Clinic of Infectious Diseases, Catholic University of Sacred Heart, Rome, Italy.
PLoS One. 2010 Oct 29;5(10):e13753. doi: 10.1371/journal.pone.0013753.
Although genotypic resistance testing (GRT) is recommended to guide combination antiretroviral therapy (cART), funding and/or facilities to perform GRT may not be available in low to middle income countries. Since treatment history (TH) impacts response to subsequent therapy, we investigated a set of statistical learning models to optimise cART in the absence of GRT information.
The EuResist database was used to extract 8-week and 24-week treatment change episodes (TCE) with GRT and additional clinical, demographic and TH information. Random Forest (RF) classification was used to predict 8- and 24-week success, defined as undetectable HIV-1 RNA, comparing nested models including (i) GRT+TH and (ii) TH without GRT, using multiple cross-validation and area under the receiver operating characteristic curve (AUC). Virological success was achieved in 68.2% and 68.0% of TCE at 8- and 24-weeks (n = 2,831 and 2,579), respectively. RF (i) and (ii) showed comparable performances, with an average (st.dev.) AUC 0.77 (0.031) vs. 0.757 (0.035) at 8-weeks, 0.834 (0.027) vs. 0.821 (0.025) at 24-weeks. Sensitivity analyses, carried out on a data subset that included antiretroviral regimens commonly used in low to middle income countries, confirmed our findings. Training on subtype B and validation on non-B isolates resulted in a decline of performance for models (i) and (ii).
Treatment history-based RF prediction models are comparable to GRT-based for classification of virological outcome. These results may be relevant for therapy optimisation in areas where availability of GRT is limited. Further investigations are required in order to account for different demographics, subtypes and different therapy switching strategies.
尽管基因耐药性检测（GRT）被推荐用于指导联合抗逆转录病毒治疗（cART），但在中低收入国家可能没有资金和/或设施来进行 GRT。由于治疗史（TH）会影响后续治疗的反应，我们研究了一组统计学习模型，以在没有 GRT 信息的情况下优化 cART。
使用 EuResist 数据库提取了有 GRT 和额外的临床、人口统计学和 TH 信息的 8 周和 24 周治疗变化期（TCE）。随机森林（RF）分类用于预测 8 周和 24 周的成功，定义为无法检测到 HIV-1 RNA，比较了包括（i）GRT+TH 和（ii）无 GRT 的 TH 的嵌套模型，使用了多次交叉验证和接收者操作特征曲线下的面积（AUC）。在 8 周和 24 周时，分别有 68.2%和 68.0%的 TCE 达到病毒学成功（n=2831 和 2579）。RF（i）和（ii）表现相当，平均（标准差）AUC 分别为 0.77（0.031）和 0.757（0.035）在 8 周时，0.834（0.027）和 0.821（0.025）在 24 周时。在包括中低收入国家常用抗逆转录病毒方案的数据集子集上进行的敏感性分析证实了我们的发现。在基于 B 亚型和非 B 分离物的验证上，模型（i）和（ii）的性能下降。
基于治疗史的 RF 预测模型与基于 GRT 的模型在病毒学结果的分类上相当。这些结果可能与 GRT 有限的地区的治疗优化相关。需要进一步的研究，以考虑不同的人口统计学、亚型和不同的治疗转换策略。
