Abstract
Background/Aim: Alectinib is recommended for anaplastic lymphoma kinase fusion gene-positive non-small cell lung cancer. We have experienced early alectinib discontinuation due to disease progression and adverse effects in real world. Because alectinib has a high protein-binding rate of >99%, low serum albumin may increase the concentration of free drug and affect efficacy and adverse events. However, no association between serum albumin and the clinical impact of alectinib has been reported. The purpose of this study was to determine the effect of serum albumin on time-to-treatment failure (TTF) in alectinib. Patients and Methods: Fifty-six patients who were admitted to four hospitals (National Hospital Organization Hokkaido Cancer Center, Sapporo Minami-Sanjo Hospital, KKR Sapporo Medical Center, Otaru General Hospital) between October 2014 and September 2020 were retrospectively evaluated to identify those treated with alectinib. Results: The multivariate analysis showed that the risk of discontinuation was significantly higher with serum albumin <3.6 g/dl compared to ≥3.6 g/dl at the start of alectinib administration (hazard ratio=3.00; 95% confidence interval=1.36-6.66; p<0.01). On Kaplan-Meier curves, TTF for serum albumin <3.6 was significantly shorter than that for ≥3.6. (median TTF: 12.1 months vs. not reach, p<0.01). Conclusion: To the best of our knowledge, this study is the first to report that serum albumin <3.6 g/dl at alectinib induction is associated with poor TTF. Low serum albumin is a poor prognostic factor in cancer patients. Thus, serum albumin levels must be measured before treatment.
Anaplastic lymphoma kinase (ALK) fusion genes are found in approximately 5% of non-small cell lung cancers and are characteristic of adenocarcinoma (1). ALK tyrosine kinase inhibitors (ALK-TKIs) such as crizotinib, alectinib, ceritinib, lorlatinib, and brigatinib inhibit the growth of ALK fusion gene positive tumor cells by inhibiting ALK tyrosine kinase activity (2-10). Alectinib has selective kinase inhibitory activity against ALK and tumor growth inhibitory effects even in cell lines with crizotinib-resistant mutations (11, 12). In addition, alectinib confers significantly longer progression-free survival (PFS) than standard chemotherapy (pemetrexed and docetaxel) as second-line therapy after crizotinib failure (ALUR study) (6) and significantly longer PFS compared to crizotinib as first-line therapy [ALESIA study (10), ALEX study (4), J-ALEX study (5)]. Based on these results, alectinib is recommended in the National Comprehensive Cancer Network guidelines (version 2022.3) for the first-line therapy of ALK fusion gene-positive non-small cell lung cancer and second-line therapy after crizotinib failure (13). Alectinib has a median (range) time-to-treatment failure (TTF) of 482 days (range=1-775 days) (14). We have experienced cases of early alectinib discontinuation owing to disease progression and adverse effects, in clinical practice. Masuda et al. reported Eastern Cooperative Oncology Group performance status (ECOG PS) ≥3, prior crizotinib administration, and presence of brain metastases as factors for alectinib TTF failure (14). However, because alectinib is orally administered, increased risk of potential drug-drug interactions and patient-to-patient pharmacokinetic variability (absorption, distribution, metabolism, and excretion) have increased the incidence of adverse events and decreased therapeutic efficacy (15). Because alectinib has a high protein-binding rate of >99% (16), low serum albumin or concomitant use of drugs with high protein binding rates may increase the concentration of free drug and affect therapeutic efficacy and the incidence of adverse events. However, no association between serum albumin and the clinical impact of alectinib has been reported. The purpose of this study was to determine the effect of serum albumin on TTF in patients receiving alectinib.
Patients and Methods
Patients. The cooperative study groups included four hospitals (National Hospital Organization Hokkaido Cancer Center, Sapporo Minami-Sanjo Hospital, KKR Sapporo Medical Center, and Otaru General Hospital). We retrospective analyzed 62 patients with recurrent and unresectable ALK fusion gene-positive lung adenocarcinoma treated with alectinib at a dose of 300 mg twice daily between October 2014 and September 2020 at each hospital. Six patients were excluded due to missing data, and the final analysis was performed on 56 patients (Figure 1).
Study flow.
Data collection. Patient data collected included sex, age, number of treatments, ECOG-PS, the presence of central nervous system (CNS) metastases, body surface area (BSA) calculated from the Du Bois formula, body mass index (BMI), serum albumin, serum creatinine, the number of combined medications, and the incidence of adverse events. The cutoff value for the number of combined medications was set to 6 because polypharmacy (6 or more medications) has been reported to increase the incidence of adverse events (17).
