Maintenance or consolidation therapy in small-cell lung cancer: A systematic review and meta-analysis

Lung Cancer 70 (2010) 119–128

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

Review

Maintenance or consolidation therapy in small-cell lung cancer: A systematic review and meta-analysis Antonio Rossi a,∗ , Marina Chiara Garassino b , Michela Cinquini c , Paola Sburlati b , Massimo Di Maio d , Gabriella Farina b , Cesare Gridelli a , Valter Torri c a

Division of Medical Oncology, “S.G. Moscati” Hospital, Città Ospedaliera, Contrada Amoretta 8, 83100 Avellino, Italy Oncology Department, “Fatebenefratelli and Oftalmico” Hospital, Milan, Italy c Laboratory of New Drug Development Strategies, Oncology Department, “Mario Negri” Institute, Milan, Italy d Clinical Trials Unit, National Cancer Institute, “G. Pascale” Foundation, Naples, Italy b

a r t i c l e

i n f o

Article history: Received 9 May 2009 Received in revised form 9 September 2009 Accepted 1 February 2010 Keywords: SCLC Maintenance therapy Consolidation therapy Meta-analysis Targeted therapy Immunotherapy Chemotherapy

a b s t r a c t Objective: To assess the role of maintenance or consolidation therapy in the treatment of small-cell lung cancer (SCLC), a meta-analysis of all published randomized clinical trials (RCTs) was performed in order to provide an overall meta-analysis and indirectly compare the effect of chemotherapy, interferons, and other biologic agents. Methods: Electronic databases were searched for publication reporting of RCTs comparing maintenance or consolidation therapy versus placebo or follow-up alone until December 2008. Hazard ratios (HRs) for progression-free survival (PFS) and overall survival (OS), with their relative 95% confidence intervals (CI), were derived. In the calculation of HRs, the “no maintenance” arm served as a reference. The a priori value of p < 0.05 was chosen as significant level for statistical tests. Results: Twenty-one RCTs, encompassing 3,688 patients, were eligible for the present analysis: 11 RCTs employing chemotherapy, 6 interferons (4 alpha and 2 gamma), and 4 other biological agents. Overall, no statistical advantage in OS (HR 0.93, 95% CI 0.87–1.00; p = 0.05) or in PFS (HR 0.98, 95% CI 0.91–1.06; p = 0.63) was reported for maintenance or consolidation therapy. Statistical evidence of different effects among the four types of therapy was detected for OS (2 test for heterogeneity: 8.07 [3 df]; p = 0.04), but not for PFS. A statistically significant reduction of mortality was detected in those studies assessing the efficacy of chemotherapy (HR 0.89, 95% CI 0.81–0.98; p = 0.02) and of interferon-alpha (HR 0.78, 95% CI 0.64–0.96; p = 0.02). Conclusions: The maintenance or the consolidation approach failed to improve the outcomes of SCLC. A survival advantage is suggested for maintenance chemotherapy and interferon-alpha, but its clinical impact needs to be confirmed by further studies. © 2010 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Small-cell lung cancer (SCLC) is characterized by a rapid doubling time, high growth fraction, and the early development of widespread metastases [1]. The therapeutic approach depends on disease extension at diagnosis. In fact, two main SCLC disease stages are identified: the limited disease (LD)-SCLC, defined as a tumor that is confined to one hemithorax and involves ipsilateral, mediastinal, or supraclavicular lymph nodes, and is located within a single radiation port, and which accounts for about one-third of cases; the remaining patients are diagnosed with extensive disease (ED), where the tumor is not confined to one hemithorax or has malignant pleural effusion [2]. In patients achieving a complete

∗ Corresponding author. Tel.: +39 0825 203573; fax: +39 0825 203556. E-mail address: arossi [email protected] (A. Rossi). 0169-5002/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2010.02.001

response, chemotherapy with concurrent thoracic irradiation was followed by prophylactic cranial irradiation (PCI), and this may be considered the standard of care for LD-SCLC. With this approach, an overall survival (OS) of 16–24 months can be reasonably expected, with 12–26% of patients surviving beyond 5 years [3–5]. In ED-SCLC achieving a complete response, chemotherapy was followed by PCI reporting an OS ranging from 7 to 12 months, with <5% of patients living beyond 2 years [3–5]. To improve this outcome, several therapeutic approaches have been tested. Among these, maintenance or consolidation treatments have been extensively studied with contrasting results. A systematic review with a meta-analysis evaluated the role of maintenance/consolidation chemotherapy. A total of 14 relevant randomized clinical trials (RCTs), encompassing 2,550 patients, reported an increase in 1-year (by 9%; from 30 to 39%) and 2-year (by 4%; from 10 to 14%) survival with maintenance/consolidation chemotherapy compared to standard treatment. The odds ratios

120

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

(ORs) for 1- and 2-year OS were 0.67 (95% confidence interval [CI] 0.56–0.79; p = 0.001), and again 0.67 (95% CI 0.53–0.86; p = 0.001). Likewise, 1- and 2-year progression-free survival (PFS) were better with maintenance/consolidation chemotherapy, with ORs of 0.49 (95% CI 0.37–0.63; p = 0.001), and 0.64 (95% CI 0.45–0.92; p = 0.015). However, some of these trials showed an increased toxicity when maintenance/consolidation chemotherapy was administered [6]. Maintenance/consolidation therapy has been also tested by including biologic agents. A possible advantage by using these agents would be their potentially better tolerability compared with conventional chemotherapy, better target selectivity, availability for chronic treatment, and, in some cases, oral administration, which would make them ideal for maintenance therapy. Therefore, it was considered important to produce a thorough picture of efficacy related to different pharmacological strategies, by performing a meta-analysis of the published data in order to assess the efficacy, in terms of OS and PFS, of the maintenance therapy, and to verify if some specific subgroup of therapy (namely chemotherapy, interferon-alpha, interferon-gamma, and other biological therapies) can have a different impact on the target outcomes.

