Poor tolerability of lenvatinib in elderly patients ≥80 years old with hepatocellular carcinoma: A multicenter observational study

Abstract

Introduction

Management of elderly patients with cancer has become a global issue. We investigated the safety and tolerability of lenvatinib in hepatocellular carcinoma (HCC) patients ≥80 years old.

Methods

We retrospectively evaluated 61 HCC patients and divided them into 2 groups: an elderly group (n = 13, ≥80 years old) and a younger group (n = 48, <80 years old). We compared the adverse events (AEs), administration period, dose intensity, objective response, and progression-free survival (PFS) between the two groups.

Results

The discontinuation of lenvatinib due to AEs was more frequent in the elderly group (8/13, 61.5%) than in the younger group (10/48, 20.8%) (P = 0.0043). Fatigue and appetite loss accounted for half of the cases discontinued due to AEs in the elderly group. The elderly group had a significantly lower 8-week-delivered dose intensity/body surface area ratio (147.2) and 8-week-relative dose intensity (50.0%) than those in the younger group (267.4, 67%) (P = 0.003, 0.029). The objective response rate was significantly lower in the elderly group (15.4%) than in the younger group (61.5%) (P = 0.021). The PFS in the elderly group tended to be shorter than that in the younger group (P = 0.058, hazard ratio [HR] 1.98). The modified albumin-bilirubin (mALBI) grade (hepatic function) (HR, 2.60; P = 0.01) and objective response (HR, 0.41; P = 0.011) were independently associated with the PFS in the multivariate analysis.

Conclusion

The management of AEs is crucial for adherence and maintaining the dose intensity of lenvatinib in elderly HCC patients.

Keywords

Hepatocellular carcinoma lenvatinib elderly tolerability

Introduction

Hepatocellular carcinoma (HCC), representing about 90% of primary liver cancers, is a major health problem worldwide. In 2017, there were 953 000 incident cases of liver cancer and 819 000 deaths globally.^1,2 Despite advances in therapeutic strategies, such as surgical resection, transplantation, percutaneous ablation, transarterial chemoembolization (TACE), and the administration of tyrosine kinase inhibitors, the mortality rates remains comparable to the incidence of HCC, reflecting the poor prognosis of this disease.

The average life expectancy during the 20th century dramatically increased in many parts of the world. The incidence of HCC increases progressively with advancing age in all populations, reaching a peak at 70 years old.² Japan has the highest life expectancy in the world, at 81.4 years for men and 87.5 for women, resulting in a dramatic increase in the chance of encountering to patients with HCC ≥80 years old.³ In fact, the incidence of HCC in Japan is reported to be the highest in the cohort of men 70 to 79 years old,^2,4 and almost half of the patients with HCC ≥75 years old treated at our institutions are over 80 years old.

In general, elderly patients have more comorbidities such as cardiovascular and cerebrovascular disease, pulmonary disease, renal disease, locomotor diseases, mental disorders, and other malignant tumors compared with

younger patients. Regarding the liver, the hepatic volume and hepatic blood flow decline between 20 and 70years old. An age-related decrease in the liver metabolic capacity is also observed, including reduced cytochrome P450 activity, a decreased liver regeneration capacity and a decline in immunity, which leads to an increased risk of drug-induced liver damage and severe viral hepatitis in the elderly patients.^5–8

Since the REFLECT study showed that lenvatinib was non-inferior to sorafenib concerning the overall survival of patients with untreated advanced HCC,⁹ several studies have demonstrated the efficacy and feasibility of lenvatinib in patients with advanced HCC.^10,11 Generally, elderly patients ≥80 years old have been underrepresented in clinical trials. Indeed, although the REFLECT study did include elderly patients ≥75 years old, they accounted for only 12.0% of all patients.⁹ Furthermore, although 1 study from Japan evaluated the efficacy and safety of lenvatinib in patients with HCC ≥75 years old,¹² we found difficulty in administering lenvatinib to HCC patients ≥ 80 years old in daily clinical practice. Therefore, the safety and efficacy of lenvatinib in HCC patients aged ≥80 years old should be explored.

In the present study, we investigated the safety, tolerability, and efficacy outcome of lenvatinib in elderly HCC patients ≥80 years old.

Materials and methods

Patients

This was a multicenter, retrospective observational study carried out at the three institutions. A total of 61 patients with HCC treated with lenvatinib from May 2018 to December 2020 were enrolled. All medical records were reviewed retrospectively for patients’ demographic and clinical data, administration period, dose intensity, adherence rates at 8 weeks, objective response, adverse effects (AEs) of lenvatinib, and survival outcomes.

The following comorbidities were evaluated: hypertension, diabetes mellitus, cardiovascular and cerebrovascular disease, pulmonary disease, renal disease, locomotor diseases, mental disorders, and other malignant tumors.

This study conformed to the Declaration of Helsinki and the current ethical guidelines and was approved by each institutional ethics board. Written informed consent for participation in this study was not obtained, as this study was retrospective and analyzed anonymously.

