Literature search strategy
A detailed review of English and Chinese literature was conducted from PubMed, Embase, Cochrane Library, CBM, CNKI, using the terms ( “liposom* and doxorubicin OR DOX-SL OR Lipodox OR Doxil OR Caelyx OR Lipo-Dox OR DaunoXome” ) and (“cancer OR tumor OR tumour OR neoplasm OR neoplasma OR neoplasia OR carcinoma”). The literature search was last updated on March 21st , 2019. We also searched the reference of the relevant review articles to seek for the potentially included studies. We did not try to contact the corresponding authors if the survival analysis data could not obtained. The PRISMA statement for reporting systematic reviews and meta-analyses was cited in this meta-analysis [10]. In addition, the ethical approval was not applied in current study because there was no patient’s privacy or clinical samples.
Inclusion and exclusion criteria
Relevant studies should meet the following criteria before being included: (1) the included studies focused on the association of cancer patients undergoing liposomal doxorubicin-based chemotherapy with prognosis; (2) the studies provided survival analysis data of HR and 95%CI or a survival curve from which we could extract the HR, 95%CI; As per the exclusion criteria: (1) no survival analysis data; (2) studies involved cell lines and animals; (3) similar or duplicate study; (4) other type articles including reviews, case reports, letters.
Data extraction and quality assessment
We extracted the relevant data including author, publication year, country, no of patients, age, cancer type, treatment arms, phase, follow-up time, survival outcomes and HR (95%CI) from included study. The data were independently extracted by 2 authors (Kaiping Zhang and Zhi Sun). Disagreements were resolved by discussion or reviewed by a third author (Min Chao).
Among the relevant data, country came from USA, UK, Greece and others; Sample size was separated into ≥100 and <100; Cancer type included soft tissue sarcoma (STS), multiple myeloma (MM), acute myelogenous leukaemia (AML), non-small cell lung cancer (NSCLC), oesophago-gastric (OG) cancer, acute lymphoblastic leukemia (ALL), metastatic breast cancer (MBC), ovarian cancer (OG). Survival outcomes contained overall survival (OS) and progression-free survival (PFS). The quality of included studies was assessed with the Newcastle-Ottawa Scale (NOS), which is composed of three domains: selection, comparability, and exposure. The NOS is a semi-quantitative scale, from which a score of 0-9 was assigned to each study. A total score of ≤3 was considered poor quality, 4-6 was considered moderate qualify, and 7-9 was deemed high quality.
Statistical analysis
We explored the prognosis of cancer patients undergoing liposomal doxorubicin-based chemotherapy by using Review Manager 5.3 (Cochrane Collaboration, Oxford, UK) and STATA 12.0 software (Stata Corpotation, College Station, TX). HR(95% CI) was obtained for assessing the prognosis of cancer patients after liposomal doxorubicin-based chemotherapy. Meanwhile, the Q statistics and I2 test were applied to calculate the heterogeneity of eligible study. P < 0.05 and/or I2 > 50% were considered as statistically heterogeneous, and random effects (DerSimonian and Laird method) model was used to pool the results [11, 12]. Otherwise, fixed effects (Mantel-Haenszel method) model was applied [13].
One-way sensitivity analyses removed each included studies were performed to assess the pooled results’ stability. Moreover, the publication bias was assessed using Begg's test. P<0.05 indicated that there was a bias of this meta-analysis [14]. Additionally, different subgroups consisted of country, cancer type and sample size were analysed in current meta-analysis.