Assessments. The primary endpoint was TTF, and the secondary endpoint was the incidence of adverse events. TTF was defined as the time from the date of alectinib initiation to the date the physician determined that alectinib should be discontinued or the study cut-off date (September 30, 2020). Adverse events were evaluated in accordance with the National Cancer Institute’s Common Terminology Criteria for Adverse Events (version 5.0) (18).
Statistical analysis. Baseline patient backgrounds were summarized as medians and interquartile range for continuous variables, and number and percentages for categorical variables. Kaplan-Meier estimate were used to assess TTF. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were estimated using Cox proportional-hazard model. Cox proportional-hazard models were used for univariate and multivariate analyses. Multivariate analysis was applied for items with p<0.2 in the univariate analysis. Serum albumin was analyzed using receiver operating characteristic (ROC) analysis to determine the cutoff value for treatment discontinuation. Differences were considered significant at levels below 5%. The statistical software used was the Bell Curve for Excel (Social Survey Research Information Co., Ltd., Tokyo, Japan).
Ethical statement. This study was performed in compliance with the ethical guidelines for medical research on human subjects. It was approved by the Ethics Committee of Hokkaido Cancer Center (Approval No: 02-38), Sapporo Minami-Sanjo Hospital (Approval No: R3-2), KKR Sapporo Medical Center (Approval No: 2020-43), and Otaru General Hospital (Approval No: 03-004). Owing to its retrospective nature, no written or oral consent was obtained from the research subjects. Information about the study was made available to the research subjects (posted in the hospital or on the hospital website), and the research subjects were guaranteed the opportunity to refuse to have the study conducted. We ensured that all patient confidential information were protected. The data were anonymized prior to handling.
Results
Baseline patient characteristics. The backgrounds of the 56 patients are presented in Table I. Patients’ demographics and medical history were as follows: (i) sex: 18 males (32%), 38 females (68%); (ii) median age: 67 (quartiles: 57 and 74) years; (iii) ECOG-PS at alectinib initiation: 0, 36 patients (64%); (iv) BSA at alectinib initiation: 1.54 m2 (quartiles: 1.43 and 1.65 m2); (v) BMI at alectinib initiation: 22 kg/m2 (quartiles: 20 and 25 kg/m2); (vi) the presence of CNS metastases: 15 patients (27%); (vii) number of 6 or more concomitant medications at alectinib initiation: 14 patients (25%); and (viii) alectinib treatment line at initiation: first-line: 34 patients (61%); second-line or later: 22 patients (39%) (Table I). Alectinib was continued in 30 patients (54%) and discontinued in 26 patients (46%). The reasons for discontinuation were disease progression in 20 patients (78%), skin rash in 3 patients (11%), and pneumonia in 3 patients (11%). The median TTF was 48 months (Figure 2).
Characteristics of patients treated with alectinib.
Kaplan-Meier curve for time-to-treatment failure in patients treated with alectinib (N=56) showed a median time-to-treatment failure of 48 months.
Effect of baseline patient characteristics on time-to-treatment failure. The factors affecting the TTF are shown in Figure 3. The cutoff of serum albumin was calculated using the receiver operating characteristic (ROC) curve (cutoff 3.6, AUC=0.64, p=0.04). Univariate analysis by Cox proportional hazard models showed that the risk of discontinuation was significantly higher with serum albumin <3.6 g/dl than ≥3.6 g/dl at alectinib initiation (HR=2.71; 95%CI=1.24-5.95; p=0.01), and the risk of discontinuation was likely to associate with higher age ≥75 years compared to <75 years at alectinib initiation (HR=0.45; 95%CI=0.15-1.31; p=0.14). The multivariate analysis showed that the risk of discontinuation was significantly higher with serum albumin <3.6 g/dl compared to ≥3.6 g/dl at alectinib initiation (HR=3.00; 95%CI=1.36-6.66; p<0.01), and the risk of discontinuation was likely to associate with higher age ≥75 years compared to <75 years at alectinib initiation (HR=0.39; 95%CI=0.13-1.14; p=0.09). The TTF was compared for serum albumin and age using Kaplan-Meier curves (Figure 4). Less than 3.6 g/dl serum albumin was associated with a significantly shorter TTF compared to ≥3.6 g/dl (median TTF; 12.1 months vs. NR, p<0.01). However, TTF was not significantly different for age (median TTF; NR vs. 35.4 months, p=0.13).
