Background The responses of cancer patients to immune checkpoint inhibitors (ICIs) vary in success. CD8+ tumor infiltrating lymphocytes (TILs) play a key role in killing tumor cells. This study aims to evaluate the prognostic role of CD8+ TILs in cancer patients treated with ICIs. Methods We systematically searched all publications from PubMed, EMBASE, and Cochrane Library until 12 Jul 2021 without any restriction of language or article types. Studies assessing high versus low CD8+ TILs in predicting efficacy and survival of various cancer patients were included. The outcomes included overall survival (OS), progression-free survival (PFS), and objective response rate (ORR). The study protocol is prospectively registered on PROSPERO (registration number CRD42021233654). Findings Findings: A total of 33 studies consisting of 2559 cancer patients were included. The result showed that high CD8+ TILs were significantly associated with better OS (HR, 0.52; 95% confidence interval: 0.41–0.67; p < 0.001), PFS (HR, 0.52; 95% confidence interval: 0.40–0.67; p < 0.001) and ORR (OR, 4.08; 95% confidence interval: 2.73–6.10; p < 0.001) in patients treated with ICIs. Subgroup analyses suggested that patients with high CD8+ TILs had a better clinical benefit, regardless of different treatments (ICI mono therapy, or combination therapy), cancer types (NSCLC, melanoma and others), and CD8+ T cells locations (intra-tumor, stroma, and invasive margin). The higher baseline circulating CD8+ T cells from peripheral blood did not contribute to the improved OS (HR, 0.93; 95% confidence interval: 0.67–1.29; p = 0.67) and PFS (HR, 0.89; 95% confidence interval: 0.60–1.32; p = 0.56) compared with the low baseline. Interpretation Interpretation: Our results suggested that high intra-tumoral, stromal, or invasive marginal, but not circulating CD8+ T cells, can predict treatment outcomes in patients with ICIs therapy across different cancers, in either single-agent ICIs or combination with other therapies. Funding Funding: China National Science Foundation (Grant No. 82,022,048, 81,871,893), Key Project of Guangzhou Scientific Research Project (Grant No. 201,804,020,030), High-level university construction project of Guangzhou medical university (Grant No. 20,182,737, 201,721,007, 201,715,907, 2,017,160,107); National key R & D Program (Grant No. 2017YFC0907903 & 2017YFC0112704) and the Guangdong high level hospital construction "reaching peak" plan.

Background: Spontaneous ventilation video-assisted thoracoscopic surgery (SV-VATS) exhibits dual intraoperative and postoperative advantages for patients with non-small cell lung cancer (NSCLC).However, there is a lack of data regarding its long-term survival superiority over the double-lumen intubated mechanical ventilation video-assisted thoracoscopic surgery (MV-VATS) or thoracotomy.Methods: A retrospective study was conducted from 2011 to 2018 in the First Affiliated Hospital of Guangzhou Medical University among patients with NSCLC who underwent the SV-VATS or the MV-VATS. Patients receiving the SV-VATS were the study group, and patients receiving the MV-VATS were the control group. Propensity score matching (PSM) was performed to establish 1:1 SV-VATS versus MV-VATS group matching to balance potential baseline confounding factors. Primary endpoints were overall survival (OS) and disease-free survival (DFS). Secondary endpoints were perioperative outcomes. The baseline information of these patients was recorded. The perioperative data and survival data were collected using a combination of electronic data record system and telephone interview. A 1:1:1 SPM was also used to compare the OS in the SV-VATS, the MV-VATS and thoracotomy group by using another database, including patients undergoing thoracotomy and the MV-VATS.Results: For the two-group comparison, after 1:1 PSM, a matched cohort with 400 (200:200) patients was generated. The median follow-up time in this cohort was 4.78 years (IQR, 3.78-6.62 years). The OS (HR =0.567, 95% CI, 0.330 to 0.974, P=0.0498) and the DFS (HR =0.546, 95% CI, 0.346 to 0.863, P=0.013) of the SV-VATS group were significantly better than the MV-VATS group. There were no statistically differences between the SV-VATS and the MV-VATS group on the operative time (158.56±40.09 vs.

No abstract