Spinner Michael A, Aleshin Alexey, Santaguida Marianne T, Schaffert Steven A, Zehnder James L, Patterson A Scott, Gekas Christos, Heiser Diane, Greenberg Peter L
Division of Hematology, Department of Medicine, Stanford University Medical Center, Stanford, CA; and.
Notable Labs, Foster City, CA.
Blood Adv. 2020 Jun 23;4(12):2768-2778. doi: 10.1182/bloodadvances.2020001934.
Precision medicine approaches such as ex vivo drug sensitivity screening (DSS) are appealing to inform rational drug selection in myelodysplastic syndromes (MDSs) and acute myeloid leukemia, given their marked biologic heterogeneity. We evaluated a novel, fully automated ex vivo DSS platform that uses high-throughput flow cytometry in 54 patients with newly diagnosed or treatment-refractory myeloid neoplasms to evaluate sensitivity (blast cytotoxicity and differentiation) to 74 US Food and Drug Administration-approved or investigational drugs and 36 drug combinations. After piloting the platform in 33 patients, we conducted a prospective feasibility study enrolling 21 patients refractory to hypomethylating agents (HMAs) to determine whether this assay could be performed within a clinically actionable time frame and could accurately predict clinical responses in vivo. When assayed for cytotoxicity, ex vivo drug sensitivity patterns were heterogeneous, but they defined distinct patient clusters with differential sensitivity to HMAs, anthracyclines, histone deacetylase inhibitors, and kinase inhibitors (P < .001 among clusters) and demonstrated synergy between HMAs and venetoclax (P < .01 for combinations vs single agents). In our feasibility study, ex vivo DSS results were available at a median of 15 days after bone marrow biopsy, and they informed personalized therapy, which frequently included venetoclax combinations, kinase inhibitors, differentiative agents, and androgens. In 21 patients with available ex vivo and in vivo clinical response data, the DSS platform had a positive predictive value of 0.92, negative predictive value of 0.82, and overall accuracy of 0.85. These data demonstrate the utility of this approach for identifying potentially useful and often novel therapeutic drugs for patients with myeloid neoplasms refractory to standard therapies.
鉴于骨髓增生异常综合征（MDS）和急性髓系白血病具有显著的生物学异质性，体外药敏筛选（DSS）等精准医学方法对于指导合理用药具有吸引力。我们评估了一种新型的全自动体外DSS平台，该平台使用高通量流式细胞术对54例新诊断或难治性髓系肿瘤患者进行检测，以评估其对74种美国食品药品监督管理局批准或正在研究的药物及36种药物组合的敏感性（原始细胞细胞毒性和分化情况）。在33例患者中对该平台进行试点后，我们开展了一项前瞻性可行性研究，纳入21例对去甲基化药物（HMA）耐药的患者，以确定该检测能否在临床可操作的时间范围内完成，并能否准确预测体内临床反应。在检测细胞毒性时，体外药敏模式具有异质性，但它们定义了对HMA、蒽环类药物、组蛋白去乙酰化酶抑制剂和激酶抑制剂具有不同敏感性的不同患者群体（各群体之间P < 0.001），并证明了HMA与维奈克拉之间存在协同作用（联合用药与单药相比P < 0.01）。在我们的可行性研究中，骨髓活检后中位15天可获得体外DSS结果，这些结果为个性化治疗提供了依据，个性化治疗通常包括维奈克拉联合用药、激酶抑制剂、分化剂和雄激素。在21例有体外和体内临床反应数据的患者中，DSS平台的阳性预测值为0.92，阴性预测值为0.82，总体准确率为0.85。这些数据证明了该方法对于为标准治疗难治的髓系肿瘤患者识别潜在有用且往往是新型治疗药物的实用性。
