Lung Cancer 89 (2015) 161–166
Contents lists available at ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Renal insufficiency is the leading cause of double maintenance (bevacizumab and pemetrexed) discontinuation for toxicity to advanced non-small cell lung cancer in real world setting Marion Sassier a,∗ , Audrey Emmanuelle Dugué a , Bénédicte Clarisse a , Paul Lesueur a , Virginie Avrillon b , Acya Bizieux-Thaminy c , Jean-Bernard Auliac d , Laure Kaluzinski e , Julie Tillon f , Gilles Robinet g , Hervé Le Caer h , Isabelle Monnet i , Anne Madroszyk j , Gabriella Boza k , Lionel Falchero l , Pierre Fournel m , Thomas Egenod n , Anne-Claire Toffart o , Nathalie Leiber a , Pascal Do a , Radj Gervais a a
Centre Franc¸ois Baclesse, Caen F-14000, France Centre Léon Bérard, Lyon F-69008, France c Centre Hospitalier Départemental, La Roche Sur Yon F-85000, France d Centre Hospitalier F. Quesnay, Mantes La Jolie F-78200, France e Centre Hospitalier Public du Cotentin, Cherbourg-Octeville F-50100, France f Centre Hospitalier, Dieppe F-76200, France g Centre Hospitalier Régional Universitaire, Brest F-29200, France h Centre Hospitalier, Draguignan F-83300, France i Centre Hospitalier Intercommunal, Créteil F-94000, France j Institut Paoli Calmettes, Marseille F-13009, France k Centre Hospitalier, Vire F-14500, France l Centre Hospitalier, Villefranche Sur Saône F-69400, France m Institut de Cancérologie Lucien Neuwirth, Saint-Priest en Jarez F-42270, France n Centre Hospitalier Universitaire, Limoges F-87000, France o Centre Hospitalier Universitaire, Grenoble F-38000, France b
a r t i c l e
i n f o
Article history: Received 23 December 2014 Received in revised form 6 May 2015 Accepted 7 May 2015 Keywords: Renal insufficiency Bevacizumab Pemetrexed Non-small cell lung cancer
a b s t r a c t Objectives: In advanced non-small cell lung cancer (NSCLC), maintenance therapy has emerged as a novel therapeutic reference for patients with non-progressive disease after platinum-based induction chemotherapy. However, the use of double maintenance (DM) with pemetrexed and bevacizumab is still being evaluated in terms of its clinical benefits and safety profile. The objective of this retrospective study was to describe the reasons for DM discontinuation in a real-world setting. Materials and methods: Patients with advanced non-squamous NSCLC were eligible if they had received at least 4 cycles of induction chemotherapy, followed by at least 1 cycle of DM. They were identified by using the oncology pharmacy database of 17 French centers. Results: Eighty-one patients who began a DM after induction chemotherapy were identified from September 2009 to April 2013. Among the 78 patients who had stopped DM at the time of the analysis, the main reasons for discontinuation were disease progression (42%), adverse events (33%), and personal preference (8%). The most frequent toxicity responsible for DM discontinuation was renal insufficiency (54%). Conclusion: For patients with advanced NSCLC eligible for DM therapy, a particular attention should be paid to potential renal failure. Kidney function should be monitored carefully before and during DM to detect and manage early this adverse event. © 2015 Elsevier Ireland Ltd. All rights reserved.
∗ Corresponding author at: Centre Franc¸ois Baclesse, Unité de Recherche Clinique, 3 Av. Général Harris, BP 5026, 14076 Caen Cedex 05, France. Tel.: +33 2 31 45 50 02; fax: +33 2 31 45 51 58. E-mail addresses: [email protected] (M. Sassier), [email protected] (A.E. Dugué), [email protected] (B. Clarisse), [email protected] (P. Lesueur), [email protected] (R. Gervais). http://dx.doi.org/10.1016/j.lungcan.2015.05.005 0169-5002/© 2015 Elsevier Ireland Ltd. All rights reserved.
