2022 Volume 45 Issue 3 Pages 333-338
Proteinuria is one of the most frequently reported adverse events leading to the discontinuation of lenvatinib treatment in patients with advanced hepatocellular carcinoma (HCC). However, there are no reports regarding the risk factors of proteinuria in patients with HCC or patients receiving lenvatinib. We retrospectively reviewed the medical records of patients with HCC receiving lenvatinib at the Kobe City Medical Center General Hospital between April 2018 and December 2020. The severity of proteinuria was graded based on the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. A multivariate Cox proportional hazards model was employed to identify the risk factors of developing grade ≥2 proteinuria. Among the 37 patients included, 3 patients had grade-1 proteinuria at baseline and 10 patients had estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 at baseline. Grades 1, 2, and 3 proteinuria were observed in 15 (40.5%), 10 (27.0%), and 2 (5.4%) patients, respectively, during lenvatinib treatment. The median value of eGFR was significantly lower in patients who developed grade ≥2 proteinuria than those with grade ≤1 proteinuria (59.6 vs. 78.1 mL/min/1.73 m2, p = 0.045). Multivariate analysis revealed that pre-existing proteinuria at baseline (hazard ratio (HR), 9.72; 95% confidence interval (CI), 1.29–52.21; p = 0.030), and eGFR <60 mL/min/1.73 m2 at baseline (HR, 4.49; 95% CI, 1.32–16.07; p = 0.017) were significantly associated with developing grade ≥2 proteinuria. These patients should be monitored carefully, and our preliminary data should be confirmed by further studies.
Hepatocellular carcinoma (HCC) is one of the most common types of liver cancers and is the second most common cause of cancer-related death worldwide.1) It ranks fifth among the causes of death from cancer in Japan.2)
Lenvatinib, an oral multi-kinase inhibitor of vascular endothelial growth factor receptor 13, platelet-derived growth factor receptor α, fibroblast growth factor receptor 1-4, and RET and KIT proto-oncogenes, demonstrated non-inferiority to sorafenib in terms of overall survival and progression-free survival among patients with advanced HCC, and it was approved in Japan for the treatment of advanced HCC in March 2018.3) However, significant adverse drug events (ADEs) are associated with lenvatinib, including hand-foot syndrome, hypertension, fatigue, diarrhea, and proteinuria.3–6) In a phase-II trial of Asian patients with HCC, proteinuria was the most frequently reported ADE leading to drug discontinuation.4)
An understanding of the predictive factors leading to proteinuria in patients receiving lenvatinib is important for managing this ADE. However, this has not yet been completely elucidated. To date, only three studies on the risk factors of developing proteinuria in patients with metastatic renal cell carcinoma (RCC) receiving oral vascular endothelial growth factor (VEGF) receptor-tyrosine kinase inhibitors (VEGFR-TKIs) such as axitinib, sorafenib, and sunitinib have been published.7–9) However, no data are available regarding the risk factors of proteinuria in patients with HCC or patients receiving lenvatinib. Therefore, in this study, we investigated the risk factors of proteinuria development in patients with HCC receiving lenvatinib.
This retrospective study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Kobe City Medical Center General Hospital (Approval No. zn210301). Patients were eligible if they were ≥20 years of age, diagnosed with HCC, and started lenvatinib monotherapy at Kobe City Medical Center General Hospital between April 1, 2018, and December 31, 2020. Patients received oral lenvatinib 12 mg/d (for patients with body weight ≥60 kg) or 8 mg/d (for patients with body weight <60 kg) once daily. Dose reduction was performed at the physician’s discretion. The primary objective of this study was to evaluate the risk factors for developing grade ≥2 proteinuria in patients with advanced HCC receiving lenvatinib.10–17) The secondary objective was to determine the association between risk factors and cumulative incidence of proteinuria. The exclusion criteria were a treatment period of less than 3 weeks after starting lenvatinib treatment, and patients with baseline proteinuria ≥2+, according to urine dipstick testing.
Data Collection and AssessmentBaseline blood and urine tests were conducted within one week before the start of lenvatinib treatment. For safety monitoring, urine dipstick tests were conducted at least twice in the first 4 week after the start of lenvatinib treatment, then at least every 4 weeks. Data on pre-existing proteinuria, age, sex, weight, Eastern Cooperative Oncology Group performance status, serum creatinine, estimated glomerular filtration rate (eGFR), systolic blood pressure (SBP), medications that affect blood pressure, comorbid with diabetes, and prior treatments were retrospectively collected from the electronic medical records. The eGFR was calculated using the formula developed by the Japanese Society of Nephrology.18) The best tumor response was evaluated using computed tomography scans as complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD), based on the Response Evaluation Criteria in Solid Tumors guideline version 1.1.19) The overall response rate (ORR) was defined as CR + PR, and disease control rate (DCR) as CR + PR + SD.