The diagnosis of HCC and the evaluation of the liver functional reserve

The diagnosis of HCC was confirmed either pathologically or using imaging techniques, including four-phase multidetector-row computed tomography (CT) or dynamic contrast-enhanced magnetic resonance imaging (MRI). The diagnosis was based on the typical hallmarks of HCC (combination of hypervascularity in the late arterial phase and wash out in the portal venous and/or delayed phase).²

Tumor-related variables, such as the maximum tumor diameter, presence of vascular invasion and presence of extrahepatic metastases, and the Barcelona Clinic Liver Cancer (BCLC) classification,² were evaluated based on these imaging techniques and other variables.

The liver functional reserve was evaluated by the Child-Pugh score and modified albumin-bilirubin (mALBI) grade.^13,14

Inclusion and exclusion criteria

The inclusion criteria were unresectable HCC patients with ≥1measurable target lesions, patients with BCLC classification B or C, patients with Child-Pugh grade A or B, and patients with an Eastern Cooperative Oncology Group performance status score of 0–2.

The exclusion criteria were patients with BCLC classification D, patients with Child-Pugh grade C, and patients with an Eastern Cooperative Oncology Group performance status score ≥3

Treatment protocol

The initial dose of lenvatinib was determined based on the body weight at inclusion, with a dose of 12 mg/day administered to patients with a body weight ≥60 kg and 8 mg/day to patients with a body weight <60 kg. At the discretion of the attending physician, the initial dose of lenvatinib was reduced based on the performance status or liver functional reserve. In any Grade ≥3 severe adverse events (AEs) or unacceptable treatment-related AEs were noted, the dose of lenvatinib was reduced, or the treatment was discontinued. Even for Grade 2 ≤ AEs, the dose was reduced or suspended at the discretion of the attending physician if necessary. AEs were evaluated in accordance with the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0. Based on the guideline provided by the manufacturer, dose reduction or temporary interruption of lenvatinib was maintained until the AEs resolved to Grade 1 or 2.

The relative dose intensity (RDI) was calculated by dividing the actual delivered dose by the standard dose, and the delivered dose intensity/body surface area ratio (DBR) was calculated by dividing the actual delivered dose by the body surface area.¹⁵ Eso et al. pointed that since the doses fix according to body weight with a cut-off of 60 kg, two patients of the same height, weighing 60 kg and 59 kg may have a significantly different RDI despite being almost equal in size.¹⁵ They then demonstrated that DBR reflected the treatment intensity of lenvatinib more accurately than the RDI.¹⁵

Aadherence rates was defined as the proportion of prescribed doses actually taken.

The evaluation of the treatment response and follow-up

Patients were followed carefully after the treatment. The treatment response (complete response [CR], partial response [PR], stable disease [SD], or progressive disease [PD]) was evaluated by four-phase multidetector-row CT or dynamic contrast-enhanced MRI four to eight weeks after the initiation of treatment, based on the Modified Response Evaluation Criteria in Solid Tumors (mRECIST) guidelines.¹⁶

The start date of the follow-up was the date of the first administration of lenvatinib. The end date of the follow-up was the time of the last follow-up covered by this study (December 2020) or the time of the patient's death.

The progression-free survival (PFS) is defined as the time from the first administration of lenvatinib to disease progression or death from any cause.

Net clinical benefit evaluation

We evaluated the net clinical benefit of lenvatinib by subtracting the percentage of all adverse events or the percentage of adverse events leading to discontinuation from the response rate in patients with a high 8-week DBR (≥242) and high 8-week RDI (≥67%).

Statistical analyses

Comparisons between the groups were performed using the Mann–Whitney U-test for continuous and ordinal variables and the χ2 -test or Fisher's exact test for categorical variables. To evaluate the predictive ability of RDI and DBR for the objective response (CR or PR), receiver-operating characteristic curves (ROCs) were generated, and the areas under the curve (AUCs) were measured. The time to discontinuation of lenvatinib due to AEs and the PFS were calculated using the Kaplan–Meier method and differences between the groups were compared by the log–rank test. To evaluate the predictive factors for the discontinuation of lenvatinib, logistic regression analyses were used to calculate the odds ratio (OR) with 95% confidence intervals (CIs). Based on previous studies,^17–19 potential predictive factors for the discontinuation of lenvatinib were determined. A sample power analysis was performed using the results of a previous study, which found that 57% of patients ≥70 years old and 22% of those < 70 years old required discontinuation of lenvatinib; we thus concluded that the minimum sample size to achieve a power of 80% with an alpha of 0.05 was 17 subjects for elderly group and 63 for younger group. Both univariate and multivariate analyses for the PFS were performed using the Cox proportional hazard model. The cut-off points of continuous variables for the univariate and multivariate analyses were determined based on the median values. Variables that proved to be significant in the univariate analyses were tested subsequently with a multivariate Cox proportional hazard model. P < 0.05 was considered statistically significant. All statistical analyses were performed using the IBM SPSS Statistics software program, version 19.0 (IBM SPSS, Chicago, IL, USA) and https://biostat.app.vumc.org/wiki/Main/PowerSampleSize.