162
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
1. Introduction In advanced non-small cell lung cancer (NSCLC), first-line induction by platinum-based doublet chemotherapy followed by maintenance therapy has emerged as a novel therapeutic reference for patients with non-progressive disease after induction, i.e. about 60% of patients. Two separate maintenance strategies have evolved: the introduction of an additional agent immediately after completion of induction chemotherapy (switch maintenance), or the continuation of the non-platinum partner initially introduced during induction (continuation maintenance) [1]. Both strategies have been shown to improve Progression Free Survival (PFS) and/or Overall Survival (OS) for patients with at least stable disease after induction chemotherapy. Bevacizumab, an anti-vascular endothelial growth factor (VEGF) inhibitor monoclonal antibody, and pemetrexed, a multitarget antifolate agent, have both proven to be important components in the first-line induction and maintenance setting. Thus, bevacizumab was associated to first-line chemotherapy followed by bevacizumab continuation maintenance until disease progression in two phase III studies [2,3]. The JMEN study examined pemetrexed as switch maintenance therapy after platinum-based chemotherapy [4], while the PARAMOUNT study compared continuation maintenance by pemetrexed with best supportive care [5,6]. Recent phase III studies have also investigated the value of double maintenance (DM) by bevacizumab and pemetrexed administered every 21 days. In particular, the PointBreak study compared a regimen of carboplatin, pemetrexed plus bevacizumab followed by pemetrexed plus bevacizumab maintenance, with a regimen of carboplatin, paclitaxel plus bevacizumab followed by bevacizumab maintenance [7]. From the randomization before induction, median OS was similar in both arms and the PFS was statistically significantly longer in the DM arm. In a preplanned analysis among the maintenance population, PFS and OS were improved in patients receiving DM vs. bevacizumab alone (respectively 8.6 vs. 6.9 months and 17.7 vs. 15.7 months). The AVAPERL study compared continuation maintenance with bevacizumab monotherapy and bevacizumab plus pemetrexed. The DM arm had a significantly longer PFS, as measured from the time of randomization after induction (7.4 vs. 3.7 months; P < 0.001) and from the start of induction by platin, pemetrexed, and bevacizumab (10.2 vs. 6.6 months; P < 0.001). OS from the start of induction was also numerically longer by nearly 4 months for DM, although the difference was not significant (19.8 vs. 15.9 months; P = 0.32) [8,9]. The value of maintenance therapy is now statistically established, but the use of DM is still controversial while waiting for additional studies assessing its clinical benefits and safety profile. Currently, DM is not standard practice but remains an option used by some physicians with selected patients. The purpose of our retrospective study was to describe in a real-world setting the frequency of DM discontinuation for adverse events, together with the prevalence and type of toxicities occurring during DM.
2. Patients and methods 2.1. Eligibility All patients older than 18 years with advanced non-squamous NSCLC were eligible if they had received at least 4 cycles of induction chemotherapy with platinum, pemetrexed, and bevacizumab, followed by at least 1 cycle of DM (pemetrexed and bevacizumab). Patients were identified from September 2009 to April 2013 by using the oncology pharmacy database of 17 French centers.
The study was conducted in accordance with the Good Clinical Practice guidelines. All patients were monitored for survival and DM discontinuation until June 2014. 2.2. Endpoints All charts were retrospectively reviewed to collect clinical and laboratory data, including renal function (before induction and before DM) assessed by estimated glomerular filtration rate (eGFR), calculated by the abbreviated Modification of Diet in Renal Disease (aMDRD) formula from serum creatinine. In this multicenter observational retrospective study, the primary objective was to describe the reasons for DM discontinuation defined as definitive interruption of any of two or both drugs. The associated reason was determined by the referring physician and classified as “disease progression”, “adverse event” or “other reason”. Treatment following DM discontinuation for adverse events and OS were also assessed. 2.3. Statistical analysis Categorical data were described by frequencies and percentages, while numerical data were described by mean and standard deviation or median and extreme values if necessary. OS was estimated with the Kaplan–Meier method. The probability to stop DM for a given adverse event was estimated over time by the Fine & Gray method, assuming that stopping for another reason was a competing risk [10]. The time point of discontinuation was set at 20 days after the last injection. Because the proportional hazard assumption was not verified, we only tested the probability of discontinuing DM for renal insufficiency by a Chi squared or Fisher’s exact test, according to the following factors: initial metastatic status, older than 65 years, history of cardiovascular disease, impaired (grade 1 and above according to the National Cancer Institute Common Terminology Criteria for Adverse Events CTCAE v.4) renal function before induction and before DM, and at least one cycle of cisplatin during induction chemotherapy. Alpha risk was set at 0.05 for each statistical analysis. All figures and analyses were produced by using the R software [11]. 3. Results The study population consisted of 81 patients identified in the participating centers (Table 1). Thirty-two (46%) had an impaired renal function (only grade 1 or 2) before induction treatment. Table 1 Baseline characteristics (n = 81).
