Topotecan for Relapsed Small-cell Lung Cancer: Systematic Review and Meta-Analysis of 1347 Patients

Topotecan is the most reliable chemotherapy regimen for relapsed small-cell lung carcinoma (SCLC). The efficacy and adverse effects of topotecan as reported by previous studies varied greatly. The inclusion criterion was a prospective study that was able to provide data for 6-month over-all survival (OS) rate, 1-year OS rate, objective responses, and/or adverse effects of single agent topotecan as a second line chemotherapy for SCLC, written in English language as a full article. Any topotecan regimen were allowed. Binary data were meta-analyzed with the random-model generic inverse variance method. We included 14 articles consisted of 1347 patients. Pooled values were estimated as follows.  Six-month OS rate: 37% (95% CI: 28–46%). One-year OS rate: 9% (95% CI: 5–13%). Response rate: 5% (95% CI: 1–8%).  Six-month OS rate: 57% (95% CI: 50–64%). One-year OS rate: 27% (95% CI: 22–32%). Response rate: 17% (95% CI: 11–23%).  Grade III/IV neutropenia 69% (95% CI: 58–80%). Grade III/IV thrombopenia 41% (95% CI: 34–48%). Grade III/IV anemia 24% (95% CI: 17–30%). Non-hematorogical events were rare. Chemotherapy-related death 2% (95% CI: 1–3%). In conclusion, Topotecan provided a possibly promising outcome for sensitive-relapse SCLC and poor outcome for refractory relapse SCLC. Adverse events were mainly hematological.

Scientific RepoRts | 5:15437 | DOi: 10.1038/srep15437 provide data about survival, objective response, and AEs of TOP when prescribed as the second-line chemotherapy for patients with SCLC.

Methods
Institutional review board approval and patient consent were not required because of the review nature of this study. Study search. Systematic searching was conducted to find eligible articles. The inclusion criterion for a study to be included in the current meta-analysis was a prospective study that was able to provide data for the 6-month over-all survival (OS) rate, the 1-year OS rate, objective responses, and/or AEs of single agent TOP as second line chemotherapy for SCLC, written in English language as a full article. Any TOP regimen prescribed for both intravenous and oral administration was allowed. Conference abstracts and duplicate use of the same data were excluded. Two investigators independently searched for eligible articles using the PubMed, Web of Science, and Cochrane databases as of February, 2015. The following search formula was used for PubMed: ("small-cell lung cancer" OR "small-cell lung carcinoma" OR "SCLC") AND (relapsed OR second-line OR 2nd-line OR "second line" OR "previously treated") AND (nogitecan OR hycamtin OR topotecan OR NGT).
Outcome. Survival was evaluated as 6-month OS rate and 1-year OS rate. If 6-month and/or 1-year survival rate was not directly provided in the article, it was estimated from the survival curve using Parmar's method 21 .
For response analysis, response rate (RR), disease control ratio (DCR), complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD), not assessable (NA) were evaluated. Both SD and no change were merged as SD. NA, non-evaluable, and early death before response assessment were counted as NA. RR included CR and PR. DCR included CR, PR, and SD. For response assessment, the total number of patients evaluated was equal to the numbers of patients with CR, PR, SD, PD, and NA.
Grade III and IV hematological toxicity including neutropenia, thrombopenia, and anemia; and grade III and IV non-hematological toxicity including fatigue, asthenia, nausea/vomiting, diarrhea, anorexia, dyspnea, and fever were assessed for AE assessment. Febrile neutropenia of any grade was counted. Death that was clearly documented as due to AE was counted. Other minor non-hematological AEs were sometimes mentioned in articles; however they were not assessed for the current systematic review. This was because very rare AEs were usually not mentioned in most articles, thus, including these very rare AEs might have resulted in publication bias. AEs were analyzed based on the number of patients, not on the number of chemotherapy courses.
Survival analysis and response analysis were conducted for patients with sensitive relapse and refractory relapse separately 6 . AE analysis was conducted for both relapses collectively.
For a subgroup analysis, we divided TOP regimens into three groups: intravenous days 1-5 administration, intravenous weekly administration, and oral administration. Subgroup analyses were not performed for outcomes for refractory relapse because these analyses did not include sufficient numbers of original cohorts.
All analysis was performed in Review Manager ver 5.3 (Cochrane Collaboration, Oxford, UK). Figures illustrated using Review Manager were adjusted as necessary.
The number of patients in a study who were treated with TOP ranged from 17 to 309. The total number of patients in all the studies was 1347. In most of the studies, patients with performance status of 1 and men were the majority. Mean or median age presented for each study ranged from 58 to 68 years. Two studies did not mention types of relapse. The cutoff between refractory and sensitive relapse was 90 days except for one study that used 60 days. Intravenous TOP 1.5 mg/m 2 on days 1-5 every 3 weeks was the most preferred regimen. Two studies from Japan used intravenous TOP in a low dose of 1.0 mg/m 2 on day 1-5 every 3 weeks. Three studies used weekly intravenous TOP in higher doses of 4.0 or 6.0 mg/m 2 . Three studies used oral TOP 2.3 mg/m 2 on days 1-5 every 3 weeks (Table 1).

