3.1 Characteristics of the included studies
During the initial search, we obtained 9567 publications. As a result of eliminating duplicates and screening titles and abstracts, 26 studies, including 16 Clinical trials (N = 5696 patients), finally
met our criteria (Fig. 1). A total of 8 pharmacological interventions were studied in these clinical trials, representing the following classes of therapies: bevacizumab、trebananib, cediranib, niraparib, nintedanib, olaparib, rucaparib, and fuzuloparib. A detailed description of the included studies can be found in Table 2. Trial duration ranged from 1 to 5 years.
Table 2
Baseline characteristics of the studies included in this NMA.
Auther
|
Clinical Trial Number
|
Year
|
Sample Size (E/C)
|
Study phase
|
Treatment
|
Median age(E/C)
|
Outcomes
|
Sandro Pignata et al. [32]
|
NCT01802749
|
2021
|
203/203
|
Phase II
|
Bevacizumab
|
61.0/61.0
|
PFS; OS; Toxicity
|
Carol Aghajanian et al. [27–28]
|
NCT00434642
|
2015
|
242/242
|
Phase III
|
Bevacizumab
|
60.0/61.0
|
PFS; OS; Toxicity
|
Robert L.Coleman et al. [29]
|
NCT00565851
|
2017
|
337/337
|
Phase III
|
Bevacizumab
|
59.5/60.6
|
PFS; OS; Toxicity
|
Beth Y. Karlan et al. [33]
|
NCT00479817
|
2011
|
53/55
|
Phase II
|
Trebananib
|
59.0/62.0
|
PFS; OS; Toxicity
|
Christian Marth et al. [34]
|
NCT01281254
|
2016
|
114/109
|
Phase III
|
Trebananib
|
61.0/60.0
|
PFS; OS; Toxicity
|
Bradley J. Monk et al.[35–36]
|
NCT01204749
|
2016
|
458/461
|
Phase III
|
Trebananib
|
60.0/60.0
|
PFS; OS; Toxicity
|
JonathanA.Ledermann et al. [37–38]
|
NCT00532194
|
2021
|
164/118
|
Phase III
|
Cediranib
|
62.0/62.0
|
PFS; OS; Toxicity
|
JonathanA.Ledermann et al.[39]
|
NCT00710762
|
2011
|
43/40
|
Phase II
|
Nintedanib
|
60.0/63.0
|
PFS; OS; Toxicity
|
X. H. Wu et al. [40]
|
NCT03705156
|
2021
|
177/88
|
Phase III
|
Niraparib
|
53/55
|
PFS; OS; Toxicity
|
Mansoor R. Mirza et al. [24–26]
|
NCT01847274
|
2020
|
372/181
|
Phase III
|
Niraparib
|
57/58
|
PFS; OS; Toxicity
|
Andrés Poveda et al. [15–16]
|
NCT01874353
|
2021
|
196/99
|
Phase III
|
Olaparib
|
56/56
|
PFS; OS; Toxicity
|
JonathanA.Ledermann et al. [17–19]
|
NCT00753545
|
2016
|
136/129
|
Phase II
|
Olaparib
|
58/59
|
PFS; OS; Toxicity
|
Amit M Oza et al.[14]
|
NCT01081951
|
2014
|
81/81
|
Phase II
|
Olaparib
|
59/62
|
PFS; OS; Toxicity
|
JonathanA.Ledermann et al. [21–23]
|
NCT01968213
|
2021
|
375/189
|
Phase III
|
Rucaparib
|
61/62
|
PFS; OS; Toxicity
|
Rebecca Kristeleit et, al. [41]
|
NCT02855944
|
2022
|
107/54
|
Phase III
|
Rucaparib
|
58/58
|
PFS; OS; Toxicity
|
Ning Li, et al. [42]
|
NCT03863860
|
2022
|
167/85
|
Phase III
|
fuzuloparib
|
54/54
|
PFS; OS; Toxicity
|
3.2 PFS
3.2.1 All comer population
The PFS network relationship in the overall population is shown in Supplementary Fig. 3A; 16 studies reported outcomes related to the HRs of PFS. We compared the 9 interventions included directly.
The results showed that compared with the placebo, There were seven treatments that improved PFS, and the HR ranged from 1.35 to 3.99.
The interventions with signifificant differences from placebo are bevacizumab (HR = 1.83, 95%CI: 1.47 to 2.33), trebananib(HR = 1.35, 95%CI: 1.03 to 1.71), cediranib (HR = 1.79, 95%CI: 1.78 to 2.72), niraparib (HR = 2.9, 95%CI: 2.15 to 3.95), olaparib (HR = 2.74, 95%CI: 2.05 to 3.6) ,rucaparib (HR = 2.61, 95%CI: 1.81 to 3.57) and fuzuloparib (HR = 3.99, 95%CI: 2.41 to 6.6).