Age (y) at induction, median, range Male gender, n (%) Disease stage at diagnostic 1A 1B 2B 3A 3B 4 Adenocarcinoma Renal function before induction aMDRD eGFR (ml/min/1.73 m2 ) ≥90 (gr 0) 89–60 (gr 1) 59–30 (gr 2) Months from diagnosis to induction start
Description
Number of patients
58 [28–71] 44 (54%)
81 81 81
2 (2%) 2 (2%) 1 (1%) 3 (4%) 5 (6%) 68 (84%) 79 (98%)
37 (54%) 31 (45%) 1 (1%) 1.2 [0–42]
81 69
77
y, year; eGFR, estimated glomerular filtration rate; aMDRD, abbreviated Modification of Diet in Renal Disease.
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
163
Table 2 Induction treatment (N = 81). Description Type of induction cycles 60 (74%) Cisplatin only Carboplatin only 12 (15%) Both 9 (11%) Number of induction cycles 4 55 (68%) 3 (4%) 5 23 (28%) 6 Response after induction Stable disease 32 (40%) 43 (53%) Partial response 2 (2%) Complete response 4 (5%) Unkown (not progressive) Renal function before DM aMDRD eGFR (ml/min/1.73 m2 ) 23 (32%) ≥90 (gr 0) 42 (58%) 89–60 (gr 1) 7 (10%) 59–30 (gr 2) 9 (–) ND Data are presented as number (percent). aMDRD, abbreviated Modification of Diet in Renal Disease; eGFR, estimated glomerular filtration rate; DM, double maintenance; ND, no data.
Characteristics of induction treatment are presented in Table 2: patients received a median of 4 cycles (range: 4–6) of chemotherapy with platinum (carboplatin or cisplatin), pemetrexed, and bevacizumab. Among this population who began DM, disease response after induction phase was objective for 45 patients (55%), stable for 32 patients (40%), and unknown (not progressive) for 4 patients (5%). Before DM, 49 (68%) of patients had an impaired renal function (from whom 29 already had a renal function impairment before induction). Concerning DM, 3 patients were still receiving it at the time of the analysis and had already received 18, 27, and 29 cycles respectively. The 78 patients who had stopped DM had received between 1 and 31 cycles (median: 4.5) of bevacizumab and pemetrexed during the DM phase, corresponding to a median of 3.5 months (range: 0.7–24.8). Reasons for discontinuation were (1) progression (42%), (2) adverse event (33%), (3) personal preference (8%), or other (17%, mainly for surgery or radiotherapy). Reported adverse events resulting in discontinuation of DM were renal insufficiency (14/26; 54%), edema (3/26; 11%), hematologic toxicities (2/26; 8%), asthenia (2/26; 8%), high blood pressure (1/26; 4%), and other toxicities (4/26; 16%). As the most frequent toxicity responsible for DM discontinuation was renal insufficiency, we present the probability over time of stopping DM for renal insufficiency, for another toxicity or for another reason in Fig. 1. The overall median time to discontinuation was 7.3 months (95% CI: 6.2–9.8) and 3.6 months (95% CI: 2.9–5.0) from induction and DM start, respectively. As the proportional hazard assumption was not verified, only the proportion of DM discontinuation for renal insufficiency was tested. The occurrence of a renal impairment before DM was significantly linked to a higher probability of discontinuing the DM for renal insufficiency, whereas no other factor was found to be significantly related (Table 3). Time to DM discontinuation is shown on Fig. 2, according to have an impaired renal function. After discontinuation of DM for renal insufficiency, 50% (7/14) of patients have continued to receive bevacizumab alone, 3 patients began a second line and 4 patients had a treatment break. None of these patients continued the treatment with pemetrexed alone. Median follow-up was 18.3 months from initiation of induction (range: 4.6–36 months) and 37 (46%) patients died. Overall survival (OS) is shown in Fig. 3. Median OS from induction start was 23.5 months (95% CI: 18.3–Inf) in the whole cohort, 22.5 months (95% CI: 17.9–Inf) in patients with stage IV disease at diagnosis, and was not reached (95% CI: 28.3–Inf) in the others. Median OS from DM
Fig. 1. Cumulative probability, along time, of stopping double maintenance (DM) for renal insufficiency, other toxicity or other reason (progression, patient’s desire, death, and other) while considering each motive as a competing risk; ongoing patients were censored. N = 81 patients.
