RESUMO
Large language models (LLMs) may facilitate and expedite systematic reviews, although the approach to integrate LLMs in the review process is unclear. This study evaluates GPT-4 agreement with human reviewers in assessing the risk of bias using the Risk Of Bias In Non-randomised Studies of Interventions (ROBINS-I) tool and proposes a framework for integrating LLMs into systematic reviews. The case study demonstrated that raw per cent agreement was the highest for the ROBINS-I domain of 'Classification of Intervention'. Kendall agreement coefficient was highest for the domains of 'Participant Selection', 'Missing Data' and 'Measurement of Outcomes', suggesting moderate agreement in these domains. Raw agreement about the overall risk of bias across domains was 61% (Kendall coefficient=0.35). The proposed framework for integrating LLMs into systematic reviews consists of four domains: rationale for LLM use, protocol (task definition, model selection, prompt engineering, data entry methods, human role and success metrics), execution (iterative revisions to the protocol) and reporting. We identify five basic task types relevant to systematic reviews: selection, extraction, judgement, analysis and narration. Considering the agreement level with a human reviewer in the case study, pairing artificial intelligence with an independent human reviewer remains required.
RESUMO
Clinical trials of prostate-specific membrane antigen (PSMA) targeted radiopharmaceuticals have shown encouraging results. Some agents, like lutetium-177 [177Lu]Lu-PSMA-617 ([177Lu]Lu-PSMA-617), are already approved for late line treatment of metastatic castration-resistant prostate cancer (mCRPC). Projections are for continued growth of this treatment modality; [177Lu]Lu-PSMA-617 is being studied both in earlier stages of disease and in combination with other anti-cancer therapies. Further, the drug development pipeline is deep with variations of PSMA-targeting radionuclides, including higher energy alpha particles conjugated to PSMA-honing vectors. It is safe to assume that an increasing number of patients will be exposed to PSMA-targeted radiopharmaceuticals during the course of their cancer treatment. In this setting, it is important to better understand and mitigate the most commonly encountered toxicities. One particularly vexing side effect is xerostomia. In this review, we discuss the scope of the problem, inventories to better characterize and monitor this troublesome side effect, and approaches to preserve salivary function and effectively palliate symptoms. This article aims to serve as a useful reference for prescribers of PSMA-targeted radiopharmaceuticals, while also commenting on areas of missing data and opportunities for future research.