Compared with fuzuloparib of the top SUCRA-ranked intervention,niraparib, Olaparib, and rucaparib were comparable in their efficacy, Bevacizumab, trebananib, and cediranib were significantly inferior.Detailed results are shown in Supplementary Table 3A and Fig. 4 、5、6A.
3.2.2 BRCA mutated patients
The PFS network relationship in the BRCA mutated patients is shown in Supplementary Fig. 3B; 7 studies reported outcomes related to the HRs of PFS. We compared the 5 interventions included directly.
The results showed that compared with the placebo, There were four treatments that improved PFS, and the HR ranged from 4.08 to 7.16.
The interventions with signifificant differences from placebo are niraparib (HR = 4.08, 95%CI: 1.49 to 11.33), olaparib (HR = 4.29, 95%CI: 1.9 to 9.75) ,rucaparib (HR = 4.16, 95%CI: 1.04 to 16.49) and fuzuloparib (HR = 7.16, 95%CI: 1.63 to 30.82).
Compared with fuzuloparib of the top SUCRA-ranked intervention,niraparib, Olaparib, and rucaparib were comparable in their efficacy. Detailed results are shown in Supplementary Table 3B and Fig. 4、5、6B.
3.2.3 BRCA wild-type patients
The PFS network relationship in the BRCA wild-type patients is shown in Supplementary Fig. 3C; 5 studies reported outcomes related to the HRs of PFS. We compared the 5 interventions included directly.
The results showed that the significant difference compared with placebo is niraparib (HR = 2.33, 95%CI: 1.17 to 4.7).In order of SUCRA ranking, niraparib has the highest ranking. Detailed results are shown in Supplementary Table 3C and Fig. 4 、5、6C.
3.3 OS
3.3.1 All comer population
The OS network relationship in the overall population is shown in Supplementary Fig. 3D; 15 studies reported outcomes related to the HRs of OS. We compared the 9 interventions included directly. The results showed that compared with the placebo, disappointingly, there was no statistically significant beneficial effect for OS between the maintenance treatment groups; the HR ranged from 0.89 to 1.45. Among them, the highest SUCRA ranking is niraparib. Detailed results are shown in Supplementary Table 3D and Fig. 4 、5、6D.
3.3.2 BRCA mutated patients
The OS network relationship in the BRCA mutated patients is shown in Supplementary Fig. 3E; 6 studies reported outcomes related to the HRs of OS. We compared the 4 interventions included directly.
The results showed that in BRCA-mutated patients receiving maintenance treatment, no nodes of statistical significance were reached, and the HR ranged from 0.83 to 1.42.Among them, the highest SUCRA ranking is Olaparib. Detailed results are shown in Supplementary Table 3E and Figs. 4 、5、6E.
3.3.3 BRCA wild-type patients
The OS network relationship in the BRCA wild-type patients is shown in Supplementary Fig. 3F; 4 studies reported outcomes related to the HRs of OS. We compared the 4 interventions included directly.
The results showed that BRCA wild-type patients receiving maintenance treatment also had no nodes of statistical significance reached, and the HR ranged from 1.01 to 1.2. Among them, the highest SUCRA ranking is Olaparib.Detailed results are shown in Supplementary Table 3F and Figs. 4 、5、6F.
3.4 PFS of AC at each time point
For PFS, 16 articles reported related outcomes; the network relationship is shown in Supplementary Fig. 7A. At the 8-time points of 3, 6, 9,12, 18, 24,30, and 36 months, there were sufficient data for the NMA of PFS. We summarizes the interventions' details with significant results compared with placebo in Table 4. Detailed results of comparisons of 8 interventions at each time point are shown in Supplementary Fig. 8–10.
In 3rd month, compared with placebo, Bevacizumab (HR = 3.24,95% Cl:1.74 to 6.37), Cediranib (HR = 7.62,95% Cl:1.53 to 64.15), Trebananib (HR = 1.72,95% Cl:1.0 to 3.03) ,Niraparib (HR = 2.9,95% Cl:1.57 to 5.92),Olaparib (HR = 4.35,95% Cl:2.24 to 8.77),Rucaparib (HR = 2.11,95% Cl:1.04 to 3.74) and Fuzuloparib (HR = 1.72,95% Cl:1.0 to 3.03) significantly increased PFS rates.