initiation was 20.1 months (95% CI: 15.2–Inf) in the whole cohort, 18.7 months (95% CI: 12.9–Inf) in patients with stage IV disease at diagnosis, and 23.4 months (95% CI: 23.4–Inf) in the others. 4. Discussion This study in a real life setting provides insight into the tolerance of DM with pemetrexed and bevacizumab after induction chemotherapy with platinum, pemetrexed, and bevacizumab in advanced NSCLC. In the two DM studies to date, the most common adverse events in the DM arm were nausea, hypertension, asthenia, and hematologic toxicities, whatever the grade [7,8]. Table 3 Univariate analysis between several factors and DM discontinuation for renal insufficiency. DM discontinuation for renal insufficiency
Initial metastatic status M0 M1 Age ≤65 >65 History of cardiovascular disease No Yes At least 1 course of cisplatin during induction Yes Carboplatin only Renal function before induction Impaired (gr ≥ 1) Normal (gr 0) Renal function before DM Impaired (gr ≥ 1) Normal (gr 0) Switch to carboplatin during induction Yes No
n (%)
N
4 (31) 10 (16)
13 64
11 (16) 3 (27)
67 11
6 (18) 2 (50)
33 4
P value* 0.24
0.41
0.2
1 2 (17) 12 (18)
12 66
7 (24) 6 (16)
29 37
12 (26) 1 (4)
46 23
0.42
0.047
1 1 (13) 13 (19)
8 70
*P values were assessed by Chi-squared or Fisher’s exact test as appropriate. Data are presented as number (percent). DM, double maintenance.
164
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
Fig. 2. Cumulative probability of stopping for Renal Insufficiency (RI) from double maintenance (DM) start: (A) for patients with normal (solid line, N = 37) or impaired (dotted line, N = 32) renal function at baseline, (B) for patients with normal (solid line, N = 23) or impaired (dotted line, N = 49) renal function before double maintenance; after considering other reasons of discontinuation as a competing risk (solid and dotted grey lines respectively); ongoing patients were censored.
The results show that toxicity is responsible for DM discontinuation in one third of patients and that renal insufficiency represents more than a half of these cases. Our findings highlighted that renal function impairment before DM was significantly related to an increased risk of DM discontinuation for renal insufficiency. In our population, impaired renal function was noticed for almost half of patients before induction and two thirds of patients before DM. Such observations were consistent with the prevalence of renal insufficiency among lung cancer patients reported by LaunayVacher et al. [12]. Amazingly, this toxicity has not been highlighted in DM clinical studies. Nevertheless, in the Pointbreak and Avaperl studies, discontinuation of DM was defined as the cessation of both treatments, while a switch to single maintenance (pemetrexed or bevacizumab) was considered as a discontinuation in our study. This may explain why renal toxicity was not emphasized in these DM trials. In addition, we did not use the same definition of DM discontinuation as the others so no comparison can be made. Furthermore, all patients received 4 cycles with platinum in the other studies, while 32% in
our study received more than 4 cycles during induction. Park et al. showed that the administration of four cycles of induction is a currently accepted standard and that increasing the number of courses carries a risk of renal toxicity [13]. The occurrence of renal insufficiency may be multifactorial, perhaps induced by pemetrexed and/or bevacizumab but also by concomitant treatments or comorbidities. The recent PARAMOUNT study is the only trial that reported renal toxicity during pemetrexed maintenance. It demonstrated an alteration of the glomerular filtration (grade 1 or 2) in 4.2% of patients with pemetrexed maintenance and an increase in creatinine (grade 1 or 2) in 2.8% of patients [6]. In the pemetrexed arm, discontinuation for renal insufficiency was noted in one quarter of the elderly patients (≥70 years) and in more than one third of the others (