2. Methods 2.1. Identification of eligible trials The first literature search was carried out in January 2008 and updated in December 2008. The PubMed, Medline, CancerLit, and Embase databases and Cochrane Central Register of Controlled Trials were searched for publications reporting of RCTs addressing the issue of maintenance or consolidation therapy versus placebo or follow-up alone for the treatment of SCLC. The following keywords were used in all the possible combinations: “small cell carcinoma”, “SCLC”, “maintenance therapy”, “consolidation therapy”, “lung cancer”, “oat-cell carcinoma”, and “clinical trial”. Also abstracts presented at the American Society of Clinical Oncology (ASCO), the European Society for Medical Oncology (ESMO), the European Cancer Conference (ECCO), and the International Association for the Study of Lung Cancer (IASLC) web searches up to December 2008 were analyzed. In order to correctly define trial eligibility, maintenance and consolidation therapies were defined as follows: “maintenance treatment” is the administration of additional different drugs at the end of a defined number of induction chemotherapy cycles in patients achieving a non-progressive disease. Maintenance should be prolonged until evidence of progressive disease or unacceptable toxicity. On the other hand, consolidation is the prolongation of the induction treatment with an a priori defined time and/or number of courses [7,8]. Maintenance/consolidation treatment consists of either a chemotherapeutic or a biologic agent; in addition, it consists both of drugs included in the induction regimen or other non-cross-resistant agents. Moreover, if a noncross-resistant agent, different from that used as induction, is administered as maintenance/consolidation treatment, this should also be considered as an early second-line treatment. In conclusion, our goal was to explore the effect of continuing treatment after standard induction therapy. We did not make any distinction between maintenance and consolidation approaches and between the use of a drug previously included or not in the induction therapy. Therefore, we use only the term “maintenance treatment”. We evaluated only the trials in which the randomization was done after any induction chemotherapy excluding the studies in which patients ab initio were randomized to receive a different number of chemotherapy courses.

All data were reviewed and then computed, first independently and then collaboratively, by two investigators.

2.2. Outcome definition We considered maintenance therapy as experimental arms, and placebo or control as comparators. Primary outcomes were OS, defined as the time between randomization and death for any cause, or the date of last follow-up visit for alive patients, and PFS, defined as the time between randomization and progression of disease, death without progression, or the date of last followup visit for patients alive without progression. Considering that this meta-analysis includes several different treatments with various related toxicities, we considered treatment compliance as secondary endpoint. Treatment compliance was defined as the percentage of patients who stopped maintenance therapy due to toxicity or patient’s refusal, as the best toxicity indicator available.

2.3. Statistical issues p and CI were extracted from each trial hazard ratio (HR) for treatment effect. Parmar’s method [9] was used for extracting the individual HRs from each trial. Test for treatment differences in subgroups and for interaction between each factor and treatment were calculated. Heterogeneity explained by inconsistency across study results was measured by I2 statistic [10]. The significance of the Q (Chi-squared) statistic was judged at p < 0.05. The I-squared statistic indicates the percent variability due to between-study (or inter-study) variability, as opposed to withinstudy (or intra-study) variability. An I-squared value greater than 50% was considered to be large. A fixed-effects mode was used, but in the case of unexplained statistical evidence for heterogeneity effect sizes, a random effect model was also used in order to incorporate heterogeneity among studies. This was not intended as a substitute for a thorough investigation of heterogeneity, but primarily for circumstances where heterogeneity cannot be explained [11]. Funnel plots for OS were also produced for exploratory data analysis regarding publication bias. Results are depicted in all figures as conventional meta-analysis forest plots, where HR < 1 means lower rate of events in the maintenance arm. RevMan 5, a software for preparing Cochrane reviews, was used for producing forest plot figures and computing interaction tests and log-rank tests for each singular subgroup of patients [12]. A further analysis was also considered for evaluating the main endpoints, OS and PFS, for the four groups defined by maintenance chemotherapy, interferon-alpha, interferon-gamma, and other biologic agents. Meta-analysis for specific subgroups of patients (ED versus LD) and toxicity was planned too, but was not performed due to paucity of data.

2.4. Assessment of methodological quality of included studies Each report was evaluated for quality by two authors. The scale devised by the Oxford group [13] was used to assess study quality. Since the focus was on the assessment of the potential, for bias we selected only the appropriate point of the scale system, namely randomization, allocation concealment, blinding and intention-totreat (ITT) analysis. Disagreement between reviewers regarding the evaluation allocated to each trial was resolved by discussion or a third reviewer.

Table 1 Description of trial of maintenance chemotherapy in small-cell lung cancer. Start of the accrual

End of the accrual

Induction therapy

Maintenance therapy

Study population

No. pts

PFS (months)

OS (months)

Cullen, 1986 [26]

March 1980

December 1984

VAC × 6 cycles

VAC × 8 cycles vs Control

ED in CR or GR

29 32

nr

12.2 p = 0.006 8.5

Cullen, 1986 [26]

March 1980

December 1984

VAC × 6 cycles

LD in CR or GR

16 16

nr

p = 0.13

Anonymous, 1989 [27]

June 1981

February 1985

ETO + CTX + MTX + VCR × 6 cycles + thoracic RT (in LD)

LD, ED in OR

131 134

nr

p = 0.27

Byrne, 1989 [28]

March 1981

October 1985

CDDP + ETO → CVM × 3 cycles + thoracic RT and PCI

VAC × 8 cycles vs Control Same regimen × 6 cycles vs Control CVM × 6 cycles vs Control

LD in CR

34 32

12.9 p = 0.48 16.3

14.1 p = 0.05 19.2

Ettinger, 1990 [29]

September 1982

October 1985

CAV × 6-8 cycles + PCI (in CR)