In comparison with placebo at 6th month,Bevacizumab (HR = 3.14, 95%CI: 1.93 to 5.2), Trebananib (HR = 1.79, 95%CI: 1.06 to 2.89) ,Niraparib (HR = 3.46, 95%CI: 2.02 to 6.31),Olaparib (HR = 5.63, 95%CI: 3.28 to 10.6),Rucaparib(HR = 2.91, 95%CI: 1.51 to 4.96) and Fuzuloparib (HR = 1.79, 95%CI: 1.06 to 2.89) significantly increased PFS rates.
At 9th month, the PFS for Bevacizumab (HR = 2.61, 95% Cl: 1.46 to 4.75), Niraparib (HR = 4.11, 95% Cl: 1.98 to 8.95),Olaparib (HR = 5.42, 95% Cl: 2.75 to 10.34),Rucaparib(HR = 3.62, 95% Cl: 1.57 to 7.48) and Fuzuloparib (HR = 5.13, 95%CI: 1.7 to 15.37)were significantly higher compared to that of placebo.
At 12th month, the PFS for Bevacizumab(HR = 3.08, 95%CI:1.61 to 6), Niraparib (HR = 4.66, 95%CI:2.04 to 11.06),Olaparib (HR = 7.49, 95%CI:3.72 to 15.69),Rucaparib (HR = 5.86, 95%CI:2.22 to 13.87) and Fuzuloparib (HR = 6.19, 95%CI: 1.81 to 21.92)were significantly higher compared to that of placebo.
At 18th month, the PFS was significantly higher for Bevacizumab(HR = 5.1, 95%CI: 1.34 to 19.92), Niraparib (HR = 5.67, 95%CI: 1.11 to 30.57) ,Olaparib(HR = 15.94, 95%CI: 4.15 to 77.07) and Rucaparib(HR = 7.17, 95%CI: 1.22 to 40.83) compared to placebo.
At 24th month, the PFS increased significantly when Bevacizumab (HR = 2.78, 95%CI: 1.38 to 6.26), Niraparib (HR = 3.04, 95%CI: 1.26 to 7.86) ,Olaparib(HR = 4.23, 95%CI: 1.1 to 15.48) and Rucaparib (HR = 11.04, 95%CI: 3.02 to 36.71) were compared with placebo.
In the 30th month, the PFS was significantly higher for Rucaparib (HR = 9.08, 95%CI: 1.01 to 84.13) compared to the placebo.
At 36th month, compared with the placebo, any interventions were not significantly increased
PFS rates. Therefore, in the 36th month, no interventions had significant results.
In PFS, compared with placebo, Rucaparib was significant from 6 to 30 months; Bevacizumab, Niraparib, and Olaparib significant from 3 to 24 months; Fuzuloparib significant from 3 to 12 months; trebananib was significant from 3 to 6 months;cediranib only significant at 3rd month. However, at each month point, compared with placebo, nintedanib wasn’t significantly increased PFS rates.
3.5 OS of AC at each time point
For OS, 13 articles reported related outcomes; the network relationship is shown in Supplementary Fig. 7B; there were sufficient data for the NMA of OS.We summarizes the interventions' details with significant results compared with placebo in Table 5. Detailed results of comparisons of 8 interventions at each time point are shown in Supplementary Fig. 11–13.
At 3rd month, compared with the placebo, Olaparib (HR = 4.43,95% Cl:1.21 to 17.04) significantly increased OS rates.
Compared with the placebo in the 6th month, Olaparib (HR = 6.37, 95%CI: 1.95 to 36.42)significantly increased OS rates.
In the 9th month, the OS rate for Olaparib (HR = 5.77, 95% Cl: 1.33 to 33.18)was significantly higher than that of the placebo.
In the 12th month, the OS rate for Olaparib (HR = 7.58, 95%CI:3.04 to 19.6) was significantly higher compared to that of the placebo.
In the 18th month, the OS rate was significantly higher for Olaparib(HR = 24.39, 95%CI: 3.88 to 413.19) compared to the placebo.
In the 24th month, the OS rate increased significantly when Olaparib(HR = 26.55, 95%CI: 3.58 to 2080.51) was compared with a placebo.
In the 30th month, the OS rate was significantly higher for Olaparib (HR = 10.81, 95%CI: 1.55 to 485.85) compared to the placebo.
In the 36th month, the OS rate was significantly increased by Olaparib (HR = 25.04, 95%CI: 4.72 to 415.94) compared with the placebo.
In OS, compared with placebo, the intervention measure with significant effect from 3 to 36 months was Olaparib; other interventions were not significantly increased OS rates.
3.6 PFS of BRCAm at each time point
For PFS of BRCAm, 7 articles reported related outcomes; the network relationship is shown in Fig. 7C, and there were sufficient data for the NMA. We summarize the details of the interventions with significant results compared with placebo in Table 6. Detailed results of comparisons of 8 interventions at each time point are shown in Supplementary Fig. 14–16.