Contents lists available at ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Renal insufficiency is the leading cause of double maintenance (bevacizumab and pemetrexed) discontinuation for toxicity to advanced non-small cell lung cancer in real world setting Marion Sassier a,∗ , Audrey Emmanuelle Dugué a , Bénédicte Clarisse a , Paul Lesueur a , Virginie Avrillon b , Acya Bizieux-Thaminy c , Jean-Bernard Auliac d , Laure Kaluzinski e , Julie Tillon f , Gilles Robinet g , Hervé Le Caer h , Isabelle Monnet i , Anne Madroszyk j , Gabriella Boza k , Lionel Falchero l , Pierre Fournel m , Thomas Egenod n , Anne-Claire Toffart o , Nathalie Leiber a , Pascal Do a , Radj Gervais a a
Centre Franc¸ois Baclesse, Caen F-14000, France Centre Léon Bérard, Lyon F-69008, France c Centre Hospitalier Départemental, La Roche Sur Yon F-85000, France d Centre Hospitalier F. Quesnay, Mantes La Jolie F-78200, France e Centre Hospitalier Public du Cotentin, Cherbourg-Octeville F-50100, France f Centre Hospitalier, Dieppe F-76200, France g Centre Hospitalier Régional Universitaire, Brest F-29200, France h Centre Hospitalier, Draguignan F-83300, France i Centre Hospitalier Intercommunal, Créteil F-94000, France j Institut Paoli Calmettes, Marseille F-13009, France k Centre Hospitalier, Vire F-14500, France l Centre Hospitalier, Villefranche Sur Saône F-69400, France m Institut de Cancérologie Lucien Neuwirth, Saint-Priest en Jarez F-42270, France n Centre Hospitalier Universitaire, Limoges F-87000, France o Centre Hospitalier Universitaire, Grenoble F-38000, France b
a r t i c l e
i n f o
Article history: Received 23 December 2014 Received in revised form 6 May 2015 Accepted 7 May 2015 Keywords: Renal insufficiency Bevacizumab Pemetrexed Non-small cell lung cancer
a b s t r a c t Objectives: In advanced non-small cell lung cancer (NSCLC), maintenance therapy has emerged as a novel therapeutic reference for patients with non-progressive disease after platinum-based induction chemotherapy. However, the use of double maintenance (DM) with pemetrexed and bevacizumab is still being evaluated in terms of its clinical benefits and safety profile. The objective of this retrospective study was to describe the reasons for DM discontinuation in a real-world setting. Materials and methods: Patients with advanced non-squamous NSCLC were eligible if they had received at least 4 cycles of induction chemotherapy, followed by at least 1 cycle of DM. They were identified by using the oncology pharmacy database of 17 French centers. Results: Eighty-one patients who began a DM after induction chemotherapy were identified from September 2009 to April 2013. Among the 78 patients who had stopped DM at the time of the analysis, the main reasons for discontinuation were disease progression (42%), adverse events (33%), and personal preference (8%). The most frequent toxicity responsible for DM discontinuation was renal insufficiency (54%). Conclusion: For patients with advanced NSCLC eligible for DM therapy, a particular attention should be paid to potential renal failure. Kidney function should be monitored carefully before and during DM to detect and manage early this adverse event. © 2015 Elsevier Ireland Ltd. All rights reserved.
∗ Corresponding author at: Centre Franc¸ois Baclesse, Unité de Recherche Clinique, 3 Av. Général Harris, BP 5026, 14076 Caen Cedex 05, France. Tel.: +33 2 31 45 50 02; fax: +33 2 31 45 51 58. E-mail addresses: [email protected] (M. Sassier), [email protected] (A.E. Dugué), [email protected] (B. Clarisse), [email protected] (P. Lesueur), [email protected] (R. Gervais). http://dx.doi.org/10.1016/j.lungcan.2015.05.005 0169-5002/© 2015 Elsevier Ireland Ltd. All rights reserved.
