Efficacy and safety of anti-vascular endothelial growth factor therapies in older patients for first line treatment of metastatic renal cell carcinoma.

BACKGROUND: immunotherapy became the first line treatment of metastatic renal cell carcinoma (mRCC). Nevertheless, a better understanding of the specificities of targeted therapies (TT) in the elderly population could be helpful in order to improve the management of mRCC in this population. The aim of this retrospective study was to assess efficacy and safety of sunitinib and sorafenib used as first-line TT in 70 years older patients compared to younger patients.
METHODS: Data were retrospectively collected for all consecutive mRCC patients receiving first line TT treatment by sunitinib or sorafenib for mRCC from January 2006 to November 2017. Patients were divided into two groups according to the age using a cut-off at 70 years old. Median progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier method and compared using log-rank test.
RESULTS: In total, 147 patients were included; 94 (63.9%) were <70 and 53 (36.1%) were 70 years old or more. First line TT used was sunitinib in 123 (83.7%) patients or sorafenib in 24 (16.3%) patients. Median PFS was 8 months for elderly patients vs. 6 in younger group (P=0.68). Median OS were 26 vs. 36 months (P=0.08). Severe induced toxicity was more frequent among elderly patients: 34 (64.2%) vs. 46 patients (48.9%) (P=0.07). Rate of treatment discontinuation due to toxicity was 22 patients (23.4%) in younger group vs. 28 patients (52.8%) in the elderly group (P=0.0005). Results were similar in the 2 groups regarding the type of toxicities.
CONCLUSIONS: Our results suggest similar efficacy of anti-vascular endothelial growth factor (VEGF) agents as first-line treatment for mRCC among younger and older patients with an age cut-off of 70 years. Safety results suggest that these drugs can be safely used for older patients with a need of caution regarding toxicity prevention. 2021 Translational Andrology and Urology. All rights reserved.

Introduction

Renal cell carcinoma (RCC) represents 2–3% of malignant cancers with an increasing incidence worldwide, especially in western countries (1). Today immunotherapies represents the main option for first line treatment (2). However, targeted therapies (TT) [anti-vascular endothelial growth factor (VEGF) therapies] remain essential for the subsequent lines in the treatment sequence of mRCC patients. The efficacy and safety of these drugs were largely documented in pivotal trials. Different studies have suggested that age remains a risk factor and most of diagnoses are made between 60 and 70 years old (3).

Elderly patients represent a significant proportion of the RCC population as 331,582 prevalent cases older than 70 in 2018 (4). Regarding metastatic disease, 20–30% of all RCC are metastatic (mRCC) at diagnosis while 20% of patients with initially localized RCC will encounter metastatic recurrence during follow-up. Specific data regarding the management of TT in elderly patients are scarce (5). Whereas they represent a particular population with an increased incidence of comorbidities as chronic kidney disease (CKD) or liver dysfunction, as well as polymedication. These parameters may impact the efficacy and safety of TT in this population as well as limit the use of immunotherapies (6). A better understanding of the specificities regarding the efficacy and toxicity of TT in the elderly population could be helpful in order to improve the management of mRCC in this population. The aim of this retrospective study was to assess efficacy and safety of sunitinib and sorafenib used as first-line TT in patients >70 years old compared to younger patients.

We present the following article in accordance with the STROBE reporting checklist (available at http://dx.doi.org/10.21037/tau-20-1481) (7).