At 3rd month, PFS rate of BRCAm was significantly increased by Olaparib (HR = 6.49, 95% Cl: 1.5 to 29.2),Rucaparib(HR = 10.11, 95% Cl: 1.21 to 90.29) and Fuzuloparib (HR = 28.89, 95%CI: 1.95 to 1265.75) in comparison with placebo.
At 6th month, in comparison with placebo, Niraparib (HR = 4.17, 95%CI: 1.07 to 18.96),Olaparib (HR = 7.52, 95% Cl: 2.11 to 33.02),Rucaparib(HR = 7.7, 95% Cl: 1.05 to 58.43) and Fuzuloparib (HR = 12.35, 95%CI: 1.52 to 105.68) were more effective.
In 9th month, Niraparib (HR = 9.3, 95%CI: 1.53 to 74.58), Olaparib (HR = 7.55, 95% Cl: 1.4 to 35.12), and Fuzuloparib (HR = 17.1, 95%CI: 1.09 to 287.23) significantly increased PFS rate of BRCAm compared to placebo.
At 12th month, there was a significant increase in the PFS rate of BRCAm with Niraparib (HR = 5.94, 95%CI: 1.2 to 39.58), Olaparib (HR = 8.09, 95% Cl: 2.19 to 38.03) and Fuzuloparib (HR = 35.69, 95%CI: 2.46 to 674.12) compared with placebo.
In the 18th month, the PFS rate of BRCAm was significantly higher with Olaparib (HR = 32.09, 95% Cl: 1.49 to 2116.8) than with placebo.
Among the treatments tested at 24 to 36 months, compared with placebo, any interventions were not significantly increased the PFS rate of BRCAm.
In summary,for the PFS of BRCAm, Olaparib is significant from 3 to 18 months; according to duration, Olaparib may be the first choice for improving the PFS of BRCAm.
3.7 PFS of BRCAwt at each time point
For PFS of BRCAwt, 5 articles reported related outcomes; the network relationship is shown in Fig. 7D, and there were sufficient data for the NMA.We summarizes the interventions' details with significant results compared with placebo in Table 7. Detailed results of comparisons of 8 interventions at each time point are shown in Supplementary Fig. 17–19.
In 6th month, in comparison with placebo, Niraparib (HR = 2.91, 95%CI: 1.03 to 8.47) was more effective.
In the 9th month, Niraparib (HR = 3.82, 95%CI: 1.07 to 14.05) and Rucaparib (HR = 5.95, 95%CI: 1.0 to 36.02) significantly increased the PFS rate of BRCAwt compared to placebo.
In 12th month, there was a significant increase in the PFS rate of BRCAwt with Niraparib (HR = 4.17, 95%CI: 1.01 to 17.2)and Rucaparib (HR = 7.11, 95%CI: 1.0 to 51.26) compared with placebo.
Among the treatments tested at 3 and 18 to 36 months, compared with placebo, any interventions were not significantly increased the PFS rate of BRCAwt.
In summary, for the PFS of BRCAwt, Niraparib is significant from 6 to 12 months; according to duration, niraparib may be the best choice for improving PFS of BRCAwt.
Supplementary Fig. 3A,16 studies reported outcomes related to the HRs of PFS. We compared the 9 interventions included directly.
The results showed that compared with placebo, There were seven treatments improved PFS,the HR ranged from 1.35 to 3.99.
The interventions with signifificant differences from placebo are bevacizumab (HR = 1.83, 95%CI: 1.47 to 2.33), trebananib(HR = 1.35, 95%CI: 1.03 to 1.71), cediranib (HR = 1.79, 95%CI: 1.78 to 2.72), niraparib (HR = 2.9, 95%CI: 2.15 to 3.95), olaparib (HR = 2.74, 95%CI: 2.05 to 3.6) ,rucaparib (HR = 2.61, 95%CI: 1.81 to 3.57) and fuzuloparib (HR = 3.99, 95%CI: 2.41 to 6.6).
Compared with fuzuloparib of the top SUCRA-ranked intervention,niraparib, olaparib and rucaparib were comparable with their efficacy,bevacizumab, trebananib and cediranib were significantly inferior.Detailed results are shown in Supplementary Table 3A and Fig. 4、5、6A.
3.2.2 BRCA mutated patients
The PFS network relationship in the BRCA mutated patients is shown in Supplementary Fig. 3B,7 studies reported outcomes related to the HRs of PFS. We compared the 5 interventions included directly.
The results showed that compared with placebo, There were four treatments improved PFS,the HR ranged from 4.08 to 7.16.