Results

Baseline characteristics

All of the 61 patients enrolled were eligible for evaluation. The baseline characteristics of the patients according to their age at inclusion are shown in Table 1. A total of 13 patients ≥80 years old were classified as the elderly group, and the remaining 48 patients <80 years old were classified as the younger group. Significant differences were found between the two groups in the performance status, body weight, body mass index, body surface area, ALT, estimated glomerular filtration rate (eGFR), presence of comorbidities, initial dose of lenvatinib, and adherence rates at 8 weeks.

Table 1.

Baseline characteristics of patients.

Variable	Elderly group (n = 13)	Younger group (n = 48)	P-value
Age (years)	84 (80-89)	73 (44-79)	<0.0001<0.0001<0.0001
Sex (men/women)	11 (84.6%)/2 (15.4%)	42 (87.5%)/6 (12.5%)	0.78
ECOG-PS (0/1/2) †a	6 (60%)/3(30%)/1(10%)	42(93.3%)/2(4.4%)/1(2.2%)	0.016
Body weight (kg)	57.2 (52.8-68.0)	62.9 (43.7-96.1)	0.01
Body mass index	22.3 (17.4-25.6)	23.3 (17.3-32.9)	0.031
Body surface area (m2)Body surface area (m2)Body surface area (m2)	1.6 (1.5-1.8)	1.7 (1.4-2.1)	0.025
AST (IU/l)	40 (16-140)	37 (17-132)	0.97
ALT (IU/l)	19 (9-59)	31 (15-149)	0.03
Albumin (g/dl)	3.6 (3.0-4.8)	3.8 (2.8-4.8)	0.25
Total bilirubin (mg/dl)	0.6 (0.3-1.1)	0.7 (0.2-6.2)	0.31
Creatinine (mg/dl)	0.8 (0.6-1.8)	0.8 (0.5-4.3)	0.27
eGFR (ml/min./1.73 m2) eGFR (ml/min./1.73 m2) eGFR (ml/min./1.73 m2)	54 (29-75)	68 (11-121)	0.044
Platelet count (104/mm3)Platelet count (104/mm3)Platelet count (104/mm3)	14.2 (10.3-23.1)	15.2 (5.4-34.6)	0.87
Prothrombin time (%)	94 (26-100)	95 (63-100)	0.52
Maximum tumor diameter (mm)	42 (12-148)	38 (12-140)	0.82
Child–Pugh score (5/6/7/8)	5(38.5%)/6(46.2%)/2(15.4%)/0(0%)	34(70.8%)/7(14.6%)/6(12.5%)/1(2.1%)	0.076
mALBI grade (1/2a/2b/3)	3(23.1%)/6(46.2%)/4(30.8%)/0(0%)	19(39.6%)/17(35.4%)/11(22.9%)/1(2.1%)	0.37
BCLC stage (B/C)	6(46.2%)/7(53.8%)	28(58.3%)/20(41.7%)	0.43
Portal vein invasion (yes/no)Portal vein invasion (yes/no)	2(15.4%)/11(84.6%)	10(20.8%)/38(79.2%)	0.5
Extrahepatic spread (yes/no)	5(38.5%)/8(61.5%)	10(20.8%)/38(79.2%)	0.19
AFP (ng/mL)	4.9 (1.6-48500)	12 (1.3-71100)	0.09
DCP (mAU/mL)	154 (10-91832)	187 (17-19673)	0.74
Initial dose of lenvatinib (4 mg/8 mg/12 mg)	1(7.7%)/12(92.3%)/0(0%)	6(12.5%)/27(56.3%)/15(31.3%)	0.042
Initial dose reduction of lenvatinib at the time of start (yes/no)	3(23.1%)/10(76.9%)	17(35.4%)/31(64.6%)	0.31
History of tyrosine kinase inhibitor treatment (yes/no)	0(0%)/13(100%)	2(4.2%)/46(95.8%)	0.62
Comorbidities (present/absent)	13(100%)/0(0%)	34(70.8%)/14(29.2%)	0.021
Adherence rates at 8 weeks	61.5%	85.4%	0.11

Continuous variables are expressed as median and range. Categorical variables are expressed as number and percentages.

†a

n = 55 ECOG-PS: Eastern Cooperative Oncology Group Performance status, AST: aspartate aminotransferase, ALT: alanine aminotransferase, eGFR: estimated glomerular filtration rate, mALBI grade: modified albumin-bilirubin grade, BCLC stage: Barcelona Clinical Liver Cancer stage, eGFR: estimated glomerular filtration rate, mALBI grade: modified albumin-bilirubin grade, BCLC stage: Barcelona Clinical Liver Cancer stage, AFP: α-fetoprotein, DCP: des-γ-carboxy prothrombin, Comorbidities: cardiovascular disease/ cerebro-vascular disease/ other malignant tumor/ mental disorder/ hypertension/ diabetes mellitus.

Although the initial dose of lenvatinib was significantly lower in the elderly group than in the younger group (P = 0.042), there was no significant difference in the initial dose reduction between the groups (P = 0.31).

