JO U RN A L OF GE RI A TR IC O N COL O G Y 5 (2 0 1 4 ) 7 8 –8 8
Available online at www.sciencedirect.com
ScienceDirect
Review article
Bevacizumab in elderly patients with metastatic colorectal cancer Francesco Sclafani, David Cunningham⁎ The Royal Marsden NHS Foundation Trust, London and Surrey, United Kingdom
AR TIC LE I N FO
ABS TR ACT
Article history:
The progressively ageing population combined with an increased availability of antitumoural
Received 29 May 2013
agents has created a new, challenging therapeutic scenario for oncologists. Due to the lack of
Received in revised form 19 July 2013
evidence-based data on elderly patients it is uncertain whether the criteria for assessing the
Accepted 27 August 2013
risk/benefit ratio of treatment strategies in these patients coincide with those classically used
Available online 20 September 2013
for the general population. A critical reevaluation of the role and potential options of systemic chemotherapy in elderly patients with metastatic colorectal cancer (mCRC) is warranted as the
Keywords:
historical conservative approach of oncologists may have resulted in undertreatment of this
Bevacizumab
patient population. Bevacizumab was demonstrated to improve the outcome of mCRC
Elderly
patients when used in combination with standard first and second line chemotherapy.
Colorectal cancer
However, its toxicity profile including hypertension, thromboembolic events, haemorrhage and proteinuria may raise important concerns when this anti-angiogenic agent is used in elderly patients with comorbidities. In this review article we analyse the available evidence on the safety and effectiveness of bevacizumab in elderly mCRC patients. Based on the data from subgroup or pooled analysis of prospective trials, observational cohort studies, retrospective population-based studies and a single recent randomised phase III trial, we conclude that the clinical benefit and safety profile of bevacizumab in elderly patients are not significantly different from those reported in younger patients, with the exception of an increased risk of arterial thromboembolic events. Bevacizumab should therefore be considered as a potential therapeutic option for elderly patients with mCRC. © 2013 Elsevier Ltd. All rights reserved.
Contents 1. 2. 3. 4. 5. 6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . Initial Evidence from Unselected Populations in the Pivotal Trials Evidence from Subgroup Analyses of Prospective Trials . . . . Evidence from Pooled Analyses of Prospective Trials . . . . . . Evidence from Observational Cohort Studies . . . . . . . . . . Evidence from Prospective, Randomised, Clinical Trials . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
79 79 80 81 85 86
⁎ Corresponding author at: Department of Medicine, The Royal Marsden NHS Foundation Trust Hospital, Downs Road, Sutton, Surrey, SM2 5PT, United Kingdom. Tel.: +44 208 661 3156; fax: +44 208 643 9414. E-mail address: [email protected] (D. Cunningham). 1879-4068/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jgo.2013.08.006
79
J O U RN A L OF GE RI A TR IC O N CO LOG Y 5 ( 20 1 4 ) 7 8 –88
7. Conclusions. . . . . . . . . . . . . . . . . . Disclosures and Conflict of Interest Statements . Author Contributions Author Contributions . . Acknowledgements and Funding . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .
1.
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
Introduction
In 2008, over 1.2 million new cases of colorectal cancer (CRC) were diagnosed, 608,700 individuals died of this disease worldwide, and CRC represented the third leading cause of cancer death in females and the fourth in males.1 The incidence and mortality of CRC increases with age. In the US, 90% of new cases and 94% of deaths occur in individuals ≥50 years. Although over the last two decades an increasing number of new cases have been diagnosed among individuals 75 years) (n = 99, 21% of the study population), the survival outcomes were in line with those previously reported for the whole study population.20 Indeed, despite the use of a reduced dose of capecitabine in most of the elderly patients, the addition of bevacizumab was associated with a significant improvement in PFS (HR 0.53 for capecitabine versus capecitabine plus bevacizumab, HR 0.38 for capecitabine versus capecitabine and mitomycin plus bevacizumab). Moreover, an interaction test for response rate, PFS and OS between treatment and age groups showed that the bevacizumab effect was not different between patients > 75 years and patients ≤ 75 years. It is also worth noting that a trend toward a better survival with the use of bevacizumab was observed in the elderly population (Table 1). Although potentially affected by the small sample size, this finding is likely to partly reflect a stronger impact by first line treatment on OS in the subgroup of frail, elderly patients where the use of subsequent lines of therapy is limited. The AGITG MAX trial also provided useful information on the safety profile of bevacizumab in the elderly population. Interestingly, this trial included the prospective record of co-morbidities primarily relating to cardiovascular risk factors and diabetes, hypertension, ischemic heart disease and previous cerebrovascular accident/ transient ischemic attack were present in 15%, 56%, 15% and 10% of patients >75 years, respectively. However, no significant differences in bevacizumab-related toxicities by age were observed (Table 3) and the addition of bevacizumab to chemotherapy in the elderly population was associated with an increased rate of hand-and-foot syndrome only, likely reflecting the longer duration of treatment in the investigational arms. The CAIRO 2 trial was a randomised phase III study investigating the addition of cetuximab to combination chemotherapy with