The interventions with signifificant differences from placebo are niraparib (HR = 4.08, 95%CI: 1.49 to 11.33), olaparib (HR = 4.29, 95%CI: 1.9 to 9.75) ,rucaparib (HR = 4.16, 95%CI: 1.04 to 16.49) and fuzuloparib (HR = 7.16, 95%CI: 1.63 to 30.82).
Compared with fuzuloparib of the top SUCRA-ranked intervention,niraparib, olaparib and rucaparib were comparable with their efficacy.Detailed results are shown in Supplementary Table 3B and Fig. 4、5、6B.
3.2.3 BRCA wild type patients
The PFS network relationship in the BRCA wild type patients is shown in Supplementary Fig. 3C,5 studies reported outcomes related to the HRs of PFS. We compared the 5 interventions included directly.
The results showed that signifificant differences compared with placebo is niraparib (HR = 2.33, 95%CI: 1.17 to 4.7).In order of SUCRA ranking, niraparib has the highest ranking.Detailed results are shown in Supplementary Table 3C and Fig. 4、5、6C.
3.3 OS
3.3.1 All comer population
The OS network relationship in the overall population is shown in Supplementary Fig. 3D,15 studies reported outcomes related to the HRs of OS. We compared the 9 interventions included directly.The results showed that compared with placebo, disappointingly there was no statistically significant beneficial effect for OS between the maintenance treatment groups,the HR ranged from 0.89 to 1.45.Among them, the highest SUCRA ranking is niraparib.Detailed results are shown in Supplementary Table 3D and Fig. 4、5、6D.
3.3.2 BRCA mutated patients
The OS network relationship in the BRCA mutated patients is shown in Supplementary Fig. 3E,6 studies reported outcomes related to the HRs of OS. We compared the 4 interventions included directly.
The results showed that BRCA mutated patients receiving maintenance treatment no nodes of statistical significance were reached,the HR ranged from 0.83 to 1.42.Among them, the highest SUCRA ranking is olaparib.Detailed results are shown in Supplementary Table 3E and Fig. 4、5、6E.
3.3.3 BRCA wild type patients
The OS network relationship in the BRCA wild type patients is shown in Supplementary Fig. 3F,4 studies reported outcomes related to the HRs of OS. We compared the 4 interventions included directly.
The results showed that BRCA wild type patients receiving maintenance treatment also no nodes of statistical significance were reached,the HR ranged from 1.01 to 1.2.Among them, the highest SUCRA ranking is olaparib.Detailed results are shown in Supplementary Table 3F and Fig. 4、5、6F.
3.4 PFS of AC at each time point
For PFS, 16 articles reported related outcomes,network relationship is shown in Supplementary Fig. 7A. At the 8-time points of 3, 6, 9,12, 18, 24,30 and 36 month, there were sufficientdata for the NMA of PFS.We summarize the details of the interventions with significant results compared with placebo in Table 3.Detailed results of comparisons of 8 interventions at each time point are shown in the Supplementary Fig. 8–10.
At 3rd month, compared with placebo, Bevacizumab (HR = 3.24,95% Cl:1.74 to 6.37), Cediranib (HR = 7.62,95% Cl:1.53 to 64.15),Trebananib (HR = 1.72,95% Cl:1.0 to 3.03) ,Niraparib (HR = 2.9,95% Cl:1.57 to 5.92),Olaparib (HR = 4.35,95% Cl:2.24 to 8.77),Rucaparib (HR = 2.11,95% Cl:1.04 to 3.74) and Fuzuloparib (HR = 1.72,95% Cl:1.0 to 3.03) significantly increased PFS rates.
In comparison with placebo at 6th month,Bevacizumab (HR = 3.14, 95%CI: 1.93 to 5.2), Trebananib (HR = 1.79, 95%CI: 1.06 to 2.89) ,Niraparib (HR = 3.46, 95%CI: 2.02 to 6.31),Olaparib (HR = 5.63, 95%CI: 3.28 to 10.6),Rucaparib(HR = 2.91, 95%CI: 1.51 to 4.96) and Fuzuloparib (HR = 1.79, 95%CI: 1.06 to 2.89) significantly increased PFS rates.
At 9th month, the PFS for Bevacizumab (HR = 2.61, 95% Cl: 1.46 to 4.75), Niraparib (HR = 4.11, 95% Cl: 1.98 to 8.95),Olaparib (HR = 5.42, 95% Cl: 2.75 to 10.34),Rucaparib(HR = 3.62, 95% Cl: 1.57 to 7.48) and Fuzuloparib (HR = 5.13, 95%CI: 1.7 to 15.37)were significantly higher compared to that of placebo.