AEs and the administration period of lenvatinib

A total of 95.1% (58/61) experienced treatment-related AEs. No fatal treatment-related AEs were observed. The most common AEs were hypothyroidism (52.5%), followed by fatigue (36.1%), hypertension (34.4%), diarrhea (29.5%), hand-foot skin reaction (24.6%), proteinuria (24.6%), and appetite loss (22.3%).

Over 30% of patients suffered fatigue, appetite loss, and hypothyroidism in the elderly group, while over 30% of patients suffered hypothyroidism, hypertension, fatigue, and diarrhea in the younger group. The frequency of appetite loss (both any grade and Grade ≥3) was significantly higher in the elderly group than in the younger group (46.2% vs. 16.7%, P = 0.025, 38.4% vs. 0.02%, P = 0.0011), while the frequency of other AEs was comparable between the two groups (Table 2).

Table 2.

Summary of adverse events.

Event, n (%)	Elderly group (n = 13)	Younger group (n = 48)	P-value
Hypertension
Any grade	3 (23.1%)	18 (37.5%)	0.27
Grade ≥3	0 (0%)	3 (6.3%)	0.48
Hand–foot skin reaction
Any grade	3 (23.1%)	12 (25.0%)	0.60
Grade ≥3	1 (7.7%)	3 (6.3%)	0.63
Fatigue
Any grade	6 (46.2%)	16 (33.3%)	0.39
Grade ≥3	0 (0%)	0 (0%)	1
Appetite loss
Any grade	6 (46.2%)	8 (16.7%)	0.025
Grade ≥3	5 (38.4%)	1 (0.02%)	0.0011
Diarrhea
Any grade	3 (23.1%)	15 (31.3%)	0.42
Grade ≥3Grade ≥3	0 (0%)	1 (0.02%)	0.79
Proteinuria
Any grade	3 (23.1%)	12 (25.0%)	0.60
Grade ≥3Grade ≥3	3 (23.1%)	5 (10.4%)	0.22
Hypothyroidism
Any grade	4 (30.8%)	28 (58.3%)	0.072
Grade ≥3	1 (7.7%)	3 (6.3%)	0.63
Others
Any grade	3 (23.1%)	２７ (56.3%)	0.034
Grade ≥3	2 (15.4%)	8 (16.7%)	0.64
Discontinuation of Lenvatinib due to adverse events	8 (61.5%)	10 (20.8%)	0.0043
Cause of discontinuation	Fatigue 4 †b	Proteinuria 2
	Appetite loss 4 †b	Fatigue 1
	Liver damage 2	Hepatic encephalopathy 1
	Hepatic encephalopathy 1	Liver damage 1
	Proteinuria 1	Hand–foot skin reaction 1
		Renal insufficiency 1
		Deterioration of PS 1
		Anemia 1
		Tumorlysis syndrome 1

†b

overlapped PS: performance status.

The discontinuation of lenvatinib due to AEs was more frequent in the elderly group (61.5%, 8/13) than in the younger group (20.8%, 10/48) (P = 0.0043) (Table 2). Then, we compared the discontinuation rates in the two groups by matching performance status and BMI, which were significantly different in the two groups. In patients with a BMI of 22 (median of the elderly group) or higher and performance status 0, the discontinuation rates in the two groups were compared. The discontinuation rate was still significantly higher in the elderly group: 3/4 (75.0%) in the elderly group and 5/27 (18.5%) in the younger group (P = 0.043). The AEs leading to the discontinuation of lenvatinib were fatigue (8.2%, 5/61), followed by appetite loss (6.6%, 4/61), proteinuria (4.9%, 3/61), liver damage (4.9%, 3/61), and hepatic encephalopathy (3.3%, 2/61). Among them, fatigue and appetite loss accounted for half of the cases (4/8 and 4/8, respectively [both overlapping]) of discontinuation due to AEs in the elderly group. The time to discontinuation of lenvatinib due to AEs was significantly shorter in the elderly group (median 134 days) than in the younger group (not reached during the period) (P = 0.001) (Figure 1). The discontinuation rate at 6 months and 12 months was 38.5% (5/13) and 61.5% (8/13) in the elderly group and 18.8% (9/48) and 20.8% (10/48) in the younger group, respectively (P = 0.13, P = 0.0074). The shortest time from treatment initiation to discontinuation of lenvatinib is seven days in 2 patients in the elderly group.

A multivariate logistic regression analysis identified elderly group (OR 4.94; P = 0.023) and presence of hypothyroidism (OR 0.28; P = 0.047) as independently associated with the discontinuation of lenvatinib (Table 3).

Table 3.

Results of a multivariate analysis for predicting discontinuation of lenvatinib.

Factors	Univariate	Multivariate
Factors	P-value	Odds ratio (95% CI) P-value
Age group (≥80 years old)	0.007	4.94 (1.25-19.5) 0.023
Sex (female)	0.60
Body weight (≥ 60.7 kg)	0.52
Creatinine (≥0.84 mg/dl)	0.48
eGFR (< 67 ml/min./1.73 m2)	0.058
mALBI grade (2a/2b/3)	0.39
BCLC stage (C)	0.59
Adverse event: hypertension (yes)	0.48
Adverse event: hand–foot skin reaction (yes)	0.71
Adverse event: fatigue (yes)	0.77
Adverse event: appetite loss (yes)	0.22
Adverse event: diarrhea (yes)	0.42
Adverse event: proteinuria (yes)	0.31
Adverse event: hypothyroidism (yes)	0.016	0.28 (0.079-0.98) 0.047
Initial dose of lenvatinib (12 mg)	0.78

CI: confidence interva, eGFR: estimated glomerular filtration rate, mALBI grade: modified albumin-bilirubin grade,.