capecitabine and oxaliplatin (CAPOX) plus bevacizumab in previously untreated mCRC patients.21 The median age of patients in this trial was 62 for both arms and more than 60% of patients had an ECOG performance status of 0. The CAIRO 2 study failed to meet its endpoint as the addition of cetuximab to CAPOX plus bevacizumab resulted in a significant decrease in PFS and a poorer quality of life. A subsequent subgroup analysis investigated efficacy, safety, drug administration and quality of life in patients > 75, 70–75 and < 75 years.22 Given the negative result of the study and the detrimental effect associated with the use of cetuximab, this subgroup analysis was restricted to patients assigned to the control arm (n = 368). The effect of CAPOX plus bevacizumab was similar between the three age groups, in terms of response rate (38% versus 39% versus 48%, p = 0.335), disease control rate (100% versus 89% versus 93%, p = 0.313) and PFS (13.5 versus 9.6 versus 10.6 months, p = 0.908). However, elderly patients were found to have a trend toward a worse
81
J O U RN A L OF GE RI A TR IC O N CO LOG Y 5 ( 20 1 4 ) 7 8 –88
median OS (13.1 versus 17.6 versus 20.3 months, p = 0.063) (Table 1). Although the incidence of grade 3–4 toxicities was not significantly different for the three age groups (82% versus 72% versus 75%, p = 0.656), a lower and significant median number of cycles with CAPOX plus bevacizumab was administered to the elderly patients (6 versus 8.5 versus 10, p = 0.036). Moreover, a difference in the reasons for end of treatment was observed and elderly patients had to discontinue treatment due to unacceptable toxicity more often than did younger patients (35% versus 40% versus 19%). Interestingly, this finding would confirm previous evidence suggesting that, even in the presence of comparable tolerability, physicians are conservative when deciding on therapy in elderly patients23 and the threshold for discontinuation of treatment is much lower in elderly than in younger populations. Although the authors reported that no difference in the relative dose intensity of capecitabine and oxaliplatin was observed in the three age groups, the incidence of bevacizumab-related toxicities was not analysed and it is not known whether these toxicities may have played a determinant role in the earlier treatment discontinuation observed in the elderly groups. Alongside treatment tolerability, it is established that quality of life represents an important endpoint of treatment in elderly populations. There is a lack of information on the quality of life in the elderly compared with younger patients with mCRC. Interestingly the subgroup analysis of the CAIRO
2 trial showed that quality of life was not significantly different in the three age groups and a similar percentage of patients in these groups had a 10 point increase in quality of life (33% in patients > 75 years versus 15% in patients 70–75 years versus 17% in patients < 75 years, p = 0.588).
4. Evidence from Pooled Analyses of Prospective Trials One of the main limitations of the subgroup analyses of prospective controlled trials is the relatively small sample size. It is worth noting that elderly patients in the above discussed AGITG MAX and CAIRO 2 trials represented approximately 20% of the whole study populations. Overall, the information on the safety and efficacy of bevacizumab in the elderly provided by these two subgroup analyses was based on 99 and 81 patients, respectively. It is reasonable to hypothesise that the relatively small sample size may have negatively affected the power of the statistical analysis to demonstrate significant differences between elderly and younger patients. This inherent limitation of subgroup analysis was partly overcome by the availability of pooled data from different clinical trials. The first available pooled analysis investigating the safety and efficacy of bevacizumab in the elderly population used data from patients enrolled in the pivotal AVF2107g trial
Table 1 – Main studies investigating the effectiveness of adding bevacizumab to standard chemotherapy in elderly versus younger patients with metastatic colorectal cancer. Author
Trial
Age groups (yrs)
N (%)
Backbone chemotherapy
Line
RR (%)
Price20
AGITG MAX
≥75 75 70–75 80 years) did not differ significantly from that observed in younger patients (9.8 months).30 This result is especially interesting if we consider that the percentage of patients with an ECOG performance status ≥2 in the elderly group was significantly higher than that in the
Available online at www.sciencedirect.com
ScienceDirect
Review article
Bevacizumab in elderly patients with metastatic colorectal cancer Francesco Sclafani, David Cunningham⁎ The Royal Marsden NHS Foundation Trust, London and Surrey, United Kingdom
AR TIC LE I N FO
ABS TR ACT
Article history:
The progressively ageing population combined with an increased availability of antitumoural
Received 29 May 2013
agents has created a new, challenging therapeutic scenario for oncologists. Due to the lack of
Received in revised form 19 July 2013
evidence-based data on elderly patients it is uncertain whether the criteria for assessing the
Accepted 27 August 2013
risk/benefit ratio of treatment strategies in these patients coincide with those classically used
Available online 20 September 2013
for the general population. A critical reevaluation of the role and potential options of systemic chemotherapy in elderly patients with metastatic colorectal cancer (mCRC) is warranted as the
Keywords:
historical conservative approach of oncologists may have resulted in undertreatment of this
Bevacizumab
patient population. Bevacizumab was demonstrated to improve the outcome of mCRC
Elderly
patients when used in combination with standard first and second line chemotherapy.