Methods

Study population

All consecutive mRCC patients >18 years receiving first line TT treatment by sunitinib or sorafenib for mRCC from January 2006 to November 2017 were included in a unique center. All clinical and tumor baseline characteristics were retrospectively collected for each patient, including toxicity related to treatment. Sunitinib was orally administered for 4 consecutive weeks, followed by a 2-week break (dose of 50 mg/day), whereas sorafenib was orally administered continuously at 400 mg twice daily. Patients underwent physical examination associated with blood cell count and serum chemistry tests at baseline and on days 14 and 28 of the first treatment cycle and at least monthly during TT exposure. Results were analyzed according to normal institutional values. For patients treated by sunitinib, an electrocardiogram (ECG) was performed at baseline. All the toxicities were graded according to NCI-CTCAE, version 3.0. When grade 3 non-hematological toxicity occurred, treatment was interrupted until toxicity was resolved and the dose was reduced thereafter by one level at the physician’s discretion. Treatment was definitively stopped in case of grade 4 non-hematological toxicity. Toxicity was classified as ‘mild to moderate’ corresponding to grade 0 to 2 toxicity or ‘severe’ corresponding to grade 3 to 4 toxicity. Baseline comorbidities were collected and analyzed according to the Charlson score (8). The prognostic score of each patient was calculated from the 6 clinical and biological data composing the International Metastatic Renal Cell Carcinoma Database Consortium model (IMDC) (9). These data were Karnofsky performance status (KPS), time from diagnosis to first-line TT, haemoglobin concentration, neutrophil count, platelet count, and serum calcium concentration. Follow-up was based on a physical examination and CT-scan performed every 3 months. Response was evaluated using Response Evaluation Criteria in Solid Tumors (10). There were no missing data for any of the patients included in the study. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by institutional ethics committee of CHU de Rouen (No.: E2021-22) and individual consent for this retrospective analysis was waived.

Statistical analysis

Patients were divided into two groups according to the age with a cut-off at 70 years old. This cut-off was previously used in analysis of elderly sunitinib-treated patients or sorafenib-treated patient with mRCC (11-13). Patients characteristics were compared between the age groups using Chi-2 test as appropriate. Overall survival (OS) was defined as time from treatment initiation to date of death because of any cause or censored at the last follow-up. Progression-free survival (PFS) was defined as time from treatment initiation to date of disease progression according to the RECIST criteria or clinically. Median PFS and OS for comparison of sunitinib or sorafenib-treated patients aged 70 or >70 years were estimated by Kaplan-Meier method and compared using log-rank test. The limit of significance was fixed at P value less than 0.05. Statistical analyses were carried out using Medcalc version 12.0 (Medcalc software bvba, Ostend, Belgium).

Results

Baseline characteristics

A total of 167 patients with a first line treatment for mRCC were reviewed. Only those treated with sunitinib or sorafenib were included, representing 147 patients (). 94 (63.9%) were men; 94 (63.9%) were <70 and 53 (36.1%) were >70 years (). The median age in the <70 and the >70 age group was 58.5 years (range, 38–69 years) and 74 years (range, 70–87 years) respectively. First line TT used was sunitinib in 123 (83.7%) patients or sorafenib in 24 (16.3%) patients. The predominant histology was clear cell renal carcinoma (131 patients, 89%). As reported in , baseline characteristics were mainly similar. However, significant differences between the two groups were found. Patients >70 years had a higher incidence of CKD (defined by GFR <60 mL/min). More patients <70 years presented thrombocytosis and/or high neutrophil count than in the >70 years old age group (respectively 14.8% vs. 3.77%, P=0.0431). Intermediate IMDC score were significantly higher in the >70 years old group. Treatment related toxicity was more frequently the reason for treatment discontinuation in the elderly group: 28 (52.8%) patients vs. 22 (24.4%) in the younger patients group (P=0.002). The median follow-up was 21 months for the both groups.

Figure 1

Flow chart. mRCC, metastatic renal cell carcinoma.

