Soldatos Theodoros G, Kaduthanam Sajo, Jackson David B
Molecular Health GmbH, 69115 Heidelberg, Germany.
J Pers Med. 2019 Sep 5;9(3):43. doi: 10.3390/jpm9030043.
The molecular characterization of patient tumors provides a rational and highly promising approach for guiding oncologists in treatment decision-making. Notwithstanding, genomic medicine still remains in its infancy, with innovators and early adopters continuing to carry a significant portion of the clinical and financial risk. Numerous innovative precision oncology trials have emerged globally to address the associated need for evidence of clinical utility. These studies seek to capitalize on the power of predictive biomarkers and/or treatment decision support analytics, to expeditiously and cost-effectively demonstrate the positive impact of these technologies on drug resistance/response, patient survival, and/or quality of life. Here, we discuss the molecular foundations of these approaches and highlight the diversity of innovative trial strategies that are capitalizing on this emergent knowledge. We conclude that, as increasing volumes of clinico-molecular outcomes data become available, in future, we will begin to transition away from expert systems for treatment decision support (TDS), towards the power of AI-assisted TDS-an evolution that may truly revolutionize the nature and success of cancer patient care.
患者肿瘤的分子特征分析为指导肿瘤学家进行治疗决策提供了一种合理且极具前景的方法。尽管如此,基因组医学仍处于起步阶段,创新者和早期采用者仍需承担很大一部分临床和财务风险。全球涌现出众多创新的精准肿瘤学试验,以满足对临床效用证据的相关需求。这些研究旨在利用预测性生物标志物和/或治疗决策支持分析的力量,迅速且经济高效地证明这些技术对耐药性/反应、患者生存率和/或生活质量的积极影响。在此,我们讨论这些方法的分子基础,并强调利用这一新兴知识的创新试验策略的多样性。我们得出结论,随着越来越多的临床分子结果数据可用,未来我们将开始从用于治疗决策支持(TDS)的专家系统转向人工智能辅助TDS的力量——这一演变可能真正彻底改变癌症患者护理的性质和成功率。