BCLC stage: Barcelona Clinical Liver Cancer stage.

Therapeutic response to lenvatinib and the association with RDI and DRW

For one patient in the elderly group and 4 in the younger group, the treatment effect could not be determined. In the elderly group, no patients achieved a CR, while 15.4% (2/13) had a PR, 46.2% (6/13) had an SD, and 30.8% (4/13) had a PD. In the younger group, 12.5% (6/48) patients had a CR, while 39.5% (19/48) had a PR, 31.3% (15/48) had an SD, and 8.3% (4/48) had a PD. The objective response rate (ORR) and disease control rate (DCR) were 15.4% (2/13) and 61.5% (8/13), respectively, in the elderly group and 52.1% (25/48) and 83.3% (40/48), respectively, in the younger group (P = 0.021, 0.055).

In the 55 patients for whom we were able to determine a treatment effect, ROC curves of the 4-week, 8-week, 12-week RDI, and 8-week DBR were constructed for the objective response (CR or PR), and we then compared the areas under the ROC (AUROC). The 8-week DBR (0.692) and 8-week RDI (0.679) had higher AUC values than the 4-week RDI (0.565) and 12-week RDI (0.586).

Patients with CR or PR had a significantly higher 8-week DBR (290.6) and 8-week RDI (78.6%) than those with SD or PD (189.7, 50.9%) (P = 0.014, 0.02). Conversely, patients with a high 8-week DBR (≥242) and high 8-week RDI (≥67%) achieved a greater objective response than those with a low 8-week DBR (<242) and low 8-week RDI (<67%) (P = 0.015, 0.026). Patients with mALBI 1 had a significantly higher 8-week DBR (298.1) and 8-week RDI (92.9%) than those with mALBI 2a or 2b (202.1, 62.5%) (P = 0.006, 0.003). Patients in the elderly group had a significantly lower 8-week DBR (147.2) and 8-week RDI (50.0%) than those in the younger group (267.4, 67%) (P = 0.003, 0.029).

Net clinical benefit evaluation

In patients with a high 8-week DBR (≥67%), objective response rate (0.61) - the percentage of all adverse events (1)<0, objective response rate (0.61) - the percentage of adverse events leading to discontinuation (0.097)>0.

In patients with a high 8-week RDI (≥67%), objective response rate (0.59) - the percentage of all adverse events (1)<0, objective response rate (0.59) - the percentage of adverse events leading to discontinuation (0.088)>0.

The PFS and prognostic factors

The median follow-up duration was 330 days (range,30-930). During the follow-up period, 10 (76.9%) patients in the elderly group and 15 (31.3%) patients in the younger group died.

The cumulative PFS at 50, 100, 150, and 200 days was 75.0%, 66.7%, 58.3%, and 41.7% in the elderly group and 95.7%, 84.5%, 70.2%, and 70.2% in the younger group, respectively. The PFS in the elderly group tended to be shorter than that in the younger group (P = 0.058, hazard ratio [HR] 1.98, 95% CI 0.96–4.1) (Figure 2). The shortest time from treatment initiation to outcome analysis is 34 days in one patient in the elderly group.

The results of the univariate and multivariate analyses are shown in Table 4. The multivariate analysis revealed that the mALBI grade (1 vs. 2a or 2b or 3) (P = 0.01, HR 2.60; 95% CI 1.26–5.36) and objective response (P = 0.011, HR 0.41; 95% CI 0.2–0.81) were independently associated with the PFS.

Table 4.

Results of a multivariate analysis for factors associated with the PFS.

	Univariate	Multivariate
Factors	P-value	Hazard ratio (95% CI) P-value
Age group (≥80 years old)Age group (≥80 years old)	0.065
Sex (female)	0.14
Body weight (≥ 60.7 kg)Body weight (≥ 60.7 kg)	0.75
Child-Pugh score ≥ 7Child-Pugh score ≥ 7	0.27
mALBI grade (2a/2b/3)	0.042	2.60 (1.26-5.36) 0.010
BCLC stage (C)	0.39
Portal vein invasion (yes)Portal vein invasion (yes)	0.22
Extrahepatic spread (yes)	0.90
AFP (≥ 9.2 ng/mL)AFP (≥ 9.2 ng/mL)	0.11
Objective response (PR/CR)	0.063	0.41 (0.20-0.81) 0.011
Initial dose of lenvatinib (12 mg)	0.40
8w-RDI (≥ 67%)8w-RDI (≥ 67%)	0.074
8w-DBR (≥ 242)8w-DBR (≥ 242)	0.15

PFS: progression-free survival, CI: confidence interva, mALBI grade: modified albumin-bilirubin gradePFS: progression-free survival, CI: confidence interva, mALBI grade: modified albumin-bilirubin grade.