The severity of axitinib-associated proteinuria was graded based on the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. The data cut-off date was May 31, 2021.
Statistical AnalysisCategorical data are presented as the number of patients (percentage) and were compared between groups using chi-squared test or Fisher’s exact test, as appropriate. Continuous data are presented as median (interquartile range), and Mann–Whitney U test was used to compare the groups. Univariate and multivariate Cox proportional hazards models were used to identify the risk factors for developing grade ≥2 proteinuria. We considered that eGFR <60 mL/min/1.73 m2 indicated decreased kidney function based on Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guidelines.20) Variables with a p-value <0.05 in the univariate analyses were evaluated as potential covariates in the multivariate analysis. The cumulative incidence of grade ≥2 proteinuria was described using Kaplan–Meier method with log-rank test. All statistical analyses were performed using JMP 13.2.1 (SAS Institute Inc., Cary, NC, U.S.A.) and EZR 1.41 (Saitama Medical Center, Jichi Medical University, Saitama, Japan) softwares.21) EZR is a graphical user interface for R (R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics. Statistical significance was defined as a two-tailed p-value <0.05.
Between April 2018 and December 2020, 43 consecutive patients with advanced HCC started treatment with lenvatinib, of which six patients were excluded because their treatment period was less than 3 weeks after starting lenvatinib treatment (n = 4), or proteinuria ≥2+, according to urine dipstick testing at baseline (n = 2). The remaining 37 patients were the study subjects. The baseline characteristics of the patients are presented in Table 1. The patients consisted of 29 (78.4%) men and eight (21.6%) women. Three patients (8.1%) had proteinuria >1 + at baseline, according to urine dipstick testing. Most patients (n = 25) received lenvatinib as the first TKI treatment, followed by the second (n = 5) and third (n = 7) TKI treatments. Twenty-five patients received a standard dose of lenvatinib, whereas 12 patients had started a one-level dose down of lenvatinib. The median concentration of albumin was significantly higher in patients who developed grade ≥2 proteinuria than those with grade ≤1 proteinuria (3.9 vs. 3.4 g/dL, p = 0.018). The proportion of patients with albumin-bilirubin (ALBI) grades was significantly different between patients who developed grade ≥2 proteinuria and those with grade ≤1 proteinuria (p = 0.048). The median value of eGFR was significantly lower in patients who developed grade ≥2 proteinuria than those with grade ≤1 proteinuria (59.6 vs. 78.1 mL/min/1.73 m2, p = 0.045). In addition, median duration of lenvatinib treatment was significantly longer in patients who developed grade ≥2 proteinuria than those with grade ≤1 proteinuria (8.1 vs. 1.9 months, p = 0.007). There were no other remarkable differences in patient characteristics between the groups.
| Proteinuria | p-Values | ||
|---|---|---|---|
| Grades 2–3 (n = 12) | Grades 0–1 (n = 25) | ||
| [Patient characteristics] | |||
| Age (years), median (IQR) | 77 (66–82) | 74 (67–80) | 0.570 |
| Male sex, n (%) | 11 (91.7%) | 18 (72.0%) | 0.232 |
| Weight (kg), median (IQR) | 68.5 (61.5–72.2) | 61.5 (53.1–73.6) | 0.277 |
| ECOG PS, n (%) | |||
| 0 | 8 (66.7%) | 11 (44.0%) | |
| 1 | 4 (33.3%) | 11 (44.0%) | |
| 2 | 0 (0%) | 3 (12.0%) | |
| Etiology, n (%) | |||
| Hepatitis B | 1 (8.3%) | 6 (24.0%) | 0.245 |