At 12th month, the PFS for Bevacizumab(HR = 3.08, 95%CI:1.61 to 6), Niraparib (HR = 4.66, 95%CI:2.04 to 11.06),Olaparib (HR = 7.49, 95%CI:3.72 to 15.69),Rucaparib (HR = 5.86, 95%CI:2.22 to 13.87) and Fuzuloparib (HR = 6.19, 95%CI: 1.81 to 21.92)were significantly higher compared to that of placebo.
At 18th month, the PFS was significantly higher for Bevacizumab(HR = 5.1, 95%CI: 1.34 to 19.92), Niraparib (HR = 5.67, 95%CI: 1.11 to 30.57) ,Olaparib(HR = 15.94, 95%CI: 4.15 to 77.07) and Rucaparib(HR = 7.17, 95%CI: 1.22 to 40.83) compared to placebo.
At 24th month, the PFS increased significantly when Bevacizumab (HR = 2.78, 95%CI: 1.38 to 6.26), Niraparib (HR = 3.04, 95%CI: 1.26 to 7.86) ,Olaparib(HR = 4.23, 95%CI: 1.1 to 15.48) and Rucaparib (HR = 11.04, 95%CI: 3.02 to 36.71) were compared with placebo.
At 30th month, the PFS was significantly higher for Rucaparib (HR = 9.08, 95%CI: 1.01 to 84.13) compared to placebo.
At 36th month, compared with placebo, any interventions were not significantly increased
PFS rates. Therefore, at 36rd month,there were no interventions with significant results.
In PFS, compared with placebo, Rucaparib was significant from 6 to 30 months; Bevacizumab, Niraparib and Olaparib significant from 3 to 24 months; Fuzuloparib significant from 3 to 12 months;trebananib significant from 3 to 6 months;cediranib only significant at 3rd month. However,at each month point, compared with placebo, nintedanib wasn’t significantly increased PFS rates.
3.5 OS of AC at each time point
For OS, 13 articles reported related outcomes, network relationship is shown in Supplementary Fig. 7B,there were sufficientdata for the NMA of OS.We summarize the details of the interventions with significant results compared with placebo in Table 4.Detailed results of comparisons of 8 interventions at each time point are shown in the Supplementary Fig. 11–13.
At 3rd month, compared with placebo, Olaparib (HR = 4.43,95% Cl:1.21 to 17.04) significantly increased OS rates.
In comparison with placebo at 6th month,Olaparib (HR = 6.37, 95%CI: 1.95 to 36.42)significantly increased OS rates.
At 9th month, the OS rate for Olaparib (HR = 5.77, 95% Cl: 1.33 to 33.18)was significantly higher compared to that of placebo.
At 12th month, the OS rate for Olaparib (HR = 7.58, 95%CI:3.04 to 19.6) was significantly higher compared to that of placebo.
At 18th month, the OS rate was significantly higher for Olaparib(HR = 24.39, 95%CI: 3.88 to 413.19) compared to placebo.
At 24th month, the OS rate increased significantly when Olaparib(HR = 26.55, 95%CI: 3.58 to 2080.51) was compared with placebo.
At 30th month, the OS rate was significantly higher for Olaparib (HR = 10.81, 95%CI: 1.55 to 485.85) compared to placebo.
At 36th month, OS rate was significantly increased by Olaparib (HR = 25.04, 95%CI: 4.72 to 415.94) in comparison with placebo.
In OS,compared with placebo, the intervention measures with significant effect from 3 to 36 months was Olaparib, other interventions were not significantly increased OS rates.
3.6 PFS of BRCAm at each time point
For PFS of BRCAm, 7 articles reported related outcomes,network relationship is shown in Supplementary Fig. 7C, there were sufficientdata for the NMA.We summarize the details of the interventions with significant results compared with placebo in Table 5.Detailed results of comparisons of 8 interventions at each time point are shown in the Supplementary Fig. 14–16.
At 3rd month, PFS rate of BRCAm was significantly increased by Olaparib (HR = 6.49, 95% Cl: 1.5 to 29.2),Rucaparib(HR = 10.11, 95% Cl: 1.21 to 90.29) and Fuzuloparib (HR = 28.89, 95%CI: 1.95 to 1265.75) in comparison with placebo.
At 6th month, in comparison with placebo, Niraparib (HR = 4.17, 95%CI: 1.07 to 18.96),Olaparib (HR = 7.52, 95% Cl: 2.11 to 33.02),Rucaparib(HR = 7.7, 95% Cl: 1.05 to 58.43) and Fuzuloparib (HR = 12.35, 95%CI: 1.52 to 105.68) were more effective.