BCLC stage: Barcelona Clinical Liver Cancer stage, AFP: α-fetoprotein, CR: complete response, PR: partial response.

RDI: relative dose intensity, DBR: dose intensity/body surface area ratio.

Discussion

In the current study, treatment discontinuation as a result of adverse drug reactions was more frequent in the elderly group than in the younger group. Futhermore, the dose intensity of lenvatinib at eight weeks and the ORR were significantly lower in the elderly group than in the younger group. Consequently, the elderly group tended to show a shorter PFS than the younger group.

The discontinuation rate due to AEs was significantly higher in the elderly group (61.5%) than in the younger group (20.8%) in the current study. The period leading up to discontinuation due to AEs was also significantly shorter in the elderly group (134 days) than in the younger group (not reached during the observation period). Shimose et al. demonstrated that age ≥71years old (discontinuation rate 57%) was independently associated with discontinuation of lenvatinib due to AEs in patients with HCC treated with lenvatinib, which is similar to the results of our study.¹⁷ In the elderly group, fatigue and appetite loss accounted for half of the cases discontinued due to AEs in our study. Ohki et al. reported that the most common reason for dose reduction or discontinuation of lenvatinib in patients with HCC treated with lenvatinib was appetite loss.¹⁸ Similarly, Shimose et al. identified ALBI grade 2 (discontinuation rate 70%), fatigue grade ≥3 (discontinuation rate 83%), and appetite loss ≥2 (discontinuation rate 87%) as splitting variables for predicting discontinuation of lenvatinib in patients ≥71 years old using a decision tree analysis.¹⁷ Hiraoka et al. also reported that the period leading up to discontinuation due to AEs was significantly shorter in those with the AEs of appetite loss than in those without appetite loss and was associated with a reduced body mass index.¹⁹

The phase 2 study of lenvatinib in patients with advanced HCC showed that the median body weight was lower in patients with an early dose withdrawal or reduction (54.1 kg) than in those without it (67.6 kg).²⁰ A previous dose-finding study for lenvatinib in patients with HCC indicated that a greater lenvatinib area under the plasma concentration curve and lower body weight resulted in earlier drug withdrawal or dose reduction.²¹ A recent study further showed that the incidence of anorexia was significantly higher in the group with a high median plasma trough concentration of lenvatinib than in the group with a low concentration.²² These findings support our results that elderly patients, whose body mass index and body weight (22.3, 57.2 kg) were significantly lower than those of younger patients (23.3, 62.9 kg), more frequently suffered from appetite loss than younger patients, resulting in their more frequent discontinuation of lenvatinib. Although the initial dose of lenvatinib was significantly lower in the elderly group than in the younger group in the current study, there were no significant differences regarding the initial dose reduction between the two groups, and the body weight was significantly lower in the elderly group than in the younger group. Therefore, the lower RDI and DBR at eight weeks in the elderly group than in the younger group is attributed to not the lower initial dose of lenvatinib but the higher discontinuation rate due to AEs among elderly patients, leading to a tendency toward a shorter PFS. In the setting of differentiated thyroid cancer, Yamazaki et al. showed that initial dose reduction of lenvatinib did not affect its efficacy and tolerability.²³

In addition to their lower body mass index and body weight, the factors of sarcopenia and frailty, which are prevalent among elderly patients, may have affected appetite loss and the discontinuation of lenvatinib, although we were unable to evaluate sarcopenia or frailty in this study. Endo et al. showed that the time to discontinuation of lenvatinib due to AEs for HCC patients with a decreased grip strength was significantly shorter than that for patients with a normal grip strength.²⁴ Uojima et al. also reported that HCC patients with a low skeletal muscle mass measured by computed tomography had a significantly higher withdrawal rate of lenvatinib due to AEs than those with high skeletal muscle mass.²⁵ Recently, Hiraoka et al. showed that the frequency of the AEs appetite loss was higher in HCC patients with sarcopenia than in those without it.²⁶

In contrast, some AEs are reported to be associated with favorable outcomes in HCC patients treated with lenvatinib. For example, Shimose et al. showed that patients who developed hypertension or hand-foot skin reaction had a significantly longer survival than those who did not.¹⁷ Hiraoka et al. also showed that patients with hand-foot skin reaction had a better objective response rate and longer time to progress than those without it.¹⁹ Shomura et al. reported that the patients who had Grade 2/3 hypothyroidism had a significantly longer overall survival than those who had Grade 0/1 hypothyroidism.²⁷ The occurrence of hypothyroidism was a favorable factor affecting medication adherence in the current study (OR: 0.28, P = 0.047). Why hypertension, hand-foot skin reaction, and hypothyroidism affected medication adherence or favorable survival outcomes remains unclear; however, these AEs are relatively easily managed with adequate medications, such as antihypertensive drugs, steroids, and thyroid hormone replacement, which may have resulted in maintaining effective blood concentrations of lenvatinib.¹⁷