| Hepatitis C | 5 (41.7%) | 13 (52.0%) | |
| Others | 6 (50.0%) | 6 (24.0%) | |
| Child–Pugh score, n (%) | |||
| 5–6 | 12 (91.7%) | 21 (84.0%) | 1.000 |
| 7–8 | 1 (8.3%) | 4 (16.0%) | |
| BCLC staging, n (%) | |||
| B | 2 (16.7%) | 2 (8.0%) | 0.582 |
| C | 10 (83.3%) | 23 (92.0%) | |
| Portal vein invasion, n (%) | 1 (8.3%) | 7 (28.0%) | 0.232 |
| Extrahepatic metastasis, n (%) | 7 (58.3%) | 12 (48.0%) | 0.728 |
| SBP (mmHg) | 124 (113–130) | 120 (110–129) | 0.432 |
| Use of antihypertensive agents, n (%) | 8 (66.7%) | 14 (56.0%) | 0.724 |
| RAS inhibitors | 5 (41.7%) | 8 (32.0%) | |
| Calcium channel blockers | 7 (58.3%) | 12 (48.0%) | |
| Comorbid with diabetes, n (%) | 2 (16.7%) | 10 (40.0%) | 0.263 |
| Line of TKI therapy, n (%) | |||
| 1st | 7 (58.3%) | 18 (72.0%) | |
| 2nd | 2 (16.7%) | 3 (12.0%) | |
| 3rd | 3 (25.0%) | 4 (16.0%) | |
| Prior TKI therapy, n (%) | |||
| Regorafenib | 4 (33.3%) | 4 (16.0%) | 0.395 |
| Sorafenib | 4 (33.3%) | 7 (28.0%) | 1.000 |
| Initial dose of lenvatinib, n (%) | |||
| 12 mg | 7 (58.3%) | 6 (24.0%) | 0.067 |
| 8 mg | 5 (41.7%) | 19 (76.0%) | |
| Initial dose reduction, n (%) | |||
| Yes | 3 (25.0%) | 9 (36.0%) | 0.711 |
| No | 9 (75.0%) | 16 (64.0%) | |
| Duration of lenvatinib treatment, months (IQR) | 8.1 (4.6–18.4) | 1.9 (1.0–5.8) | 0.007 |
| [Blood tests at baseline] | |||
| Alb (g/dL), median (IQR) | 3.9 (3.6–4.0) | 3.4 (3.1–3.8) | 0.018 |
| ALBI grade, n (%) | |||
| 1 | 6 (50.0%) | 4 (16.0%) | 0.048 |
| 2 | 6 (50.0%) | 20 (80.0%) | |
| 3 | 0 (0%) | 1 (4.0%) | |
| AFP (ng/mL), median (IQR) | 62.4 (7.4–1737.5) | 66.8 (5.7–7098.5) | 0.733 |
| eGFR (mL/min/1.73 m2), median (IQR) | 59.6 (47.4–85.0) | 78.1 (66.5–96.3) | 0.045 |
| [Urine test at baseline] | |||
| Pre-existing proteinuria, n (%) | 2 (16.7%) | 1 (4.0%) | 0.241 |
IQR, interquartile range; ECOG PS, Eastern Cooperative Oncology Group performance status; BCLC, Barcelona Clinic Liver Cancer; ALBI grade, albumin-bilirubin grade; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure; RAS, renin-angiotensin system; TKI, tyrosine kinase inhibitor.
Among the 37 patients, overall response was evaluated in 35 patients. Tumor response was determined as CR, PR, SD, and PD in 0 (0.0%), 12 (34.3%), 12 (34.3%), and 11 (31.4%) patients, respectively, with an ORR of 34.3% and a DCR of 68.6%. Two patients could not be evaluated because they discontinued lenvatinib due to ADEs before response evaluation. According to proteinuria, grades 1, 2, and 3 proteinuria were observed in 15 (40.5%), 10 (27.0%), and 2 (5.4%) patients, respectively, during lenvatinib treatment, of which, 4 out of 10 (40.0%) and 2 out of 2 (100%) patients who developed grades 2 and 3 proteinuria required dose reduction or temporary interruption of lenvatinib, respectively.
Risk Factors of ProteinuriaTo investigate the risk factors of developing grade ≥2 proteinuria, univariate and multivariate Cox proportional hazards models were employed. In the univariate analysis, pre-existing proteinuria at baseline (hazard ratio (HR), 8.60; 95% confidence interval (CI), 1.21–40.94; p = 0.035), and eGFR <60 mL/min/1.73 m2 at baseline (HR, 4.28; 95% CI, 1.27–15.00; p = 0.020) were significantly associated with the development of grade ≥2 proteinuria (Table 2). In the multivariate analysis, pre-existing proteinuria at baseline (HR, 9.72; 95% CI, 1.29–52.21; p = 0.030), and eGFR <60 mL/min/1.73 m2 at baseline (HR, 4.49; 95% CI, 1.32–16.07; p = 0.017) were also significantly associated with the development of grade ≥2 proteinuria.