Univariate and multivariable analysis of time-to-treatment failure according to baseline patient characteristics. Serum albumin <3.6 g/dl compared to ≥3.6 g/dl at alectinib administration (HR=2.93, 95%CI=1.32-6.51, p=0.01).
Kaplan-Meier curves for time-to-treatment failure (TTF) according to albumin (Alb) (A) and age (B) in patients treated with alectinib. A) Alb <3.6 g/dl vs. Alb ≥3.6 g/dl, median TTF 12.1 months vs. not reached (NR), p<0.01. B) Age ≥75 years vs. age <75 years, median TTF NR vs. 35.4 months, p=0.13.
Adverse events. The incidences of adverse events are listed in Table II. More than 5% of the incidence of all-grade adverse events were observed in 16 patients (29%) with increased aspartate aminotransferase (AST), 10 patients (18%) with increased alanine aminotransferase (ALT), 12 patients (21%) with increased total bilirubin, 13 patients (23%) with skin rash, and 5 patients (9%) with pneumonia. The incidence of grade ≥3 adverse events were skin rash in 2 patients (4%) and pneumonia in 2 patients (4%). The number of patients who discontinued due to the incidence of adverse events was 3 (5%) with skin rash and 3 (5%) with pneumonia. Serum albumin in the group of patients who discontinued treatment owing to the incidence of adverse events was not significantly different from that in the group of patients who discontinued treatment owing to disease progression (serum albumin median 4.0 (quartiles: 3.6 and 4.1) vs. serum albumin median 3.7 (quartiles: 3.3 and 4.1), p=0.58).
The incidence of adverse events with alectinib.
Effect of adverse events on time-to-treatment failure. The effect of the incidence of adverse events on TTF is shown in Figure 5. Univariate analysis using Cox proportional hazard models showed that increased ALT, AST, total bilirubin, and skin rash were not significantly different.
Univariate analysis of time-to-treatment failure in adverse event (not significantly different results).
Discussion
The purpose of this study was to determine the effect of serum albumin on the TTF of alectinib. The results of this study showed that a serum albumin of less than 3.6 g/dl at the start of treatment was associated with a shorter TTF.
First, we investigated the possibility of serum albumin as a risk factor for discontinuation of alectinib administration (Figure 2). In this study, a serum albumin of less than 3.6 g/dl was an independent risk factor for discontinuation. Since alectinib has a high protein binding rate of 99% (16), a decrease in serum albumin would be expected to increase the free drug concentration of alectinib. However, because alectinib has a distribution volume of 475 l and a hepatic extraction ratio of 0.63 (16), the apparent clearance of total drug concentration may be increased and the blood concentration may be decreased, and free drug concentration may not be increased. Serum albumin is synthesized in the liver and is commonly used as a nutritional indicator reflecting relatively long-term nutritional status with a long biological half-life of 14 to 21 days. Low serum albumin indicates a state of low nutrition. Cancer patients are more likely to be undernourished due to the incidence of adverse events and thus have a poorer prognosis (19). Low serum albumin has been reported to be a poor prognostic factor (20, 21). Low serum albumin in patients with esophageal and gastric cancers (22), and patients with recurrent or metastatic gastric cancer has been associated with shorter overall survival (23). In addition, high C-reactive protein and low serum albumin have been reported to be poor prognostic factors in patients with metastatic breast cancer (24). These reports support the results of this study, in which low serum albumin was a risk factor for discontinuation of alectinib treatment.