Table 1

Clinical and tumor baseline characteristics

Characteristics	<70 years old (n=94), n (%)	>70 years old (n=53), n (%)	P
Male	65 (69.1)	29 (54.7)	0.08
Karnofsky index <80%	34 (36.2)	24 (45.3)	0.27
Comorbidity
Charlson score >2	38 (40.4)	30 (56.6)	0.06
Cardiovascular	25 (26.6)	27 (50.9)	0.0053*
Diabetes	14 (14.9)	9 (17.0)	0.92
Obesity	22 (23.4)	8 (15.1)	0.32
Clear cell carcinoma	85 (90.4)	46 (86.8)	0.81
Metastatic at diagnosis	46 (48.9)	20 (37.7)	0.2
Brain metastasis	11 (11.7)	2 (3.8)	0.1
Time from diagnosis to targeted therapy <1 year	56 (59.6)	24 (45.3)	0.09
Type of targeted therapy
Sunitinib	75 (79.8)	48 (90.6)	0.14
Sorafenib	19 (20.2)	5 (9.4)	0.14
Anemia	29 (30.9)	16 (30.2)	0.93
Thrombocytosis	14 (14.9)	2 (3.8)	0.038*
High neutrophil count	19 (20.2)	5 (9.4)	0.043*
Hypercalcemia	8 (8.5)	1 (1.9)	0.1
LDH >1.5 ULN	14 (14.9)	5 (9.4)	0.34
CKD (clearance <60 mL/min)	28 (29.8)	35 (66.0)	<0.0001*
Hypo-albumin	29 (30.9)	19 (35.8)	0.65
IMDC prognostic score			0.057
Good	27 (28.7)	13 (24.5)
Intermediate	39 (41.5)	32 (60.4)
Poor	28 (29.8)	8 (15.1)

*, P<0.05. TT, targeted therapy; LDH, lactates dehydrogenase; ULN, upper limit of normal.

Survival analysis

The survival was similar between the two groups. The median PFS was 8 months in the >70 years old groups vs. 6 in the younger patients group (P=0.68) (). Similarly, median OS were not significantly different between the two groups (respectively 26 vs. 36 months, P=0.08) ().

Figure 2

Relationship between progression probability and time according to the age.

Figure 3

Relationship between survival probability and time according to the age.

Toxicity

A total of 80 patients (54.4%) encountered one or more severe treatment-induced toxicity event; 79.5% of the major toxicity events occurred during the first five months following treatment-initiation (). The incidence of severe induced toxicity was found with a trend to be more frequent among elderly patients: 34 patients (64.2%) vs. 46 patients (48.9%) P=0.07. For 50 patients (34%) the first line TT was definitely stopped due to treatment-induced toxicity. A significant difference between the two groups was found for the rate of treatment discontinuation due to toxicity with respectively 22 patients (23.4%) in the younger group vs. 28 patients (52.8%) in the elderly group (P=0.0005). Finally, results were similar in the 2 groups regarding to the type of induced toxicities. Especially, no significant difference was found for dermatologic and digestive toxicity as well as fatigue and hypertension ().

Figure 4

Occurrence of first severe toxicity event from treatment initiation.

Table 2

Detailed toxicities according to the age

Variables	<70 years old (n=94), n (%)	>70 years old (n=53), n (%)	P
Grade 3/4 toxicity events	46 (48.9)	34 (64.2)	0.07
Dermatologic (all)	60 (63.8)	29 (54.7)	0.36
Digestive (all)	53 (56.4)	31 (58.5)	0.94
Fatigue	35 (37.2)	27 (50.9)	0.14
Cardiovascular (all)	20 (21.3)	12 (22.6)	1
hematological	10 (10.6)	9 (17.0)	0.39
Renal (all)	13 (13.8)	5 (9.4)	0.6
Hypothyroidism	8 (8.5)	3 (5.7)	0.76
Neurologic (all)	1 (1.1)	4 (7.5)	0.1
Respiratory (all)	2 (2.1)	0 (0)	0.74

Discussion

As mentioned, literature data about the use of tyrosine kinase inhibitor (TKI) for the treatment of mRCC in elderly patients is scarce. study emphasized the fact that despite high cancer frequency in this population there is a lack of data as well as a fear of side effect and a minimization of potential benefit (5). Data come from clinical trials involving a selected population not representative of elderly population. In this study we report further informations about this topic. Especially we showed that the efficacy and the safety of sunitinib and sorafenib for patients older than 70 could be comparable to younger patients as we found no difference in terms of OS and PFS as well as toxicity.