| Variables | Univariate analyses | Multivariate analysis | ||
|---|---|---|---|---|
| HR (95% CI) | p-Value | HR (95% CI) | p-Value | |
| Pre-existing proteinuria | 8.60 (1.21–40.94) | 0.035 | 9.72 (1.29–52.21) | 0.030 |
| eGFR <60 mL/min/1.73 m2 | 4.28 (1.27–15.00) | 0.020 | 4.49 (1.32–16.07) | 0.017 |
| Age ≥75 years | 1.65 (0.49–5.78) | 0.414 | NE | |
| Male sex | 2.40 (0.44–44.48) | 0.357 | NE | |
| Body weight <60 kg | 1.01 (0.97–1.05) | 0.650 | NE | |
| Child–Pugh B | 0.98 (0.05–5.36) | 0.984 | NE | |
| BCLC stage B | 1.01 (0.15–3.95) | 0.989 | NE | |
| Albumin ≤3.5 mg/dL | 2.88 (0.82–13.23) | 0.100 | NE | |
| ALBI grade 1 | 2.51 (0.78–8.10) | 0.121 | NE | |
| SBP ≥120 mmHg | 1.69 (0.53–6.42) | 0.385 | NE | |
| Concomitant use of antihypertensive agents | 1.26 (0.40–4.74) | 0.699 | NE | |
| Comorbid with diabetes mellitus | 0.38 (0.06–1.47) | 0.171 | NE | |
| Prior TKI treatment | 2.30 (0.68–10.38) | 0.187 | NE | |
| Initial dose reduction | 0.55 (0.12–1.88) | 0.352 | NE | |
| Initial dose of 8 mg | 0.58 (0.24–1.44) | 0.235 | NE | |
HR, hazard ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; BCLC, Barcelona Clinic Liver Cancer; ALBI grade, albumin-bilirubin grade; SBP, systolic blood pressure; TKI, tyrosine kinase inhibitor. NE indicates that the covariate was not evaluated in the multivariate analysis because it was not significant in the univariate analyses.
The Kaplan–Meier curves for the cumulative incidences of proteinuria after starting lenvatinib treatment are shown in Fig. 1. The cumulative incidence of developing grade ≥2 proteinuria (Fig. 1A) was significantly higher in patients with pre-existing proteinuria at baseline than in those without (p = 0.002). The cumulative incidence of developing grade ≥2 proteinuria (Fig. 1B) was significantly higher in patients with baseline eGFR of <60 mL/min/1.73 m2 than in those with eGFR of ≥60 mL/min/1.73 m2 (p = 0.009).
Cumulative incidences of grade 2 proteinuria were compared (A) between patients with and without pre-existing proteinuria, and (B) between patients with eGFR <60 mL/min/1.73 m2 and eGFR ≥60 mL/min/1.73 m2 at baseline. Log-rank tests were performed to compare the differences between the groups.
The incidence of developing each grade of proteinuria in patients with and without pre-existing proteinuria, as well as in patients with lower and higher eGFR at baseline, are shown in Table 3. The incidence of developing each grade of proteinuria was significantly different between the lower and higher eGFR groups (p = 0.004).
| Variables | n | Grade | |||
|---|---|---|---|---|---|
| 0 | 1 | 2 | 3 | ||
| Pre-existing proteinuria | |||||
| Yes | 3 | 0 (0%) | 1 (33.3%) | 2 (66.7%) | 0 (0%) |
| No | 34 | 10 (29.4%) | 14 (41.2%) | 8 (23.5%) | 2 (5.9%) |
| eGFR at baseline | |||||
| <60 mL/min/1.73 m2 | 10 | 4 (40.0%) | 0 (0%) | 5 (50.0%) | 1 (10.0%) |
| ≥60 mL/min/1.73 m2 | 27 | 6 (22.2%) | 15 (55.6%) | 5 (18.5%) | 1 (3.7%) |
eGFR, estimated glomerular filtration rate.
The multivariate Cox proportional hazards model used in this study showed that the risk of developing grade ≥2 proteinuria was significantly higher in patients with pre-existing proteinuria and eGFR <60 mL/min/1.73 m2 at baseline. This is the first study to investigate the risk factors of proteinuria in patients with HCC receiving lenvatinib in a real-world clinical practice setting.
