable yielded a slope of −119.1 and standardized regression coefficient of −0.350 (p=0.009). Additionally, the multiple correlation coefficient adjusted for degrees of freedom was 0.189 (Figure 3A). Thus, Alb was identified as an independent factor for predicting the Cmin ss of Osimertinib.
Univariate and multivariate Cox-proportional hazards model progression-free survival.
Incidence and frequency of adverse drug reactions in the patients. In total, 56 patients experienced adverse drug effects. Adverse reactions that occurred in the eligible patients are shown in Table IV. Acne-like skin rash occurred in 20 (35.7%) patients, pruritic skin in 8 (14.3%), dry skin in 6 (10.7%), onychomycosis in 20 (35.7%), pneumonia in 5 (8.9%), diarrhea in 24 (42.9%), stomatitis in 13 (23.2%), decreased appetite in 2 (3.6%), increased ALT in 8 (14.3%), and increased AST in 10 (17.9%), Nausea in 2 (3.6%), thrombocytopenia in 13 (23.2%), neutropenia in 5 (8.9%), leukopenia in 17 (30.4%), hemoglobin (Hb) decreased in 13 (23.2%). Serious adverse reactions of Grade 3 or higher occurred as diarrhea in one patient each with [Cmin ss≥155] and [Cmin ss<155]. Stomatitis, thrombocytopenia, and anemia were observed in one patient each with [Cmin ss<155] (Table IV).
Adverse events.
Discussion
This study investigated the risk factors affecting PFS and blood levels of Osimertinib in 56 patients who were treated with 80 mg of daily dose of Osimertinib. The cut-off trough values in exposure-response analysis has been reported to be 166 ng/ml, which is the geometric mean of trough concentrations calculated from doses mentioned in several previous studies (8, 9). However, this is not supported by efficacy, and the pharmacokinetic threshold for Osimertinib remains exploratory. Therefore, in this study, to explore the possibility of a trough concentration associated with efficacy, we used a value of 155 ng/ml, calculated using ROC curves, dividing the eligible patients into two groups using 10.1 months of PFS from the international phase III study (AURA3 study) as the boundary (15).
PFS was significantly prolonged in patients with [Cmin ss<155] compared to those with [Cmin ss≥155] (18.7 months vs. 31.2 months; p=0.04) (Figure 2). Alb levels were also significantly higher in the [Cmin ss<155] group (3.7±0.4 g/dl vs. 3.9±0.3 g/dl; p=0.04) (Table II). However, multivariate analysis results indicated that low Alb levels may result in progression of non-small cell lung cancer, as CNS metastasis and Alb <3.7 g/dl were identified as factors shortening PFS (Table III). Therefore, high blood levels of Osimertinib (Cmin ss≥155) are unlikely to affect PFS, as CNS is the most common site of distant metastases in advanced non-small cell lung cancer, which severely affects the clinical presentation and quality of life of these patients. We found that CNS metastasis is one of the main prognostic factors in patients with non-small cell lung cancer, which was in agreement with the findings of Boosman et al. (9).
A significant negative correlation was observed between Osimertinib Cmin ss and Alb [r=−0.408 (p=0.002)] (Figure 3A). Moreover, Alb was also confirmed as an independent factor predicting Osimertinib Cmin ss in multiple regression analysis. Fiala et al. reported that hypoalbuminemia is associated with poor outcomes in patients with non-small cell lung cancer treated with erlotinib (16). Osimertinib reportedly has a high protein-binding rate and high affinity for Alb. Moreover, it has also been reported that Osimertinib-Alb conjugates remain in the human plasma for approximately 21 days after the oral administration of the drug (17). Furthermore, the free drug concentration at the therapeutic target site is considered to be correlated with the pharmacological effects of Osimertinib (18). Liu et al. and Dickinson et al. reported that covalent binding between Osimertinib and human plasma Alb is irreversible (19, 20). This suggests that, if a patient is hypoalbuminemic, the amount of Alb to which Osimertinib binds will decrease, thereby increasing the amount of free Osimertinib. Yokota et al. also reported a significant negative correlation between Osimertinib AUC (0-24) and serum Alb levels, with serum Alb level being the only independent factor predicting Osimertinib AUC (0-24) in multiple regression analysis (21). Therefore, the amount of Osimertinib distributed in the body may be significantly affected by the serum Alb concentration. It is conceivable that an increase in the release rate may also increase the hepatic intrinsic clearance, thereby preserving metabolic function in the liver. However, blood levels have been reported to be elevated in patients with hypoalbuminemia. This may be due to the effects of inflammation on metabolic functions.