Both the RDI and DBR at eight weeks were significantly lower in the elderly group than in the younger group in our study. Several studies have reported that the dose intensity of lenvatinib influenced the response to lenvatinib or the survival outcome. Takahashi et al. showed that patients with an RDI at 8 weeks ≥75% experienced a higher response rate and longer PFS than those with an RDI <75%.²⁸ Kirino et al. also showed that patients with an RDI at 4 weeks ≥70% had a higher disease control rate, longer duration of lenvatinib therapy, and longer overall survival than those with an RDI <70%.²⁹ In contrast, Eso et al. demonstrated that the predictive ability of the DBR at eight weeks for the response to lenvatinib was superior to that of RDI at eight weeks, which is consistent with our results. Futhermore, they showed that a high DBR at eight weeks was independently associated with a longer PFS in their multivariate analysis.¹⁵ In our study, both the RDI and DBR at 8 weeks were significantly lower in the elderly group (50%, 147.2) than in the younger group (67%, 267.4), which may have led to the lower ORR and shorter PFS in the elderly group (15.4%, 176 days) than in the younger group (52.0%, 300 days).

The results of the current study indicate that the management of AEs, especially appetite loss and fatigue, is crucial for ensuring adherence and a good RDI of lenvatinib in elderly HCC patients, which affects the disease control and survival outcome. Several previous studies have proposed effective solutions to the AEs of lenvatinib. Colombo et al. showed that the occurrence of primary adrenal insufficiency might be a common cause of fatigue during lenvatinib and vandetanib treatment in patients with advanced thyroid cancer, and replacement treatment with cortisone acetate was associated with the substantial relief of fatigue.³⁰

Iwamoto et al. showed that fatigue was the most common reason for dose reduction and discontinuation of lenvatinib in patients with HCC, and 5 days-on/2 days-off administration of lenvatinib was associated with better tolerability against lenvatinib and significantly prolonged the administration period and the survival time.³¹ Ohkubo et al. showed that lenvatinib therapy induced carnitine insufficiency in patients with HCC and carnitine supplementation ameliorated the lenvatinib-induced fatigue.³²

There are several limitations associated with the current study. First, this was a retrospective study with a small sample size. Furthermore, the number of cases was less than that estimated for sample power calculation. Second, we were unable to match both groups using propensity score matching due to the small sample size, which may have led to inconsistent results by Tada et al. that the efficacy and safety of lenvatinib were comparable between elderly (≥75 years old) and younger (<75 years old) patients.¹² Third, due to the multicenter, retrospective observational nature of the study, we were unable to obtain sufficient data, including the prevalence of sarcopenia and frailty. Fourth, although the REFLECT trial did not include Child B patients, some patients have no other treatment options other than lenvatinib, so even in Child B patients, lenvatinib was administered with caution and with attention to adverse events in our study. This might have a negative impact on the efficacy and safety of lenvatinib. Fifth, although lenvatinib dose reduction and discontinuation were made by attending physician decision even for Grade<2 AEs in our study, this might have led to investigation bias from attending physician. These methodological drawbacks might have affected the significance of our conclusions. A large-scale prospective study is required to confirm the findings of the current study.

In conclusion, we showed that the discontinuation of lenvatinib due to AEs was more frequent in elderly patients than in younger patients, which led to a lower dose intensity and objective response of lenvatinib in elderly patients than in the younger patients. Ultimately, elderly patients tended to show a shorter PFS than younger patients. The management of AEs, especially appetite loss and fatigue, is crucial for ensuring the adherence and maintaining a good dose intensity of lenvatinib among elderly HCC patients.

Footnotes

Acknowledgements

The authors thank the staff of Division of Gastroenterology and Hepatology, the Jikei University Daisan Hospital, the Jikei University Kashiwa Hospital, and Jikei University School of Medicine for their cooperation in this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Akiyoshi Kinoshita

References

Global Burden of Disease Cancer Collaboration, Fitzmaurice

Abate

Abbasi

, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived With disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: a systematic analysis for the global burden of disease study. JAMA Oncol 2019; 5: 1749–1768

European Association for the Study of the Liver. EASL Clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2018; 69: 182–236

Ministry of Health, Labor and Welfare 2019. In: Japanese simple life table. 2019. https://www.mhlw.go.jp/toukei/saikin/hw/life/life19/index.html Accessed 7 March 2021

Tateishi

Uchino

Fujiwara

, et al. A nationwide survey on non-B, non-C hepatocellular carcinoma in Japan: 2011–2015 update. J Gastroenterol 2019; 54: 367–376

Frith

Jones

Newton

. Chronic liver disease in an ageing population. Age Ageing 2009; 38: 11–18.

Gan

Chitturi

Farrell

. Mechanisms and implications of age-related changes in the liver: nonalcoholic fatty liver disease in the elderly. Curr Gerontol Geriatr Res 2011; 2011: 831536.

Carrion

Martin

. Viral hepatitis in the elderly. Am J Gastroenterol 2012; 107: 691–697.

Kinoshita

Koike

Nishino

. Clinical features and prognosis of elderly patients with hepatocellular carcinoma not indicated for surgical resection. Geriatr Gerontol Int 2017; 17: 189–201.