In a phase-II trial of Asian patients with HCC, proteinuria was the most frequently reported ADE leading to drug discontinuation.4) Therefore, accumulating information on the risk factors of proteinuria is important for considering the potential protective approach for proteinuria that can lead to discontinuation of lenvatinib treatment. However, no study has been reported significant risk factor for developing proteinuria in patients receiving lenvatinib. Previous reports in real-world setting have reported the risk factors of developing grade ≥2 proteinuria in patients with cancer receiving monoclonal antibodies targeting the human vascular endothelial growth factor pathway.13,14) In addition, several studies pointed out the clinically importance of grade ≥2 proteinuria after started lenvatinib treatment.15–17) Thus, we investigated the risk factors of developing grade ≥2 proteinuria in this study. Two reports from prospective clinical trials and a retrospective study report have described the risk factors of developing proteinuria in patients with RCC receiving VEGFR-TKIs. A phase-II trial of axitinib monotherapy in patients with metastatic RCC (n = 64) showed that baseline urine protein levels and lower eGFR were significantly associated with proteinuria.7) In that study, proteinuria was a major cause of drug discontinuation. Another report comprised a pooled analysis from two phase-III randomized controlled trials in patients with metastatic RCC receiving pazopanib or sunitinib (n = 1392), and this showed that Asian ethnicity, diabetes, hypertension, pre-existing proteinuria, and prior nephrectomy were significant risk factors of proteinuria.9) Baek et al. retrospectively investigated the risk factors associated with proteinuria in patients with metastatic RCC treated with sunitinib and reported that hypertension, hyperlipidemia, and baseline eGFR <60 mL/min/1.73 m2 were significant risk factors of any-grade of proteinuria.8) In our study, pre-existing proteinuria and eGFR <60 mL/min/1.73 m2 at baseline were significant risk factors for developing grade ≥2 proteinuria. The results of this study were consistent with those of the previous three reports and suggested that patients who have these risk factors at baseline should be monitored carefully when starting treatment with VEGFR-TKIs.
In the present study, grade ≥2 and grade 3 proteinuria were observed in 32.4 and 5.4% of patients, respectively. The reported incidence of grade ≥2 and grade 3 proteinuria is 15–44 and 5–25%, respectively.3,4,6,16,17,22–26) Although the incidence of grade ≥2 proteinuria in the present study was almost average within these ranges, the proportion of grade 3 proteinuria was relatively lower than that reported previously. We speculate that this discordance was due to the proportions of patients with pre-existing proteinuria at baseline, which was reported to affect the risk of developing any-grade and grade 3 proteinuria.3) The REFLECT study was a phase-III, randomized trial which compared the efficacy and safety of lenvatinib and sorafenib in first-line therapy for patients with unresectable HCC. In that study, sixteen-percent of patients had pre-existing proteinuria, and grade ≥2 and grade 3 proteinuria were observed in 16.4 and 5.9% of patients, respectively.3,27) In contrast, only 3 (8.1%) patients had pre-existing proteinuria at baseline in our study may lead to relatively lower proportion of grade 3 proteinuria (5.4%) out of grade ≥2 proteinuria (32.4%).
Twelve patients started a one-level dose reduction of lenvatinib based on the physician’s discretion, and 24 patients started initial daily dose of 8 mg. However, initial dose reduction or 8 mg dose of lenvatinib were not affect the development of grade 2 proteinuria in our univariate Cox proportional hazards models. These results were consisted the results of previous pharmacokinetic study of lenvatinib. Hata et al. investigated the associations of plasma lenvatinib trough concentration and efficacy and adverse events in 21 Japanese patients with hepatocellular carcinoma, and reported that plasma lenvatinib concentration did not affect the development of proteinuria.22) Taken together, initial dose reduction of lenvatinib could not contribute to reduce the risk of developing proteinuria.
There are some limitations to our study. First, this was a single-center, small, retrospective study. The study subjects comprised a heterogeneous population. Nevertheless, the multivariate analysis showed that patients with pre-existing proteinuria at baseline and eGFR <60 mL/min/1.73 m2 were significantly associated with developing grade ≥2 proteinuria. Second, although recent studies have reported that evaluating the urine protein/creatinine ratio (UPCR) might be appropriate for assessing proteinuria in patients with HCC or thyroid cancer receiving tyrosine kinase inhibitors including lenvatinib,15,28,29) we did not use the UPCR, because it has not been utilized in the CTCAE criteria and the protocols of randomized controlled trials. Third, we did not precisely quantitate 24-h urine protein levels based on this retrospective observational study design. Our preliminary findings should be confirmed by further studies.
In conclusion, we investigated the risk of developing grade ≥2 proteinuria in patients with HCC receiving lenvatinib. The results of our study suggest that patients with pre-existing proteinuria at baseline and eGFR <60 mL/min/1.73 m2 are significantly associated with developing grade ≥2 proteinuria. These patients should be monitored carefully, and our preliminary findings should be confirmed by prospective studies with a larger sample size.
The authors declare no conflict of interest.