At 9th month, Niraparib (HR = 9.3, 95%CI: 1.53 to 74.58),Olaparib (HR = 7.55, 95% Cl: 1.4 to 35.12) and Fuzuloparib (HR = 17.1, 95%CI: 1.09 to 287.23) significantly increased PFS rate of BRCAm compared to placebo.
At 12th month, there was a significant increase in PFS rate of BRCAm with Niraparib (HR = 5.94, 95%CI: 1.2 to 39.58),Olaparib (HR = 8.09, 95% Cl: 2.19 to 38.03) and Fuzuloparib (HR = 35.69, 95%CI: 2.46 to 674.12) compared with placebo.
At 18th month, PFS rate of BRCAm was significantly higher with Olaparib (HR = 32.09, 95% Cl: 1.49 to 2116.8) than with placebo.
Among the treatments tested at 24 to 36 months, compared with placebo, any interventions were not significantly increased the PFS rate of BRCAm.
In summary,for PFS of BRCAm,Olaparib significant from 3 to 18 months,according to duration olaparib may be the first choice for improving PFS of BRCAm.
3.7 PFS of BRCAwt at each time point
For PFS of BRCAwt, 5 articles reported related outcomes, network relationship is shown in Supplementary Fig. 7D,there were sufficientdata for the NMA.We summarize the details of the interventions with significant results compared with placebo in Table 6.Detailed results of comparisons of 8 interventions at each time point are shown in the Supplementary Fig. 17–19.
At 6th month, in comparison with placebo, Niraparib (HR = 2.91, 95%CI: 1.03 to 8.47) was more effective.
At 9th month, Niraparib (HR = 3.82, 95%CI: 1.07 to 14.05) and Rucaparib (HR = 5.95, 95%CI: 1.0 to 36.02) significantly increased PFS rate of BRCAwt compared to placebo.
At 12th month, there was a significant increase in PFS rate of BRCAwt with Niraparib (HR = 4.17, 95%CI: 1.01 to 17.2)and Rucaparib (HR = 7.11, 95%CI: 1.0 to 51.26) compared with placebo.
Among the treatments tested at3and 18 to 36 months, compared with placebo, any interventions were not significantly increased the PFS rate of BRCAwt.
In summary,for PFS of BRCAwt, Niraparib significant from 6 to 12 months,according to duration niraparib may be the best choice for improving PFS of BRCAwt.
Table 3
PFS for each time point for interventions that were significant compared to placebo (shown as HR and 95% confidence intervals).
Time point
|
bevacizumab
|
cediranib
|
fuzuloparib
|
nintedanib
|
niraparib
|
olaparib
|
rucaparib
|
trebananib
|
3 month
|
3.24 (1.74, 6.37)
|
7.62 (1.53, 64.15)
|
4.45 (1.64, 12.5)
|
×
|
2.9 (1.57, 5.92)
|
4.35 (2.24, 8.77)
|
2.11 (1.04, 3.74)
|
1.72 (1, 3.03)
|
6 month
|
3.14 (1.93, 5.2)
|
×
|
5.14 (2.18, 12.33)
|
×
|
3.46 (2.02, 6.31)
|
5.63 (3.28, 10.6)
|
2.91 (1.51, 4.96)
|
1.79 (1.06, 2.89)
|
9 month
|
2.61 (1.46, 4.75)
|
×
|
5.13 (1.7, 15.37)
|
×
|
4.11 (1.98, 8.95)
|
5.42 (2.75, 10.34)
|
3.62 (1.57, 7.48)
|
×
|
12 month
|
3.08 (1.61, 6)
|
×
|
6.19 (1.81, 21.92)
|
×
|
4.66 (2.04, 11.06)
|
7.49 (3.72, 15.69)
|
5.86 (2.22, 13.87)
|
×
|
18 month
|
5.1 (1.34, 19.92)
|
×
|
×
|
×
|
5.67 (1.11, 30.57)
|
15.94 (4.15, 77.07)
|
7.17 (1.22, 40.83)
|
×
|
24 month
|
2.78 (1.38, 6.26)
|
×
|
×
|
×
|
3.04 (1.26, 7.86)
|
4.23 (1.1, 15.48)
|
11.04 (3.02, 36.71)
|
×
|
30 month
|
×
|
×
|
×
|
×
|
×
|
×
|
9.08 (1.01, 84.13)
|
×
|
36 month
|
×
|
×
|
×
|
×
|
×
|
×
|
×
|
×
|
×: the treatment on the top is not significant compared to the Control group on the left.
|
Table 4
OS for each time point for interventions that were significant compared to placebo(shown as HR and 95% confidence intervals).
