Background: Early relapse in multiple myeloma (MM) is a major predictor of poor prognosis, regardless of cytogenetic risk or treatment intensity. Methods: Here we analyzed 1026 MM patients treated across 12 Italian hematology centers. FHR was defined as progression-free survival (PFS) ≤18 months in transplant-eligible (TE) and ≤12 months in non-transplant-eligible (NTE) patients. Logistic regression and ROC analysis were used to identify significant predictors of FHR and build a risk score. Results: FHR status was identified in 175 patients (17%). These patients had significantly shorter PFS (7 vs. 57.5 months) and overall survival (19 months vs. not reached;p< 0.001). FHR status was associated with higher median LDH, lower Hb level, higher creatinine level and lower platelets count. Modified EASIX formula was built by these significant continuous variables, to be tested in a logistic analysis: [(LDH × creatinine)/(Hb × PLT) × 100]. A significantly higher rate of FHR was found with a score > 2.0 (89% vs. 11%,p< 0.001). Multivariate logistic analysis selected the above formula, ECOG PS ≥ 2 and ISS III as factors associated with FHR. Scoring these variables according to OR, three groups of patients were segregated with a rate of FHR patients of 7%, 29.5%, and 63.5%, respectively. Treatment with anti-CD38 monoclonal antibodies was associated with lower FHR frequency. Conclusions: This study proposes a simple, clinically applicable model to identify FHR MM patients early in their disease course. However, very in-depth biological tools, not available in clinical practice, are needed to identify singularly risk of becoming FHR.
背景：多发性骨髓瘤（MM）的早期复发是预后不良的主要预测因素，无论细胞遗传学风险或治疗强度如何。方法：本研究分析了意大利12个血液中心收治的1026例MM患者。早期复发高风险（FHR）定义为：适合移植（TE）患者无进展生存期（PFS）≤18个月，不适合移植（NTE）患者PFS≤12个月。采用逻辑回归和ROC分析确定FHR的显著预测因子并构建风险评分。结果：共175例患者（17%）被确定为FHR状态。这些患者的PFS（7个月 vs. 57.5个月）和总生存期（19个月 vs. 未达到；p<0.001）均显著缩短。FHR状态与较高的中位乳酸脱氢酶（LDH）、较低的血红蛋白（Hb）水平、较高的肌酐水平及较低的血小板计数相关。基于这些显著的连续变量构建了改良EASIX公式用于逻辑分析：[（LDH × 肌酐）/（Hb × 血小板计数）× 100]。评分>2.0的患者FHR发生率显著更高（89% vs. 11%，p<0.001）。多变量逻辑分析筛选出上述公式、ECOG PS ≥ 2及ISS III期作为与FHR相关的因素。根据比值比（OR）对这些变量进行评分，可将患者分为三组，其FHR发生率分别为7%、29.5%和63.5%。抗CD38单克隆抗体治疗与较低的FHR发生率相关。结论：本研究提出了一种简单、临床适用的模型，可在疾病早期识别FHR MM患者。然而，要单独识别进展为FHR的风险，仍需目前临床实践中尚不具备的深度生物学检测工具。
肿瘤(癌症)患者之家