CAV up to 28 cycles vs Control

ED in CR

18 18

5.54a p = 0.15 2.77a

9.5 p = 0.09 6.8

Ettinger, 1990 [29]

September 1982

October 1985

CAV-HEM × 6-8 cycles + PCI (in CR)

CAV-HEM up to 28 cycles vs Control

ED in CR

25 25

5.22a p = 0.86 4.36a

11.3 p = 0.13 14.1

Giaccone, 1993 [30]

August 1982

July 1986

CDE × 5 cycles

CDE × 7 cycles vs Control

ED, LD Non-progressed

219 215

5.82 p = 0.0004 3.75

9.0 p = 0.7 9.5

Johnson, 1993 [31]

June 1982

October 1985

CAV × 6 cycles + thoracic RT

PE × 2 cycles vs Control

LD in OR

72 79

nr

21.1 p = 0.028 13.2

Beith, 1996 [32]

January 1981

August 1985

PE × 4 cycles + thoracic RT + PCI

VAC × 10 cycle vs Control

LD, ED Non-progressed

65 64

8.55b p = 0.099 5.31b

12.5 p = 0.636 12.0

Sculier, 1996 [33]

July 1990

May 1993

IVA/IVE × 6 cycles

ETO + VDS × 12 cycles vs Control

LD, ED in OR

45 46

5.77 p = 0.003 2.77

11.1 p = 0.10 8.8

Schiller, 2001 [34]

March 1995

January 1999

PE × 4 cycles

Topotecan × 4 cycles vs Control

ED Non-progressed

112 111

3.6 p < 0.001 2.3

8.9 p = 0.43 9.3

Hanna, 2002 [35]

September 1993

June 1998

VIP × 4 cycles

Oral ETO × 3 cycles vs Control

ED Non-progressed

72 72

8.23 p = 0.0018 6.5

12.2 p = 0.0704 11.2

Han, 2008 [36]

March 2003

April 2006

IP × 8 cycles

Irinotecan × 6 cycles vs Control

ED in OR

21 24

12.0 – 9.9

17.6 – 20.5

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

Author

pts = patients; PFS = progression-free survival; OS = overall survival; ED = extensive disease; LD = limited disease; CR = complete response, PR = partial response; OR = objective response; GR = good response (no unequivocal residual disease); nr = not reported; RT = radiotherapy; PCI = prophilactic cranial irradiation; ETO = etoposide; CTX = cyclophosphamide; MTX = methotrexate; VCR = vincristine; CDDP = cisplatin; VDS = vindesine; CAV (VAC) = vincristine, doxorubicin, cyclophosphamide; CVM = methotrexate, vincristine, cyclophosphamide; HEM = hexamethylmelamine, etoposide, methotrexate; CDE = cyclophosphamide, doxorubicin, etoposide; PE = cisplatin, etoposide; IVA = ifosfamide, etoposide, adryamicin; IVE = ifosfamide, etoposide, epirubicin; VIP = etoposide, ifosfamide, cisplatin; IP = irinotecan, cisplatin. a Time-to-progression. b Disease-free survival. 121

122

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

3.2. Overall survival

Fig. 1. Flow diagram for the selection of studies included in this meta-analysis: RCTs, randomized controlled trials; pts, patients.

3. Results Of the 33 RCTs evaluated at the initial stage, 12 were excluded due to the following reasons: 4 because of patients that were randomized to the maintenance therapy from the start of induction therapy [14–17]; 3 because of data included in another article [18–20]; 2 for lack of data [21,22]; 2 for the absence of a placebo/follow-up controlled arm [23,24]; 1 due to the report of data only about peripheral blood cells [25]. A total of 21 studies met the inclusion criteria: 11 RCTs concerning maintenance chemotherapy [26–36] (Table 1); 4 RCTs with maintenance interferon-alpha [37–40] and 2 with interferongamma [41,42] (Table 2); 4 employing maintenance therapy with other biological agents [43–46] (Table 3). Fig. 1 is the flow diagram for the selection of studies included in this meta-analysis.

3.1. Quality of trials All studies were randomized, but allocation concealment was adequate only for 5 studies [19,33,41,42,45] and for two [36,40] data were not available. In only three trials blinding was performed [43,45,46], in all the others this was not done. Analysis for ITT was clearly done only in trials using targeted agents [43–46] and in two studies using interferon [41,42]. Table 4 summarizes the final results. Inspection of the funnel plot did not suggest potential publication bias (Fig. 2).