Kudo

Finn

Qin

, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 2018; 391: 1163–1173

10.

Hiraoka

Kumada

Kariyama

, et al. Therapeutic potential of lenvatinib for unresectable hepatocellular carcinoma in clinical practice: multicenter analysis. Hepatol Res 2019; 49: 111–117

11.

Takeda

Nishijima

Nasu

, et al. Long-term antitumor effect of lenvatinib on unresectable hepatocellular carcinoma with portal vein invasion. Hepatol Res 2019; 49: 594–599

12.

Tada

Kumada

Hiraoka

, et al. Safety and efficacy of lenvatinib in elderly patients with unresectable hepatocellular carcinoma: a multicenter analysis with propensity score matching. Hepatol Res 2020; 50: 75–83

13.

Pugh

Murray-Lyon

Dawson

, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60: 646–649

14.

Hiraoka

Michitaka

Kumada

, et al. Validation and potential of albumin-bilirubin grade and prognostication in a nationwide survey of 46,681 hepatocellular carcinoma patients in Japan: the need for a more detailed evaluation of hepatic function. Liver Cancer 2017; 6: 325–336

15.

Eso

Nakano

Mishima

, et al. Dose intensity/body surface area ratio is a novel marker useful for predicting response to lenvatinib against hepatocellular carcinoma. Cancers (Basel) 2019; 12: 49

16.

Lencioni

Llovet

. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 2010; 30: 52–60

17.

Shimose

Iwamoto

Niizeki

, et al. Clinical significance of adverse events for patients with unresectable hepatocellular carcinoma treated with lenvatinib: a multicenter retrospective study. Cancers (Basel) 2020; 12: 1867

18.

Ohki

Sato

Kondo

, et al. Impact of adverse events on the progression-free survival of patients with advanced hepatocellular carcinoma treated with lenvatinib: a multicenter retrospective study. Drugs Real World Outcomes 2020; 7: 141–149

19.

Hiraoka

Kumada

Atsukawa

, et al. Prognostic factor of lenvatinib for unresectable hepatocellular carcinoma in real-world conditions-multicenter analysis. Cancer Med 2019; 8: 3719–3728

20.

Ikeda

Kudo

Kawazoe

, et al. Phase 2 study of lenvatinib in patients with advanced hepatocellular carcinoma. J Gastroenterol 2017; 52: 512–519

21.

Tamai

Hayato

Hojo

, et al. Dose finding of lenvatinib in subjects With advanced hepatocellular carcinoma based on population pharmacokinetic and exposure-response analyses. J Clin Pharmacol 2017; 57: 1138–1147

22.

Hata

Suetsugu

Egashira

, et al. Association of lenvatinib plasma concentration with clinical efficacy and adverse events in patients with hepatocellular carcinoma. Cancer Chemother Pharmacol 2020; 86: 803–813

23.

Yamazaki

Iwasaki

Takasaki

, et al. Efficacy and tolerability of initial low-dose lenvatinib to treat differentiated thyroid cancer. Medicine (Baltimore) 2019; 98: e14774.

24.

Endo

Kuroda

Kanazawa

, et al. Impact of grip strength in patients with unresectable hepatocellular carcinoma treated with lenvatinib. Cancers (Basel) 2020; 12: 2146

25.

Uojima

Chuma

Tanaka

, et al. Skeletal muscle mass influences tolerability and prognosis in hepatocellular carcinoma patients treated with lenvatinib. Liver Cancer 2020; 9: 193–206

26.

Hiraoka

Kumada

Kariyama

, et al. Clinical importance of muscle volume in lenvatinib treatment for hepatocellular carcinoma: analysis adjusted with inverse probability weighting. J Gastroenterol Hepatol 2021; 36: 1812–1819

27.

Shomura

Okabe

Sato

, et al. Hypothyroidism is a predictive factor for better clinical outcomes in patients with advanced hepatocellular carcinoma undergoing lenvatinib therapy. Cancers (Basel) 2020; 12: 3078

28.

Takahashi

Moriguchi

Seko

, et al. Impact of relative dose intensity of early-phase lenvatinib treatment on therapeutic response in hepatocellular carcinoma. Anticancer Res 2019; 39: 5149–5156

29.

Kirino

Tsuchiya

Kurosaki

, et al. Relative dose intensity over the first four weeks of lenvatinib therapy is a factor of favorable response and overall survival in patients with unresectable hepatocellular carcinoma. PLoS One 2020; 15: e0231828

30.

Colombo

De Leo

Di Stefano

, et al. Primary adrenal insufficiency during lenvatinib or vandetanib and improvement of fatigue after cortisone acetate therapy. J Clin Endocrinol Metab 2019; 104: 779–784

31.

Iwamoto

Suzuki

Shimose

, et al. Weekends-Off lenvatinib for unresectable hepatocellular carcinoma improves therapeutic response and tolerability toward adverse events. Cancers (Basel) 2020; 12: 1010

32.

Okubo

Ando

Ishizuka

, et al. Carnitine insufficiency is associated with fatigue during lenvatinib treatment in patients with hepatocellular carcinoma. PLoS One 2020; 15: e0229772