Next, we investigated the effect of treatment line on TTF at alectinib initiation (Figure 3). In clinical trials, the PFS with alectinib was reported to be 34.1 months as first-line therapy and 9.6 months as second-line therapy after crizotinib failure (5, 6). In real world, Yingying et al. reported significantly prolonged PFS with first-line therapy compared with second-line therapy, and Masuda et al. found that patients previously treated with crizotinib had a shorter TTF (14). In this study, treatment line at alectinib initiation was not a risk for discontinuation. In addition, TTF with prior crizotinib treatment was not significantly different between with and without (30.6 months vs. 30.8 months, data not shown). The reasons that treatment line and prior crizotinib treatment were not a risk for discontinuation in this study could be as follows. First, PFS is defined as the time from the start of treatment to progression, while TTF reflects not only the time-to-progression, but also the time-to-discontinuation of treatment due to the incidence of adverse events or patient convenience. In the present study, the reasons for treatment discontinuation or interruption in patients receiving first line therapy were progression in 10 patients (29%) and adverse events in 5 patients (15%), and progression in 10 patients (45%) and adverse events in 1 patient (5%) among patients on second-line or later therapy (data not shown). Since a large proportion of discontinuations in the first-line treatment group were due to the incidence of adverse events, TTF was not affected as much as the difference in PFS by the number of treatment lines. Second, as opposed to PFS, TTF should also consider the impact of “beyond progressive disease (PD)”. “Beyond PD” is defined as continuing TKI treatment for as long as possible with additional local therapy in “oligoprogressive disease”, in which TKI failure is limited to bone or brain metastases and responses in other disease sites are maintained; also, prolonged PFS has been reported (25, 26). Since “beyond PD” is performed regardless of treatment line, TTF is longer than PFS. Of the 20 patients in this study who discontinued treatment because of PD, “beyond PD” was used in four patients (first-line, three patients; second-line or later, one patient), and was unlikely to have contributed to the prolongation of TTF. Therefore, “beyond PD” is unlikely to affect results where the treatment line was not a risk factor for treatment discontinuation.
Next, we investigated the incidence of adverse events as a risk factor for treatment discontinuation (Figure 5). There was a difference in the dose of alectinib: 300 mg twice daily in the J-ALEX study (5) and 600 mg twice daily in the ALEX (4) and ALESIA (10) studies. In the ALEX study, grade ≥3 adverse events were observed in 52.0% of patients; adverse events leading to dose reduction were observed in 20.4%; adverse events leading to withdrawal of treatment were observed in 26.3%; and adverse events leading to treatment discontinuation were observed in 4.5% (4). In the ALESIA study, adverse events leading to death, discontinuation, dose reduction, and withdrawal treatment were observed in 2%, 7%, 24%, and 26% of patients, respectively (10). However, in the J-ALEX study, grade ≥3 adverse events leading to treatment discontinuation occurred in 7% of patients, adverse events leading to withdrawal of treatment were observed in 34% of patients, but no deadly adverse events were reported (5). In this study, adverse events leading to treatment discontinuation were observed in 22% of patients, which was similar to and consistent with the results reported in the J-ALEX study. In addition, these reports indicate that the incidence of adverse events depends on the dose, which is a risk factor for treatment discontinuation. For the association between the incidence of adverse events and blood concentration, increased AST has been reported as a parameter that contributes to increased blood-drug concentration in the population pharmacokinetic analysis of AF-001JP (27). A dose-dependent increase in the incidence of some adverse events (hepatic dysfunction and skin rash) has also been reported (27). In this study, no serious hepatic dysfunction leading to treatment discontinuation occurred, and skin rash was observed in three patients (5%). Although we consider that patients who developed skin rash may have had elevated blood-drug concentration, we consider it to be a limitation of this study as the blood-drug concentrations were not measured in this study and are therefore unknown.
The other limitations of this study include that although the study was conducted in a multicenter setting because of the rarity of ALK fusion gene-positive non-small cell lung cancer, only a small number of patients could be studied, and it was retrospective in nature. In addition, because no patients in this study were taking concomitant drugs with high protein binding rates, we were unable to examine the clinical impact of drug interactions based on protein binding substitution. The association between the incidence of adverse events and TTF requires further case accumulation, as there were only a few cases with grade ≥3 adverse events. We suggest that patients who developed grade ≥3 adverse events should also have their blood-alectinib concentration measured to clarify the clinical impact of elevated blood-drug concentration.
This study is the first to report that serum albumin of less than 3.6 g/dl at the time of alectinib initiation is associated with poor TTF. These results suggest that it is important to measure the serum albumin before initiating treatment with alectinib.
Footnotes
Authors’ Contributions
All Authors discussed the results and commented on the manuscript. K.U., K.Y., K.G., T.H., T.H., M.I., S.T., K.Y., K.M. and H.S. confirmed medical diagnosis and designed this study. K.U., K.Y., S.T. and H.S. provided advice on statistical analysis. K.U., K.Y., K.G., T.H., T.H., M.I., S.T. and H.S. edited the manuscript.
Conflicts of Interest
The Authors declare that there are no conflicts of interest in relation to this study.
- Received May 23, 2023.
- Revision received June 15, 2023.
- Accepted June 16, 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).