The total number of patients in this meta-analysis was 3,688, and all were suitable for OS analysis. Overall survival did not significantly differ between the two arms (HR 0.93, 95% CI 0.87–1.00; p = 0.05) but a significant heterogeneity of effect among different types of treatment was detected (2 test for heterogeneity: 8.07 [3 df]; p = 0.04). The subgroup analysis reported the following results: for maintenance chemotherapy HR was 0.89 (95% CI 0.81–0.98; p = 0.02); for interferon-alpha HR was 0.78 (95% CI 0.64–0.96; p = 0.02); for interferon-gamma HR was 1.09 (95% CI 0.82–1.46; p = 0.54); and for other biological agents HR was 1.05 (95% CI 0.92–1.20; p = 0.45). However, heterogeneity explained by inconsistency across study results was detected, overall and within specific subgroups of treatment. In particular, heterogeneity was evident in maintenance chemotherapy (I2 statistic = 70.3%). Since it was not possible to try to explain this inconsistency by taking into account other factors, such as extension of disease or performance status, due to lack of data, a random effect model was used. In the chemotherapy group, estimates of effect using a random effect model produced HR 0.87 (95% CI 0.72–1.06; p = 0.17). As expected, the random effect model did not produce different estimates compared with fixed effect for interferon-alpha and interferon-gamma group, where I2 statistics was 0%, (interferon-alpha: HR 0.78, 95% CI 0.64–0.96; p = 0.02; interferon-gamma: HR 1.09, 95% CI 0.82–1.46; p = 0.54). In the group of other biologic therapies, using a random effect model showed that HR was 1.03 (95% CI 0.87–1.22; p = 0.75). Fig. 3 shows the forest plot for meta-analysis survival of patients treated with maintenance therapy. 3.3. Progression-free survival Progression-free survival data were available in 15 trials. The total of patients evaluated was 2,783. Progression-free survival did not significantly differ between the two arms (HR 0.98, 95% CI 0.91–1.06; p = 0.63). Again high inconsistency was found across studies (I2 statistics = 70.6%) but in this case no statistical heterogeneity of effects among treatments groups was detected (2 = 1.09 [3 df]; p = 0.78). Results per each subgroup were as follows: in the RCTs administering maintenance chemotherapy, HR was 0.96 (95% CI 0.85–1.08; p = 0.51); for interferon-alpha HR was 0.84 (95% CI 0.59–1.19; p = 0.32); for interferon-gamma HR was 1.01 (95% CI 0.75–1.34; p = 0.96); and for the other biological agents HR was 1.01 (95% CI 0.90–1.14; p = 0.83). The heterogeneity persisted significantly only for maintenance chemotherapy (I2 = 84.2%), possibly meaning that the different chemotherapeutics influenced mainly this value, with the result of being more homogeneous when using interferons or other biologic agents as maintenance approaches. In order to assess whether the different effect observed in PFS compared to OS was due to a selection of trials, analysis for OS in the same subgroup of studies available for PFS analysis was further performed. In this subset of studies, inconsistency of effects on OS across trials was of minor entity (I2 statistics = 13.6%). Results of OS analysis in the two subgroups with different number of trials (the RCTs administering maintenance chemotherapy and interferonalpha) were: HR 0.99 (95% CI 0.88–1.11; p = 0.84), and HR 0.85 (95% CI 0.60–1.21; p = 0.36), respectively. Therefore, differences in results when comparing OS with PFS analyses seem to depend on study selection. Fig. 4 shows the forest plot for the meta-analysis PFS of patients treated with maintenance therapy. 3.4. Toxicity

Fig. 2. Funnel plot of standard error by log hazard ratio—endpoint: overall survival. Egger’s regression: intercept: −0.01 (95% CI: −2.79 to 0.76); p = 0.25.

We extrapolated from each evaluable trial the percentage of patients who stopped maintenance therapy due to severe side

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

123

Fig. 3. Annotated forest plot for meta-analysis survival of patients treated with maintenance therapy. The graph shows hazard ratios (with 95% confidence intervals) for mortality among studies of patients allocated to maintenance therapy or control group. Summary measures are calculated using the fixed-effects model.

124

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

Fig. 4. Annotated forest plot for meta-analysis progression-free survival of patients treated with maintenance therapy. The graph shows hazard ratios (with 95% confidence intervals) for progression among studies of patients allocated to maintenance therapy or control group. Summary measures are calculated using the fixed-effects model.

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

125

Table 2 Description of trial of maintenance therapy with interferon in small-cell lung cancer. Author

Start of the accrual

End of the accrual

Induction therapy

Maintenance therapy

Study population

No. pts

TTP (months)

OS (months)

Mattson, 1992 [37]

January 1982

May 1990

CTX + VCR + ETO × 4 cycles + thoracic RT

Natural IFN-␣ vs Control

LD, ED in OR

91 87

nr

11 p = 0.048 10

Kelly, 1995 [38]

November 1989

October 1991

PE × 6 cycles + thoracic RT + PCI (in CR)

LD in OR

64 68

9a p = 0.72 10a

13 p = 0.77 16

Tummarello, 1997 [39]

June 1990

December 1995

LD, ED in CR

14 12

12 2 = 0.12 7

15 2 = 0.13 9

Lebeau, 1999 [40]

March 1993

March 1998

CAV-E or CAV-T × 6 cycles + thoracic RT (in LD) CT not specified ± thoracic RT ± PCI

Recombinant IFN-␣2a vs Control Recombinant IFN-␣2b vs Control IFN-␣ vs Control

LD, ED in OR

84 68

nr

2-yS 22% – 2-yS 13%

Jett, 1994 [41]

1987

October 1990

LD, ED in CR

51 49

6.9 p = 0.54 8.1

13.3 p = 0.43 18.8

Van Zandwijk, 1997 [42]

April 1989

September 1993

LD, ED in CR

64 62

3.7 p = 0.652 4.2

8.9 p = 0.89 9.9

PE × 2 cycles → CAV × 4 cycles ± thoracic RT ± PCI Mainly CDE ± thoracic RT

Recombinant IFN-␥ vs Control Recombinat IFN-␥ vs Control

pts = patients; TTP = time-to-progression; OS = overall survival; ED = extensive disease; LD = limited disease; CR = complete response, OR = objective response; nr = not reported; RT = radiotherapy; PCI = prophilactic cranial irradiation; ETO = etoposide; CTX = cyclophosphamide; VCR = vincristine; CAV = vincristine, doxorubicin, cyclophosphamide; E = etoposide; T = teniposide; CDE = cyclophosphamide, doxorubicin, etoposide; PE = cisplatin, etoposide; CT = chemotherapy. a Progression-free survival.

effects or toxic death or patient’s refusal to proceed (Table 5). The data were not available because not specified in the text for 7 trials [27,29,31,33,40,45,46]. The higher percentage of patients stopping maintenance therapy due to toxicity was reported with interferons and the other biological agents. Patients seemed to be more compliant to receive maintenance chemotherapy. 4. Discussion SCLC is a poor prognosis disease. Its outcome has not improved significantly during the last decades despite several attempts. Among these, maintenance therapy has been tested in several clinical trials with the aim to overcome the onset of cell resistance. However, the performed RCTs reported contrasting results ren-

dering the interpretation of its role controversial. For this reason, we pooled all these data to understand whether the maintenance therapy might be a valid approach for improving SCLC outcome. Our findings indicate that measures of efficacy do not consistently show an overall effect for maintenance therapy, even though for some classes of drugs a statistically significant effect is demonstrated. Based on our findings a statistically significant reduction of mortality was detected in the studies assessing the efficacy of interferon-alpha and chemotherapy (although in this case with high heterogeneity). According to the HRs estimates, the improvement in OS can be projected to 3.5 weeks for interferon-alpha and 2 weeks for chemotherapy in patients with SCLC responding to induction chemotherapy. These figures correspond to an absolute improvement in survival of 4% (from 30 to 34%) at 1-year for