Discussion
TOP for patients with sensitive relapse SCLC could provide RR for 17% of cases, and disease control for 42% of cases, which resulted in 57% of 6-month OS rate and 27% of 1-year OS rate (Fig. 3). However, this regimen had poorer outcomes for patients with refractory relapse SCLC (Fig. 2). Hematologic   AEs, especially grade III/IV neutropenia were more commonly observed than non-hematological AEs. Approximately 2% of patients died from chemotherapy (Fig. 4).
In addition to the most commonly used regimen of intravenous 1.5 mg/m 2 on day 1-5 every three weeks, high dose weekly intravenous TOP regimens and oral TOP 2.3 mg/m 2 on day 1-5 every three weeks have been often selected. RCTs by Eckardt et al. in 2007 andby Pawel et al. in 2001 suggested that oral and intravenous TOP on day1-5 every three weeks have similar efficacy and AE profiles 9,16 . Thus, both regimens could be appropriate choices of treatment for sensitive relapse SCLC (Fig. 3). Oral TOP could provide a convenient administration route with a similar AE profile to intravenous TOP (Fig. 4). Weekly intravenous TOP has been found to be a treatment with good tolerability 7,18,19 . In our analysis, frequency of hematological AEs by weekly regimen were less than that by other regimens (Fig. 4). Nonetheless, RR of 2% and DCR of 24% for sensitive relapse by intravenous weekly administration may be disappointing (Fig. 3) 18,19 . Considering 6-month and 1-year OS rates by weekly regimen were not inferior to those by intravenous day 1-5 regimen and oral regimens, weekly regimen could be a choice for patients who have high risk for myelosuppression.
There is still little evidence to guide the third-line chemotherapy. Nonetheless, to improve OS time, a physician provide the third-line chemotherapy for a part of patients whose disease relapsed after the second-line chemotherapy 26,27 . Here, a clinician faces a challenge how to select a SCLC case who will gain benefit from further treatment. Retrospective chart reviews suggested that patients with normal or low lactase dehydrogenase level, good response to second-line chemotherapy, high body mass index, higher level of hemoglobin, long time to progression after the second-line chemotherapy indicated better response to the third-line chemotherapy 26,27 .
In 2014, two reports that were potentially conflicting with each other were published. Lara et al. analyzed 329 patients with extensive-stage SCLC who progressed after platinum-based chemotherapy 28 . This study indicated that platinum-sensitivity status may no longer be strongly associated with progression-free survival and OS. On the other hand, Ardizzoni et al. reported a retrospective study with 631 relapsed SCLC cases 6 . This concluded that the separation of relapsed SCLC into two types of relapses based on a TFI cut-off of 60 days was valid and could be a standard of care. The current analysis, though mostly using a TFI cut-off of 90 days, also demonstrated that all of 6-month OS rate, 1-year OS rate, RR and DCR were more favorable for patients with sensitive relapse than those with refractory relapse.
Few limitations of the current study should be mentioned. First, the meta-analysis was performed as aggregated data meta-analysis. If we could have obtained individual data from all the original studies, individual patient data meta-analysis would have been preferred. Second, regimen comparison was not done by head-to-head manner. Third, strong heterogeneities for some outcomes made it difficult to interpret results. However, we believe the current meta-analysis provided useful information for clinicians.
In conclusion, TOP provided a possibly promising outcome for patients with sensitive-relapse SCLC and poor outcome for patients with a refractory relapse SCLC. Adverse events were mainly hematological. We believe these data will be informative for physicians who take care of patients with relapsed SCLC.