Regarding the survival outcomes, no significant difference was found for PFS with median at 8 vs. 6 months (P=0.68) or median OS, respectively 26 vs. 36 months, (P=0.08).

An evolution in the management of toxicities could be underline, with better results especially in elderly population. These results were close to those previously reported on sunitinib or sorafenib treatment in this elderly population. In this context, Hutson et al. found no difference regarding the efficacy of sunitinib in terms of survival between age groups similar to ours (11) (). Similarly, Poprach et al. reported more recently comparable results in a retrospective cohort of 1,315 patients treated with sunitinib (14). In addition, using a different age cut-off did not suggested different results: using a cut off age at 65 years old, Gore et al. also described similar survival between the 2 groups with sunitinib (15). Regarding the molecule used, our cohort was mainly based on sunitinib 83.7%. Only 16.3% of the patients received sorafenib as first line treatment, this was due to a temporary authorization to use sorafenib in 2006 in this indication at our institution while waiting for sunitinib to be available from mid-2007. Sunitinib was then used as the standard treatment for first line mRCC. Literature data regarding the use of Sorafenib in elderly patients are even rarer and mainly issued from second line treatment studies. Eisen et al. reported similar outcomes for sorafenib as a second-line treatment between younger and older patients (>70 years) (12). Similar results were also found by Bukowski et al. and Procopio et al. (13,16) ().

Table 3

Main publications about targeted therapies for mRCC among elderly patients

Study	Population	design	Control	Size	Portion of elderly patients	Comparison to younger patients	Details	Severe toxicity
Sunitinib
Poprach et al. [2016]	First line clear cell and non-clear cell	Retrospective multicentric	N/A	1,315	299 (>70 yo)	Similar PFS, but different OS	PFS (months): 10.8 (95% CI, 9.8–11.8) vs. 8.8 (7.2–10.4) (P=0.321); OS (months): 31.9 (27.9–35.9) vs. 26.3 (21.3–31.2) (P=0.044)	No difference in severe toxicity. Treatment discontinuation due to AEs rate in elderly patients higher
Gore et al. [2015]	First and second-lines clear cell and non-clear cell	Retrospective multicentric	N/A	4,543	1,485 (>65 yo)	Similar PFS, OS and ORR	PFS (months): 9.2 (95% CI, 8.5–9.8) vs. 10.1 (95% CI, 8.8–10.9); OS (months): 18.8 (95% CI, 17.4–19.8) vs. 18.1 (95% CI, 16.5–20.3)	Incidence of non-hematological grade 3/4 AEs higher in >65 y group
Hutson et al. [2014]	First and second-lines clear cell and non-clear cell	Retrospective multicentric	N/A	1,059	202 (>70 yo)	Similar PFS and OS	PFS (months): 9.9 vs. 11(95% CI, 0.73–1.09), HR (0.89); OS (months): 23.6 vs. 25.6 (95% CI, 0.74–1.18), HR (0.93)	Higher AEs incidence in elderly patients: fatigue (60% vs. 69%); cough (20% vs. 29%). peripheral edema (17% vs. 27%); anemia (18% vs. 25%); decreased appetite (13% vs. 29%); and thrombocytopenia (16% vs. 25%) (P<0.05)
Sorafenib
Procopio et al. [2013]	First and second-lines clear cell and non-clear cell	Retrospective multicentric	N/A	4,584	1,382 (>65 yo), 559 (>75 yo)	Similar safety	N/A	No difference in any-grade AEs between subgroups
Bukowski et al. [2010]	First and second-lines clear cell and non-clear cell	Retrospective multicentric	N/A	2,496	736 (>70 yo)	Similar PFS and OS	PFS (weeks): 35(95% CI, 33–46) vs. 42 (95% CI, 36–48); OS (weeks): 50 (95% CI, 47–53) vs. 46 (95% CI, 42–53)	No difference in toxivities
Eisen et al. (2008)	Second-lines clear cell	Retrospective multicentric	Placebo	903	115 (>70 yo)	Similar PFS	PFS (weeks): 23.9 (95% CI, 0.47–0.66), HR (0.55) vs. 26.3 (95% CI, 0.26–0.69), HR (0.43)	Higher incidence of grade 3 toxicities and treatment discontinuation in elderly patients
Pazopanib
Vogelzang et al. [2017]	First lines clear cell and non-clear cell	Retrospective multicentric	Sunitinib	1,711	526 pazopanib, 1,185 sunitinib (>65 yo)	Higher OS with pazopanib	OS (months): 18.2 vs. 14.6 (P=0.015)	N/A
Axitinib
Motzer et al. [2013]	Second-lines clear cell	Prospective multicentric	Sorafenib	723	N/A	N/A	N/A	N/A