Table 3 Description of trial of maintenance therapy with biologic agents in small-cell lung cancer. Author

Start of the accrual

End of the accrual

Induction therapy

Maintenance therapy

Study population

No. pts

PFS (months)

OS (months)

Shepherd, 2002 [43]

January 1997

April 2002

Marimastat vs Placebo

LD, ED in OR

266 266

4.3 p = 0.81 4.4

9.3 p = 0.90 9.7

Giaccone, 2005 [44]

March 1998

October 2002

Bec2/BCG vaccine vs Control

LD in CR

257 258

5.7 p = 0.299 6.3

14.3 p = 0.283 16.34

Pujol, 2007 [45]

October 2000

January 2004

CT on investigator discretion ≥ 4 cycles CT on investigator discretion × 4–6 cycles ± thoracic RT ± PCI PCDE × 2 cycles

ED in OR

49 43

6.6 p = 0.15 6.4

11.7 p = 0.16 8.7

Arnold, 2007 [46]

March 2003

April 2006

PCDE × 4 cycles + Thalidomide vs PCDE × 4 cycles + Placebo Vandetanib vs Placebo

LD, ED in OR

53 54

2.7 p = 0.51 2.8

10.6 p = 0.90 11.9

CT on investigator discretion x ≥ 4 cycles ± thoracic RT

pts = patients; PFS = progression-free survival; OS = overall survival; ED = extensive disease; LD = limited disease; CR = complete response; OR = objective response; RT = radiotherapy; PCI = prophilactic cranial irradiation; PCDE = etoposide, cisplatin, cyclophosphamide, 4 -epidoxorubicin; CT = chemotherapy.

126

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

Table 4 Analyzed items according to Jadad’s scale. Author

Randomization

Allocation concealment

Blinding

Intention-to-treat

Cullen, 1986 [26] Anonymous, 1989 [27] Byrne, 1989 [28] Ettinger, 1990 [29] Giaccone, 1993 [30] Johnson, 1993 [31] Beith, 1996 [32] Sculier, 1996 [33] Schiller, 2001 [34] Hanna, 2002 [35] Han, 2008 [36] Mattson, 1992 [37] Kelly, 1995 [38] Tummarello, 1997 [39] Lebeau, 1999 [40] Jett, 1994 [41] Van Zandwijk, 1997 [42] Shepherd, 2002 [43] Giaccone, 2005 [44] Pujol, 2007 [45] Arnold, 2007 [46]

YES YES (stratification) YES YES (stratification) YES YES YES (stratification) YES (stratification) YES YES (stratification) YES YES (stratification) YES (stratification) YES (stratification) YES YES (stratification) YES (stratification/minimization) YES (permuted block) YES (stratification/minimization) YES (stratification) YES (stratification)

B: not clear B: not clear B: not clear B: not clear B: not clear B: not clear B: not clear A: adequate B: not clear B: not clear Unknown B: not clear B: not clear A: adequate Unknown A: adequate A: adequate B: not clear B: not clear B: not clear A: adequate

NO NO NO NO NO NO NO NO NO NO Unknown NO NO NO Unknown NO NO YES (double) NO YES (double) YES (double)

Not stated Not stated Not stated Not stated Not stated Not stated Not stated Not stated Not stated Not stated Unknown Not stated Not stated Not stated Unknown YES YES YES YES YES YES

chemotherapy and of 9% (from 30 to 39%) for interferon-alpha. The high value reported by the heterogeneity test could be related to the heterogeneous drugs employed as maintenance therapy. In some studies, only patients achieving a complete and/or partial response were candidates to receive maintenance chemotherapy; in others, only patients with non-progressing disease were eligible. Moreover, some trials randomized only LD-SCLC, others only ED-SCLC, and others both stages. This could impact on the final results of the present analysis which is not being performed on individual patients data (IPD), and it is not adequate for testing heterogeneity by extrapolating figures for each variable. The fact that our meta-analysis was not based on IPD represents a clear limitation to the interpretation of results, since this approach could tend

to overestimate treatment effects, even if evidence of publication bias seems to be excluded. Moreover, we recognise that the available data cannot help much in clarifying the risk/benefit ratio of the maintenance therapies, since toxicity could be analyzed only in a qualitative way and we do not have real data to support our assumption. Also, heterogeneity of the patient populations studied in the individual trials (ED, LD, responders, non-progressive patients) could not be analyzed thoroughly with data available. An IPD meta-analysis could improve precision in the estimates, particularly for PFS and for selected subgroup of patients (for example for ED and LD patients), and better define the cost and risk/benefit profile of the treatments. Nevertheless, some general results produced by our research seem relevant. Consistently with the previous

Table 5 Compliance of patients to maintenance therapy. Author

Maintenance therapy

No. of evaluable patients for toxicity

Withdrawn patients due to toxicity n◦ (%)

Chemotherapy Cullen, 1986 [26] Anonymous, 1989 [27] Byrne, 1989 [28] Ettinger, 1990 [29] Giaccone, 1993 [30] Johnson, 1993 [31] Beith, 1996 [32] Sculier, 1996 [33] Schiller, 2001 [34] Hanna, 2002 [35] Han, 2008 [36]

VAC ETO + CTX + MTX + VCR CVM CAV or CAV-HEM CDE PE VAC ETO + VDS Topotecan Oral ETO Irinotecan