Colorectal cancer
However, its toxicity profile including hypertension, thromboembolic events, haemorrhage and proteinuria may raise important concerns when this anti-angiogenic agent is used in elderly patients with comorbidities. In this review article we analyse the available evidence on the safety and effectiveness of bevacizumab in elderly mCRC patients. Based on the data from subgroup or pooled analysis of prospective trials, observational cohort studies, retrospective population-based studies and a single recent randomised phase III trial, we conclude that the clinical benefit and safety profile of bevacizumab in elderly patients are not significantly different from those reported in younger patients, with the exception of an increased risk of arterial thromboembolic events. Bevacizumab should therefore be considered as a potential therapeutic option for elderly patients with mCRC. © 2013 Elsevier Ltd. All rights reserved.
Contents 1. 2. 3. 4. 5. 6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . Initial Evidence from Unselected Populations in the Pivotal Trials Evidence from Subgroup Analyses of Prospective Trials . . . . Evidence from Pooled Analyses of Prospective Trials . . . . . . Evidence from Observational Cohort Studies . . . . . . . . . . Evidence from Prospective, Randomised, Clinical Trials . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
79 79 80 81 85 86
⁎ Corresponding author at: Department of Medicine, The Royal Marsden NHS Foundation Trust Hospital, Downs Road, Sutton, Surrey, SM2 5PT, United Kingdom. Tel.: +44 208 661 3156; fax: +44 208 643 9414. E-mail address: [email protected] (D. Cunningham). 1879-4068/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jgo.2013.08.006
79
J O U RN A L OF GE RI A TR IC O N CO LOG Y 5 ( 20 1 4 ) 7 8 –88
7. Conclusions. . . . . . . . . . . . . . . . . . Disclosures and Conflict of Interest Statements . Author Contributions Author Contributions . . Acknowledgements and Funding . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .
1.
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
Introduction
In 2008, over 1.2 million new cases of colorectal cancer (CRC) were diagnosed, 608,700 individuals died of this disease worldwide, and CRC represented the third leading cause of cancer death in females and the fourth in males.1 The incidence and mortality of CRC increases with age. In the US, 90% of new cases and 94% of deaths occur in individuals ≥50 years. Although over the last two decades an increasing number of new cases have been diagnosed among individuals 75 years) (n = 99, 21% of the study population), the survival outcomes were in line with those previously reported for the whole study population.20 Indeed, despite the use of a reduced dose of capecitabine in most of the elderly patients, the addition of bevacizumab was associated with a significant improvement in PFS (HR 0.53 for capecitabine versus capecitabine plus bevacizumab, HR 0.38 for capecitabine versus capecitabine and mitomycin plus bevacizumab). Moreover, an interaction test for response rate, PFS and OS between treatment and age groups showed that the bevacizumab effect was not different between patients > 75 years and patients ≤ 75 years. It is also worth noting that a trend toward a better survival with the use of bevacizumab was observed in the elderly population (Table 1). Although potentially affected by the small sample size, this finding is likely to partly reflect a stronger impact by first line treatment on OS in the subgroup of frail, elderly patients where the use of subsequent lines of therapy is limited. The AGITG MAX trial also provided useful information on the safety profile of bevacizumab in the elderly population. Interestingly, this trial included the prospective record of co-morbidities primarily relating to cardiovascular risk factors and diabetes, hypertension, ischemic heart disease and previous cerebrovascular accident/ transient ischemic attack were present in 15%, 56%, 15% and 10% of patients >75 years, respectively. However, no significant differences in bevacizumab-related toxicities by age were observed (Table 3) and the addition of bevacizumab to chemotherapy in the elderly population was associated with an increased rate of hand-and-foot syndrome only, likely reflecting the longer duration of treatment in the investigational arms. The CAIRO 2 trial was a randomised phase III study investigating the addition of cetuximab to combination chemotherapy with capecitabine and oxaliplatin (CAPOX) plus bevacizumab in previously untreated mCRC patients.21 The median age of patients in this trial was 62 for both arms and more than 60% of patients had an ECOG performance status of 0. The CAIRO 2 study failed to meet its endpoint as the addition of cetuximab to CAPOX plus bevacizumab resulted in a significant decrease in PFS and a poorer quality of life. A subsequent subgroup analysis investigated efficacy, safety, drug administration and quality of life in patients > 75, 70–75 and < 75 years.22 Given the negative result of the study and the detrimental effect associated with the use of cetuximab, this subgroup analysis was restricted to patients assigned to the control arm (n = 368). The effect of CAPOX plus bevacizumab was similar between the three age groups, in terms of response rate (38% versus 39% versus 48%, p = 0.335), disease control rate (100% versus 89% versus 93%, p = 0.313) and PFS (13.5 versus 9.6 versus 10.6 months, p = 0.908). However, elderly patients were found to have a trend toward a worse
81
J O U RN A L OF GE RI A TR IC O N CO LOG Y 5 ( 20 1 4 ) 7 8 –88
median OS (13.1 versus 17.6 versus 20.3 months, p = 0.063) (Table 1). Although the incidence of grade 3–4 toxicities was not significantly different for the three age groups (82% versus 72% versus 75%, p = 0.656), a lower and significant median number of cycles with CAPOX plus bevacizumab was administered to the elderly patients (6 versus 8.5 versus 10, p = 0.036). Moreover, a difference in the reasons for end of treatment was observed and elderly patients had to discontinue treatment due to unacceptable toxicity more often than did younger patients (35% versus 40% versus 19%). Interestingly, this finding would confirm previous evidence suggesting that, even in the presence of comparable tolerability, physicians are conservative when deciding on therapy in elderly patients23 and the threshold for discontinuation of treatment is much lower in elderly than in younger populations. Although the authors reported that no difference in the relative dose intensity of capecitabine and oxaliplatin was observed in the three age groups, the incidence of bevacizumab-related toxicities was not analysed and it is not known whether these toxicities may have played a determinant role in the earlier treatment discontinuation observed in the elderly groups. Alongside treatment tolerability, it is established that quality of life represents an important endpoint of treatment in elderly populations. There is a lack of information on the quality of life in the elderly compared with younger patients with mCRC. Interestingly the subgroup analysis of the CAIRO
2 trial showed that quality of life was not significantly different in the three age groups and a similar percentage of patients in these groups had a 10 point increase in quality of life (33% in patients > 75 years versus 15% in patients 70–75 years versus 17% in patients < 75 years, p = 0.588).
4. Evidence from Pooled Analyses of Prospective Trials One of the main limitations of the subgroup analyses of prospective controlled trials is the relatively small sample size. It is worth noting that elderly patients in the above discussed AGITG MAX and CAIRO 2 trials represented approximately 20% of the whole study populations. Overall, the information on the safety and efficacy of bevacizumab in the elderly provided by these two subgroup analyses was based on 99 and 81 patients, respectively. It is reasonable to hypothesise that the relatively small sample size may have negatively affected the power of the statistical analysis to demonstrate significant differences between elderly and younger patients. This inherent limitation of subgroup analysis was partly overcome by the availability of pooled data from different clinical trials. The first available pooled analysis investigating the safety and efficacy of bevacizumab in the elderly population used data from patients enrolled in the pivotal AVF2107g trial
Table 1 – Main studies investigating the effectiveness of adding bevacizumab to standard chemotherapy in elderly versus younger patients with metastatic colorectal cancer. Author
Trial
Age groups (yrs)
N (%)
Backbone chemotherapy
Line
RR (%)
Price20
AGITG MAX
≥75 75 70–75 80 years) did not differ significantly from that observed in younger patients (9.8 months).30 This result is especially interesting if we consider that the percentage of patients with an ECOG performance status ≥2 in the elderly group was significantly higher than that in the