mRCC, metastatic renal cell carcinoma; PFS, progression free survival; OS, overall survival; ORR: odd ratio; AE, adverse events; N/A, non-applicable; yo, years old.

Our results showed no significant difference in terms of incidence of treatment-induced toxicity; 54.4% of our population encountered severe toxicity events with a majority of them occurring during the first five months following treatment initiation (). A trend was noticed with a higher incidence of these events among elderly patient than younger patients, respectively 64.2% vs. 48.9% (P=0.07). This trend could be related to the higher rate of CKD in the older group. We previously reported the role of GFR less than 60 in mRCC as risk factor of TT-induced toxicities (17). Moreover, severe TT-induced toxicities were identified as an independent prognostic factor of TTF and OS. In addition, it is interesting to note that severe toxicity was significantly more likely to cause treatment discontinuation in the elderly group (52.8% vs. 24.4%, P=0.002). This suggest that even if the severe toxicity is not significantly more frequent in the elderly group, its occurrence significantly impacts more this population. This fact highlights the potential need to be cautious in the management of these drugs for this fragile population. The improvement made during the last years regarding the management of tolerability of anti-VEGF agents, inherited from the more than a decade experience of the use of these drugs are of real interest to optimize the tolerability of the treatment for the elderly patients. Hutson et al. reported comparable safety results between <70 and >70 years old groups (11). Nevertheless, they noticed some exceptions as a higher incidence of fatigue, cough, peripheral edema, anemia, decreased appetite, and thrombocytopenia in older patients (). Similarly, safety results of Poprach et al. and Gore et al. were not different between the age groups (14,15). This last one observed a higher incidence for non-hematological grade 3–4 adverse events (AEs) in >70 years old group (). Regarding sorafenib-related toxicity the results reported in the literature were rare but consistent. Although Eisen et al. observed higher grade 3 toxicity and treatment-discontinuation rates, no significant difference was described between younger and older patients (12). These results were in agreement with two retrospective studies of Procopio et al. and Bukowski et al. finding no difference regarding treatment-induced toxicity (13,16) (). In we reported literature data obtained from studies assessing the outcomes of anti-VEGF agent in elderly patients (18,19).

Our results confirmed the recent review from Neuzillet et al. regarding the specificities of cancer management in elderly patients. This review was also highlighting the possibility to safely use anti-VEGF in the elderly subject at the same doses as in the young subject with increased vigilance for side effects (20).

The main limitation of our study was of course its retrospective design with a small population because of unicentric characteristic. It was a trend regarding the overall survival, however, the comparison of the OS between two groups with different ages is difficult since the baseline life expectancy is not comparable between the groups. Cancer specific survival would have been more relevant, however, due to the retrospective nature of the study, this information was not available to allow us to assess it.

Conclusions

Our results suggest similar efficacy of anti-VEGF agents as first-line treatment for mRCC among younger and older patients with an age cut-off of 70 years. In addition, results about safety were close between the groups in term of incidence of severe toxicity events. However, the occurrence of such severe toxicity was significantly more impacting the treatment scheme for the older patients with higher rate of treatment discontinuation. This suggest that these drugs can be safely used for the older patients with a need of caution regarding toxicity prevention.

The article’s supplementary files as