45 131 34 43 219 79 65 45 112 72 21

2 (4.4) Not specified 9 (26.5) Not specified 3 (1.4) Not specified 3 (4.6) Not specified 1 (0.9) 33 (46) (due to PD or toxicity) 0 (0)

Interferons Mattson, 1992 [37] Kelly, 1995 [38] Tummarello, 1997 [39] Lebeau, 1999 [40] Jett, 1994 [41] Van Zandwijk, 1997 [42]

Natural IFN-␣ Recombinant IFN-␣2a Recombinant IFN-␣2b IFN-␣ Recombinant IFN-␥ Recombinant IFN-␥

91 64 14 84 51 59

4 (4.4) 43 (67) 14 (100) Not specified 9 (17.6) 7 (12)

Other biologic agents Shepherd, 2002 [43] Giaccone, 2005 [44] Pujol, 2007 [45] Arnold, 2007 [46]

Marimastat Bec2/BCG vaccine PCDE + thalidomide Vandetanib

266 257 49 53

85 (32) 41 (16) Not specified Not specified

ETO = etoposide; CTX = cyclophosphamide; MTX = methotrexate; VCR = vincristine; VDS = vindesine; CAV (VAC) = vincristine, doxorubicin, cyclophosphamide; CVM = methotrexate, vincristine, cyclophosphamide; HEM = hexamethylmelamine, etoposide, methotrexate; CDE = cyclophosphamide, doxorubicin, etoposide; PE = cisplatin, etoposide; PCDE = etoposide, cisplatin, cyclophosphamide, 4 -epidoxorubicin; PD = progressive disease.

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

meta-analysis [6], chemotherapy seems to produce a statistically significant benefit in terms of OS and PFS, but this effect is of little clinical importance considering that it consists of less than one month of increase in OS and in 4% increase of 1-year survival (which translates in 25 patients to be treated for one to survive more than 1 year). This difference may be due to the different selection of trials, and to the different method used for summarising treatment effect. According to various authors, our method should be preferable, since it does not rely on a particular time point, as with previous work [47]. Another important finding of our review is related to the low quality of published studies, which is consistent with the findings reported by a qualitative review [48]: several trials using chemotherapy as experimental arm were mostly performed in the early-1990s and even if they were RCTs, the assessment of methodological quality of included studies showed that none of these trials stated clearly that an ITT analysis was used, and allocation concealment was uncertain. Regarding interferons, a different effect of the two classes was reported: interferon-alpha seems to give a benefit both in OS and PFS. However, due to lack of information concerning the treatment compliance, it is not possible to rule out that a population with a better prognosis could have been selected. Therefore, considerations similar to the ones concerning chemotherapy may be drawn, and, although a statistical survival benefit was demonstrated, its clinical impact should be considered still experimental. The targeted agents used in these trials do not seem to add benefit. Although they represent a very heterogeneous group, which could suggest not to pool their results, we believe the value of displaying them is important, considering also the very high number of patients included in the RCTs. Despite the high pharmacological heterogeneity of these therapies, discordance was not high (I2 = 32.6%), and no evidence of effect was detected in any of these studies. These agents, which seemed to be the best candidates for a maintenance therapy due to their “hypothetical” low toxicity compared to chemotherapy, failed in demonstrating any effect. However, most of these biologic agents derived directly from phase I studies and, considering their presumable safety, were engaged directly in phase III randomized trials without any evidence of their selective clinical activity in SCLC. We strongly trust that specific early phase I–II trials should be always performed, in order to get better evidence on optimal dosage and activity of these drugs, and only if a positive result is reached specifically in SCLC, phase III studies might be planned. Finally, the adverse effects of the considered maintenance approaches were different due to the different agents employed, and so comparisons between trials were not possible. Consequently, we measured the compliance to treatment through the percentage of patients who withdrew therapy due to toxicity or refusal. Interestingly, the compliance seemed somehow better for chemotherapy than for the other agents, probably because the administered chemotherapeutics had been largely investigated in SCLC with a well known activity and safety profile which resulted easier to manage. In conclusion, the positive results in OS reported by maintenance chemotherapy and interferon-alpha should be considered with caution, but may supply indications for further studies in this field. Our data suggest that it is worth to continue this research on maintenance therapy which cannot be considered for routinely use. As a matter of facts, it is possible that maintenance strategy has failed due to the choice of administered drugs and not to the poor worthiness of the approach. In order to individuate the best candidate for the maintenance therapy, all the steps of drugs development are needed, starting from preclinical studies evaluating the sensitivity of SCLC to late phase III trials. Once demonstrated its efficacy, if the drug has a good and safe profile it may be considered