High free blood levels of Osimertinib may have a negative impact on patient prognosis due to off-target genetic effects. The off-target effect is a phenomenon in which a molecularly targeted drug inhibits or activates another molecule (off-target) that is different from its original target (on-target) (22). This effect has been identified as a factor contributing to Osimertinib resistance in patients with non-small cell lung cancer (20). A study on dabrafenib/trametinib in patients with BRAF-mutant metastatic melanoma reported that high concentrations of dabrafenib may have adverse pharmacodynamic effects on some kinases, thus reducing the efficacy of drug therapy, including OS (23). Rodier et al. also reported that OS in the Osimertinib group with high Cmin ss was approximately twice that in the group with low Cmin ss (8). In this context, the high blood concentration of Osimertinib in patients with low Alb concentrations may have induced an off-target effect, causing a reduction in pharmacological efficacy. However, the mechanism by which the off-target effect is induced at high concentrations has not been clarified.
The reason for the high trough concentration of Osimertinib may be the decrease in Cytochrome (CYP) activity associated with cancer inflammation. It has been reported that elevated CRP levels decrease the activity of P450 metabolic enzymes (24, 25). Moreover, it also reduces the activity of CYP3A4, the enzyme that metabolizes Osimertinib. In the present study, CRP was significantly higher in patients with [Cmin ss≥155] (Table II). In this context, a decrease in CYP enzyme activity would increase the amount of Osimertinib remaining in the blood, which would also affect Osimertinib exposure. NLR, like CRP, is also one of the most highly regarded indicators of systemic inflammation (26). However, no studies have evaluated the causal relationship between NLR and metabolic enzymes. In the present study, we found a positive correlation between Cmin ss and NLR of 0.343 (p=0.010). Moreover, we also found that Cmin ss increased with increasing inflammatory response (Figure 3B). Therefore, whether a new inflammatory biomarker, such as NLR, can be used as an indicator of its effect on pharmacokinetics needs to be examined.
The reason why Osimertinib was not a factor in shortening survival despite its increased blood levels in the present study may be due to the decreased metabolic enzyme activity caused by inflammation, which may have decreased the apparent drug clearance, thereby increasing blood levels independent of efficacy. Therefore, it is likely that there was no association between blood levels of this drug and PFS.
Contrastingly, exposure to Osimertinib was not significantly associated with the occurrence of toxicity (Table IV). Brown et al. reported a relationship between increased steady-state AUC and increased risk of adverse events such as skin rash, diarrhea, and QTc prolongation (7). In contrast, Rodier et al. reported no correlation between exposure dose and adverse events, which is in concordance with our study (8). However, given that safety was reported at a dose of 160 mg/day in a previous phase II study and that only 17% of the patients were dose-adjusted (27, 28), it seems unlikely that dose-limiting toxicity would occur at the current dose.
This study has a few limitations. First, it was a single-center study. Therefore, the number of patients included in the study was relatively small and multiple samples could not be collected from all the patients, which made it difficult to follow the changes over time. In future, it will be necessary to establish a system for collecting multiple specimens. Second, competitive inhibition of CYPs by multiple drugs was not examined in this study as there were no applicable patients.
Conclusion
In this study, the blood levels of Osimertinib may have increased due to a combination of factors, including decreased hepatic metabolic function and decreased Alb production caused by systemic inflammation in cancer patients. Therefore, it can be considered that Alb levels influence the blood concentration of Osimertinib. The more hypoalbuminemic a person is, the higher is their blood concentration of Osimertinib. In contrast, Osimertinib Cmin ss did not affect PFS. Further evaluation of the relationship between serum Alb and blood Osimertinib levels is warranted.
Footnotes
Authors’ Contributions
HY conceived the idea of the study and drafted the original manuscript. SH made significant contributions to the design of the study and the interpretation of data. SH, TT, IK, and OK contributed to the interpretation of the results. HT, HT, and WA supervised the conduct of this study. SH and other Authors substantially contributed to the revision of the manuscript drafts. All Authors have approved the submitted version of the manuscript and agreed to be accountable for any part of the study.
Conflicts of Interest
There are no conflicts of interest to declare in relation to this study.
- Received June 26, 2023.
- Revision received July 24, 2023.
- Accepted July 27, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).