162
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
1. Introduction In advanced non-small cell lung cancer (NSCLC), first-line induction by platinum-based doublet chemotherapy followed by maintenance therapy has emerged as a novel therapeutic reference for patients with non-progressive disease after induction, i.e. about 60% of patients. Two separate maintenance strategies have evolved: the introduction of an additional agent immediately after completion of induction chemotherapy (switch maintenance), or the continuation of the non-platinum partner initially introduced during induction (continuation maintenance) [1]. Both strategies have been shown to improve Progression Free Survival (PFS) and/or Overall Survival (OS) for patients with at least stable disease after induction chemotherapy. Bevacizumab, an anti-vascular endothelial growth factor (VEGF) inhibitor monoclonal antibody, and pemetrexed, a multitarget antifolate agent, have both proven to be important components in the first-line induction and maintenance setting. Thus, bevacizumab was associated to first-line chemotherapy followed by bevacizumab continuation maintenance until disease progression in two phase III studies [2,3]. The JMEN study examined pemetrexed as switch maintenance therapy after platinum-based chemotherapy [4], while the PARAMOUNT study compared continuation maintenance by pemetrexed with best supportive care [5,6]. Recent phase III studies have also investigated the value of double maintenance (DM) by bevacizumab and pemetrexed administered every 21 days. In particular, the PointBreak study compared a regimen of carboplatin, pemetrexed plus bevacizumab followed by pemetrexed plus bevacizumab maintenance, with a regimen of carboplatin, paclitaxel plus bevacizumab followed by bevacizumab maintenance [7]. From the randomization before induction, median OS was similar in both arms and the PFS was statistically significantly longer in the DM arm. In a preplanned analysis among the maintenance population, PFS and OS were improved in patients receiving DM vs. bevacizumab alone (respectively 8.6 vs. 6.9 months and 17.7 vs. 15.7 months). The AVAPERL study compared continuation maintenance with bevacizumab monotherapy and bevacizumab plus pemetrexed. The DM arm had a significantly longer PFS, as measured from the time of randomization after induction (7.4 vs. 3.7 months; P < 0.001) and from the start of induction by platin, pemetrexed, and bevacizumab (10.2 vs. 6.6 months; P < 0.001). OS from the start of induction was also numerically longer by nearly 4 months for DM, although the difference was not significant (19.8 vs. 15.9 months; P = 0.32) [8,9]. The value of maintenance therapy is now statistically established, but the use of DM is still controversial while waiting for additional studies assessing its clinical benefits and safety profile. Currently, DM is not standard practice but remains an option used by some physicians with selected patients. The purpose of our retrospective study was to describe in a real-world setting the frequency of DM discontinuation for adverse events, together with the prevalence and type of toxicities occurring during DM.
2. Patients and methods 2.1. Eligibility All patients older than 18 years with advanced non-squamous NSCLC were eligible if they had received at least 4 cycles of induction chemotherapy with platinum, pemetrexed, and bevacizumab, followed by at least 1 cycle of DM (pemetrexed and bevacizumab). Patients were identified from September 2009 to April 2013 by using the oncology pharmacy database of 17 French centers.
The study was conducted in accordance with the Good Clinical Practice guidelines. All patients were monitored for survival and DM discontinuation until June 2014. 2.2. Endpoints All charts were retrospectively reviewed to collect clinical and laboratory data, including renal function (before induction and before DM) assessed by estimated glomerular filtration rate (eGFR), calculated by the abbreviated Modification of Diet in Renal Disease (aMDRD) formula from serum creatinine. In this multicenter observational retrospective study, the primary objective was to describe the reasons for DM discontinuation defined as definitive interruption of any of two or both drugs. The associated reason was determined by the referring physician and classified as “disease progression”, “adverse event” or “other reason”. Treatment following DM discontinuation for adverse events and OS were also assessed. 2.3. Statistical analysis Categorical data were described by frequencies and percentages, while numerical data were described by mean and standard deviation or median and extreme values if necessary. OS was estimated with the Kaplan–Meier method. The probability to stop DM for a given adverse event was estimated over time by the Fine & Gray method, assuming that stopping for another reason was a competing risk [10]. The time point of discontinuation was set at 20 days after the last injection. Because the proportional hazard assumption was not verified, we only tested the probability of discontinuing DM for renal insufficiency by a Chi squared or Fisher’s exact test, according to the following factors: initial metastatic status, older than 65 years, history of cardiovascular disease, impaired (grade 1 and above according to the National Cancer Institute Common Terminology Criteria for Adverse Events CTCAE v.4) renal function before induction and before DM, and at least one cycle of cisplatin during induction chemotherapy. Alpha risk was set at 0.05 for each statistical analysis. All figures and analyses were produced by using the R software [11]. 3. Results The study population consisted of 81 patients identified in the participating centers (Table 1). Thirty-two (46%) had an impaired renal function (only grade 1 or 2) before induction treatment. Table 1 Baseline characteristics (n = 81).