127

for maintenance therapy. In this view, new targeted agents may be of further help in this purpose. Conflict of interest statement None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Acknowledgment We acknowledge Joanna Landi for her valuable technical assistance. References [1] Elias AD. Small cell lung cancer: state-of-the-art therapy in 1996. Chest 1997;112(Suppl. 4):251S–8S. [2] Patel AM, Dunn WF, Trastek VF. Staging systems of lung cancer. Mayo Clin Proc 1993;68:475–82. [3] Samson DJ, Sidenfeld J, Simon GR, Turrisi III AT, Bonnell C, Ziegler KM, et al. Evidence for management of small cell lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132:314S–23S. [4] Simon GR, Turrisi A. Management of small cell lung cancer: ACCP evidencebased clinical practice guidelines (2nd edition). Chest 2007;132:324S–39S. [5] Sorensen M, Felip E. Small-cell lung cancer: ESMO clinical recommendations for diagnosis, treatment an follow-up. Ann Oncol 2009;20(Suppl. 4):iv71–2. [6] Bozcuk H, Artac M, Ozdogan M, Savas B. Does maintenance/consolidation chemotherapy have a role in the management of small cell lung cancer (SCLC)? Cancer 2005;104:2650–7. [7] Grossi F, Aita M, Follador A, Defferrari C, Brianti A, Sinaccio G, et al. Sequential, alternating and maintenance/consolidation chemotherapy in advanced nonsmall cell lung cancer: a review of the literature. Oncologist 2007;12:451–64. [8] Rinaldi M, Belvedere O, Cauchi C, Defferrari C, Viola G, Grossi F. Maintenance chemotherapy in non-small cell lung cancer. Ann Oncol 2006;17(Suppl. 2):ii67–70. [9] Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1998;17:2815–34. [10] Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58. [11] Higgins JPT, Green S. Analysing data and undertaking meta-analyses. In: Cochrane handbook for systematic reviews of interventions, version 5.0.1 [updated September 2008]. The Cochrane Collaboration; 2008. [12] Cochrane Collaboration. RevMan user guide; 2004. Available from URL: http://www.ccims.net/download/revman/Documentation/User%20guide.pdf [Accessed 21 Dec 2007]. [13] Jadad AR, Moore RA, Carol D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12. [14] Spiro SG, Souhami RL, Geddes DM, Ash CM, Quinn H, Harper PG, et al. Duration of chemotherapy in small cell lung cancer: a Cancer Research Campaign Trial. Br J Cancer 1989;59:578–83. [15] Zarogoulidis K, Ziogas E, Papagiannis A, Charitopoulos K, Dimitriadis K, Economides D, et al. Interferon alpha-2a and combined chemotherapy as first line treatment in SCLC patients: a randomised trial. Lung Cancer 1996;15:197–215. [16] Routsalainen TM, Halme M, Tamminen K, Szopinski J, Niiranen A, Pyrhonen S, et al. Concomitant chemotherapy and IFN-␣ for small cell lung cancer: a randomized multicenter phase III study. J Interferon Cytokine Res 1999;19:253–9. [17] Lee SM, Woll PJ, Rudd R, Ferry D, O’Brien M, Middleton G, et al. Anti-angiogenic therapy using thalidomide combined with chemotherapy in small cell lung cancer: a randomized, double-blind, placebo-controlled trial. J Natl Cancer Inst 2009;101:1049–57. [18] Einhorn LH, Crawford J, Birch R, Omura G, Johnson DH, Greco FA. Cisplatin plus etoposide consolidation following cyclophosphamide, doxorubicin and vincristine in limited small-cell lung cancer. J Clin Oncol 1984;6:451–6. [19] Tummarello D, Graziano F, Mari D, Cetto G, Pasini F, Santo A, et al. Small cell lung cancer (SCLC): a randomized trial of cyclophosphamide, adryamicin, vincristine plus etoposide (CAV-E) or teniposide (CAV-T) as induction treatment, follone in complete responders by alfa-interferon or no treatment, as maintenance therapy. Anticancer Res 1994;14:2221–8. [20] Mattson K, Niiranen A, Ruotsalainen T, Maasilta P, Halme M, Pyrhonen S, et al. Interferon maintenance therapy for small cell lung cancer: improvement in long-term survival. J Interferon Cytokine Res 1997;17:103–7. [21] Maurer LH, Tulloh M, Weiss RB, Blom J, Leone L, Glidewell O, et al. A randomized combined modality trial in small cell carcinoma of the lung. Comparison of combination chemotherapy-radiation therapy versus cyclophosphamide-radiation therapy effects of maintenance chemotherapy and prophylactic whole brain irradiation. Cancer 1980;45:30–9. [22] Rigas JR, Denham CA, Rinaldi D, Moore T, Smith JW, Winston II RD, et al. Adjuvant target therapy in unresectable lung cancer: the results of two randomized placebo-controlled trials of BAY 12-9566, a matrix metalloproteinase inhibitor (MMPI). Lung Cancer 2003;41(Suppl. 2):S34 [Abstr O-107].