Time point
|
olaparib
|
bevacizumab
|
cediranib
|
nintedanib
|
niraparib
|
rucaparib
|
trebananib
|
fuzuloparib
|
3 month
|
4.43 (1.21, 17.04)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
6 month
|
6.37 (1.95, 36.42)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
9 month
|
5.77 (1, 33.18)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
12 month
|
7.58 (3.04, 19.6)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
18 month
|
24.39 (3.88, 413.19)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
24 month
|
26.55 (3.58, 2080.51)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
30 month
|
10.81 (1.55, 485.85)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
36 month
|
25.04 (4.72, 415.94)
|
×
|
×
|
×
|
×
|
×
|
×
|
-
|
×: the treatment on the top is not significant compared to the Control group on the left.
|
Table 5 BRCAm PFS for each time point for interventions that were significant compared to placebo (shown as HR and 95% confidence intervals).
|
Time point
|
niraparib
|
olaparib
|
rucaparib
|
fuzuloparib
|
3 month
|
×
|
6.49 (1.5, 29.2)
|
10.11 (1.21, 90.29)
|
28.89 (1.95, 1265.75)
|
6 month
|
4.17 (1.07, 18.96)
|
7.52 (2.11, 33.02)
|
7.7 (1.05, 58.43)
|
12.35 (1.52, 105.68)
|
9 month
|
9.3 (1.53, 74.58)
|
7.55 (1.4, 35.12)
|
×
|
17.1 (1.09, 287.23)
|
12 month
|
5.94 (1.2, 39.58)
|
8.09 (2.19, 38.03)
|
×
|
34.69 (2.46, 674.12)
|
18 month
|
×
|
32.09 (1.49, 2116.84)
|
×
|
×
|
24 month
|
×
|
×
|
×
|
×
|
30 month
|
×
|
×
|
×
|
×
|
36 month
|
×
|
×
|
×
|
×
|
×: the treatment on the top is not significant compared to the Control group on the left.
Table 6
BRCAwt PFS for each time point for interventions that were significant compared to placebo (shown as HR and 95% confidence intervals)
Time point
|
niraparib
|
olaparib
|
rucaparib
|
fuzuloparib
|
3 month
|
×
|
×
|
×
|
×
|
6 month
|
2.91 (1.03, 8.47)
|
×
|
×
|
×
|
9 month
|
3.82 (1.07, 14.05)
|
×
|
5.95 (1, 36.02)
|
×
|
12 month
|
4.17 (1.01, 17.2)
|
×
|
7.11 (1, 51.26)
|
×
|
18 month
|
×
|
×
|
×
|
×
|
24 month
|
×
|
×
|
×
|
×
|
30 month
|
×
|
×
|
×
|
×
|
36 month
|
×
|
×
|
×
|
×
|
×: the treatment on the top is not significant compared to the Control group on the left.
3.8 Heterogeneity and network meta-regression analysis
In order to better explain the heterogeneity, we performed a sensitivity analysis of the outcomes. Meta-regression analysis was performed on PFS and OS with the BRCA genes status as the covariate.
For PFS, the network relationship is shown in Supplementary Fig. 20A; after meta-regression analysis, in the BRCA-mutation group,niraparib (HR = 4, 95%CI: 2.0 to 7.7), Olaparib (HR = 4.3, 95%CI: 2.4 to 7.1) ,rucaparib (HR = 4.0, 95%CI: 1.6 to 9.3) and fuzuloparib (HR = 6.6, 95%CI: 2.3 to 18)are significantly compared with placebo.In the BRCA wild-type group,niraparib (HR = 2.4, 95%CI: 1.3 to 4.5)is significantly compared with placebo; Olaparib,rucaparib, and fuzuloparib are not consistently better than placebo.The detailed results are shown in Fig. 3.
For OS, the network relationship is shown in Supplementary Fig. 20B; after meta-regression analysis,regardless of BRCA gene status,there was no statistically significant beneficial effect for OS between the maintenance treatment groups. The detailed results are shown in Fig. 4.
For PFS at each time point, the network relationship is shown in Supplementary Fig. 20C; in the BRCA-mutation group, Olaparib and niraparib significantly increased PFS rates, olaparib from 3rd month to12th month, niraparib from 6th month to12th month,rucaparib and fuzuloparib increased PFS rates only at one month.In the BRCA wild-type group, Olaparib and rucaparib were consistently better than the placebo in the 12th month.The detailed results are shown in Fig. 5.
For all outcomes, it was revealed in Brooks-GelmanRubin diagnostic that the inferential iterations for each MCMC were stable and reproducible. We used the history feature to confirm the convergence of the model in all outcomes as well.