Age (y) at induction, median, range Male gender, n (%) Disease stage at diagnostic 1A 1B 2B 3A 3B 4 Adenocarcinoma Renal function before induction aMDRD eGFR (ml/min/1.73 m2 ) ≥90 (gr 0) 89–60 (gr 1) 59–30 (gr 2) Months from diagnosis to induction start
Description
Number of patients
58 [28–71] 44 (54%)
81 81 81
2 (2%) 2 (2%) 1 (1%) 3 (4%) 5 (6%) 68 (84%) 79 (98%)
37 (54%) 31 (45%) 1 (1%) 1.2 [0–42]
81 69
77
y, year; eGFR, estimated glomerular filtration rate; aMDRD, abbreviated Modification of Diet in Renal Disease.
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
163
Table 2 Induction treatment (N = 81). Description Type of induction cycles 60 (74%) Cisplatin only Carboplatin only 12 (15%) Both 9 (11%) Number of induction cycles 4 55 (68%) 3 (4%) 5 23 (28%) 6 Response after induction Stable disease 32 (40%) 43 (53%) Partial response 2 (2%) Complete response 4 (5%) Unkown (not progressive) Renal function before DM aMDRD eGFR (ml/min/1.73 m2 ) 23 (32%) ≥90 (gr 0) 42 (58%) 89–60 (gr 1) 7 (10%) 59–30 (gr 2) 9 (–) ND Data are presented as number (percent). aMDRD, abbreviated Modification of Diet in Renal Disease; eGFR, estimated glomerular filtration rate; DM, double maintenance; ND, no data.
Characteristics of induction treatment are presented in Table 2: patients received a median of 4 cycles (range: 4–6) of chemotherapy with platinum (carboplatin or cisplatin), pemetrexed, and bevacizumab. Among this population who began DM, disease response after induction phase was objective for 45 patients (55%), stable for 32 patients (40%), and unknown (not progressive) for 4 patients (5%). Before DM, 49 (68%) of patients had an impaired renal function (from whom 29 already had a renal function impairment before induction). Concerning DM, 3 patients were still receiving it at the time of the analysis and had already received 18, 27, and 29 cycles respectively. The 78 patients who had stopped DM had received between 1 and 31 cycles (median: 4.5) of bevacizumab and pemetrexed during the DM phase, corresponding to a median of 3.5 months (range: 0.7–24.8). Reasons for discontinuation were (1) progression (42%), (2) adverse event (33%), (3) personal preference (8%), or other (17%, mainly for surgery or radiotherapy). Reported adverse events resulting in discontinuation of DM were renal insufficiency (14/26; 54%), edema (3/26; 11%), hematologic toxicities (2/26; 8%), asthenia (2/26; 8%), high blood pressure (1/26; 4%), and other toxicities (4/26; 16%). As the most frequent toxicity responsible for DM discontinuation was renal insufficiency, we present the probability over time of stopping DM for renal insufficiency, for another toxicity or for another reason in Fig. 1. The overall median time to discontinuation was 7.3 months (95% CI: 6.2–9.8) and 3.6 months (95% CI: 2.9–5.0) from induction and DM start, respectively. As the proportional hazard assumption was not verified, only the proportion of DM discontinuation for renal insufficiency was tested. The occurrence of a renal impairment before DM was significantly linked to a higher probability of discontinuing the DM for renal insufficiency, whereas no other factor was found to be significantly related (Table 3). Time to DM discontinuation is shown on Fig. 2, according to have an impaired renal function. After discontinuation of DM for renal insufficiency, 50% (7/14) of patients have continued to receive bevacizumab alone, 3 patients began a second line and 4 patients had a treatment break. None of these patients continued the treatment with pemetrexed alone. Median follow-up was 18.3 months from initiation of induction (range: 4.6–36 months) and 37 (46%) patients died. Overall survival (OS) is shown in Fig. 3. Median OS from induction start was 23.5 months (95% CI: 18.3–Inf) in the whole cohort, 22.5 months (95% CI: 17.9–Inf) in patients with stage IV disease at diagnosis, and was not reached (95% CI: 28.3–Inf) in the others. Median OS from DM
Fig. 1. Cumulative probability, along time, of stopping double maintenance (DM) for renal insufficiency, other toxicity or other reason (progression, patient’s desire, death, and other) while considering each motive as a competing risk; ongoing patients were censored. N = 81 patients.