128

A. Rossi et al. / Lung Cancer 70 (2010) 119–128

[23] Broder LE, Sridhar KS, Selawry OS, Charyulu KN, Rao RK, Saldana MJ, et al. A randomized clinical trial in bronchogenic small-cell carcinoma evaluating alternating maintenance therapy of vincristine, adryamicin, procarbazine, and etoposide (VAPE) with cyclophosphamide, CCNU, and methotrexate (CCM) versus CCM maintenance alone in complete responders following VAPE induction and late intensification. Am J Clin Oncol 1994;17:527–37. [24] Pandya KJ, Dahlberg S, Hidalgo M, Cohen RB, Lee MW, Schiller JH, et al. A randomized, phase II trial of two dose levels of temsirolimus (CCI-779) in patients with extensive-stage small-cell lung cancer who have responding or stable disease after induction chemotherapy: a trial of the Eastern Cooperative Oncology Group (E1500). A preliminary report. J Thorac Oncol 2007;2:1036–41. [25] Pujol JL, Gibney DJ, Su JQ, Maksymiuk AW, Jett JR. Immune response induced in small-cell lung cancer by maintenance therapy with interferon-␥. J Natl Cancer Inst 1993;85:1844–50. [26] Cullen M, Morgan D, Gregory W, Robinson M, Cox D, McGivern D, et al. Maintenance chemotherapy for anaplastic small cell carcinoma of the bronchus: a randomized, controlled trial. Cancer Chemother Pharmacol 1986;17:157–60. [27] Anonymous. Controlled trial of twelve versus six courses of chemotherapy in the treatment of small-cell lung cancer. Br J Cancer 1989;59:584–90. [28] Byrne MJ, van Hazel G, Trotter J, Cameron F, Shepherd J, Cassidy B, et al. Maintenance chemotherapy in limited small cell lung cancer: a randomised controlled clinical trial. Br J Cancer 1989;60:413–8. [29] Ettinger DS, Finkelstein DM, Abeloff MD, Ruckdeschel JC, Aisner SC, Eggleston JC, A phase III study of the Eastern Cooperative Oncology Group. A randomized comparison of standard chemotherapy versus alternating chemotherapy and maintenance versus no maintenance therapy for extensive stage small-cell lung cancer. J Clin Oncol 1990;8:230–40. [30] Giaccone G, Dalesio O, McVie GJ, Kirkpatrick A, Postmus PE, Burghouts JThM, et al. Maintenance chemotherapy in small-cell lung cancer: long-term results of a randomized trials. J Clin Oncol 1993;11:1230–40. [31] Johnson DH, Bass D, Einhorn LH, Crawford J, Perez CA, Bartolucci A, et al. Combination chemotherapy with or without thoracic radiotherapy in limited-stage small-cell lung cancer. J Clin Oncol 1993;11:1223–9. [32] Beith JM, Clarke SJ, Woods RL, Bell DR, Levi JA. Long-term follow-up of a randomised trial of combined chemoradiotherapy induction treatment, with and without maintenance chemotherapy in patients with small cell carcinoma of the lung. Eur J Cancer 1996;32A:438–43. [33] Sculier JP, Paesmans M, Bureau G, Giner V, Lecomte J, Michel J, et al. Randomized trial comparing induction chemotherapy versus induction chemotherapy followed by maintenance chemotherapy in small-cell lung cancer. J Clin Oncol 1996;14:2337–44. [34] Schiller JH, Adak S, Cella D, DeVore III FR, Johnson DH. Topotecan versus observation after cisplatin plus etoposide in extensive-stage small-cell lung cancer: E7593—a phase III trial of the Eastern Cooperative Oncology Group. J Clin Oncol 2001;19:2114–22. [35] Hanna NH, Sandler AB, Loehrer Sr PJ, Ansari R, Jung SH, Lane K, et al. Maintenance daily oral etoposide versus no further therapy following induction chemotherapy with etoposide plus ifosfamide plus cisplatin in extensive smallcell lung cancer: a Hoosier Oncology Group randomized study. Ann Oncol 2002;13:95–102. [36] Han JY, Kim HT, Lim KY, Yoon SJ, Lee DH, Lee JS. Randomized phase II study of maintenance irinotecan therapy versus observation following induction

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

chemotherapy with irinotecan and cisplatin in extensive disease small cell lung cancer. J Thorac Oncol 2008;3:1039–45. Mattson K, Niiranen A, Pyrhonen S, Holsti LR, Holsti P, Kumpulainen E, et al. Natural interferon alfa as maintenance therapy for small cell lung cancer. Eur J Cancer 1992;28A:1387–91. Kelly K, Crowley JJ, Bunn PA, Hazuka MB, Beasley K, Upchurch C, et al. Role of recombinant interferon alfa-2a maintenance in patients with limited-stage small-cell lung cancer responding to concurrent chemoradiation: a Southwest Oncology Group Study. J Clin Oncol 1995;13:2924–30. Tummarello D, Mari D, Graziano F, Isidori P, Cetto G, Pasini F, et al. A randomized, controlled phase III study of cyclophosphamide, doxorubicin, and vincristine with etoposide (CAV-E) or teniposide (CAV-T), followed by recombinant interferon-␣ maintenance therapy or observation, in small cell lung carcinoma patients with complete responses. Cancer 1997;80:2222–9. Lebeau B, De La Salmoniere P, Ozenne G, Blanchon F, Leclerc P, Coertmeur D, et al. ␣Interferon (␣IFN) as maintenance therapy for small cell lung cancer (SCLC). Proc Am Soc Clin Oncol 1999;18:475a [Abstr. 1832]. Jett JR, Maksymiuk AW, Su JQ, Mailliard JA, Krook JE, Tschetter LK, et al. Phase III trial of recombinant interferon gamma in complete responders with small cell lung cancer. J Clin Oncol 1994;12:2321–6. van Zandwijk N, Groen HJM, Postmus PE, Burghouts JTM, ten Velde GPM, Ardizzoni A, et al., A randomised phase III study of the EORTC Lung Cancer Cooperative Group. Role of recombinant interferon-gamma maintenance in responding patients with small cell lung cancer. Eur J Cancer 1997;33:1759– 66. Shepherd FA, Giaccone G, Seymour L, Debruyne C, Bezjak A, Hirsh V, et al. Prospective, randomized, double-blind, placebo-controlled trial of marimastat after response to first-line chemotherapy in patients with small-cell lung cancer: a trial of the National Cancer Institute of Canada-Clinical Trials Group and the European Organization for Research and Treatment of Cancer. J Clin Oncol 2002;20:4434–9. Giaccone G, Debruyne C, Felip E, Chapman PB, Grant SC, Millward M, et al. Phase III study of adjuvant vaccination with Bec2/Bacille Calmette-Guerin in responding patients with limited-disease small cell lung cancer (European Organization for Research and Treatment of Cancer 08971-08971B; Silva Study). J Clin Oncol 2005;23:6854–64. Pujol JL, Breton JL, Gervais R, Tanguy ML, Quoix E, David P, et al. Phase III double-blind, placebo-controlled study of thalidomide in extensive-disease small-cell lung cancer after response to chemotherapy: an Intergroup Study FNCLCC cleo04–IFCT 00-01. J Clin Oncol 2007;25:3945–51. Arnold AM, Seymour L, Smylie M, Ding K, Ung Y, Findlay B, et al. Phase II study of vandetanib or placebo in small-cell lung cancer patients after complete or partial response to induction chemotherapy with or without radiation therapy: National Cancer Institute of Canada Clinical Trials Group Study BR.20. J Clin Oncol 2007;25:4278–84. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MD. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 2007;8:16. Sculier JP, Berghmans T, Castaigne C, Luce S, Sotiriou C, Vermylen P, et al. Maintenance chemotherapy for small cell lung cancer: a critical review of the literature. Lung Cancer 1998;19:141–51.