initiation was 20.1 months (95% CI: 15.2–Inf) in the whole cohort, 18.7 months (95% CI: 12.9–Inf) in patients with stage IV disease at diagnosis, and 23.4 months (95% CI: 23.4–Inf) in the others. 4. Discussion This study in a real life setting provides insight into the tolerance of DM with pemetrexed and bevacizumab after induction chemotherapy with platinum, pemetrexed, and bevacizumab in advanced NSCLC. In the two DM studies to date, the most common adverse events in the DM arm were nausea, hypertension, asthenia, and hematologic toxicities, whatever the grade [7,8]. Table 3 Univariate analysis between several factors and DM discontinuation for renal insufficiency. DM discontinuation for renal insufficiency
Initial metastatic status M0 M1 Age ≤65 >65 History of cardiovascular disease No Yes At least 1 course of cisplatin during induction Yes Carboplatin only Renal function before induction Impaired (gr ≥ 1) Normal (gr 0) Renal function before DM Impaired (gr ≥ 1) Normal (gr 0) Switch to carboplatin during induction Yes No
n (%)
N
4 (31) 10 (16)
13 64
11 (16) 3 (27)
67 11
6 (18) 2 (50)
33 4
P value* 0.24
0.41
0.2
1 2 (17) 12 (18)
12 66
7 (24) 6 (16)
29 37
12 (26) 1 (4)
46 23
0.42
0.047
1 1 (13) 13 (19)
8 70
*P values were assessed by Chi-squared or Fisher’s exact test as appropriate. Data are presented as number (percent). DM, double maintenance.
164
M. Sassier et al. / Lung Cancer 89 (2015) 161–166
Fig. 2. Cumulative probability of stopping for Renal Insufficiency (RI) from double maintenance (DM) start: (A) for patients with normal (solid line, N = 37) or impaired (dotted line, N = 32) renal function at baseline, (B) for patients with normal (solid line, N = 23) or impaired (dotted line, N = 49) renal function before double maintenance; after considering other reasons of discontinuation as a competing risk (solid and dotted grey lines respectively); ongoing patients were censored.
The results show that toxicity is responsible for DM discontinuation in one third of patients and that renal insufficiency represents more than a half of these cases. Our findings highlighted that renal function impairment before DM was significantly related to an increased risk of DM discontinuation for renal insufficiency. In our population, impaired renal function was noticed for almost half of patients before induction and two thirds of patients before DM. Such observations were consistent with the prevalence of renal insufficiency among lung cancer patients reported by LaunayVacher et al. [12]. Amazingly, this toxicity has not been highlighted in DM clinical studies. Nevertheless, in the Pointbreak and Avaperl studies, discontinuation of DM was defined as the cessation of both treatments, while a switch to single maintenance (pemetrexed or bevacizumab) was considered as a discontinuation in our study. This may explain why renal toxicity was not emphasized in these DM trials. In addition, we did not use the same definition of DM discontinuation as the others so no comparison can be made. Furthermore, all patients received 4 cycles with platinum in the other studies, while 32% in
our study received more than 4 cycles during induction. Park et al. showed that the administration of four cycles of induction is a currently accepted standard and that increasing the number of courses carries a risk of renal toxicity [13]. The occurrence of renal insufficiency may be multifactorial, perhaps induced by pemetrexed and/or bevacizumab but also by concomitant treatments or comorbidities. The recent PARAMOUNT study is the only trial that reported renal toxicity during pemetrexed maintenance. It demonstrated an alteration of the glomerular filtration (grade 1 or 2) in 4.2% of patients with pemetrexed maintenance and an increase in creatinine (grade 1 or 2) in 2.8% of patients [6]. In the pemetrexed arm, discontinuation for renal insufficiency was noted in one quarter of the elderly patients (≥70 years) and in more than one third of the others (
