Disentangling the Association of Hydroxychloroquine Treatment with Mortality in Covid-19 Hospitalized Patients through Hierarchical Clustering

Abstract

The efficacy of hydroxychloroquine (HCQ) in treating SARS-CoV-2 infection is harshly debated, with observational and experimental studies reporting contrasting results. To clarify the role of HCQ in Covid-19 patients, we carried out a retrospective observational study of 4,396 unselected patients hospitalized for Covid-19 in Italy (February–May 2020). Patients’ characteristics were collected at entry, including age, sex, obesity, smoking status, blood parameters, history of diabetes, cancer, cardiovascular and chronic pulmonary diseases, and medications in use. These were used to identify subtypes of patients with similar characteristics through hierarchical clustering based on Gower distance. Using multivariable Cox regressions, these clusters were then tested for association with mortality and modification of effect by treatment with HCQ. We identified two clusters, one of 3,913 younger patients with lower circulating inflammation levels and better renal function, and one of 483 generally older and more comorbid subjects, more prevalently men and smokers. The latter group was at increased death risk adjusted by HCQ (HR[CI95%] = 3.80[3.08-4.67]), while HCQ showed an independent inverse association (0.51[0.43-0.61]), as well as a significant influence of cluster∗HCQ interaction ( p < 0.001 ). This was driven by a differential association of HCQ with mortality between the high (0.89[0.65-1.22]) and the low risk cluster (0.46[0.39-0.54]). These effects survived adjustments for additional medications in use and were concordant with associations with disease severity and outcome. These findings suggest a particularly beneficial effect of HCQ within low risk Covid-19 patients and may contribute to clarifying the current controversy on HCQ efficacy in Covid-19 treatment.

Hydroxichloroquine for COVID-19 infection: Do we have a final word after one year?

“…Using the same CORIST dataset, including more than 4,000 COVID-19 hospitalized patients, we tried to disentangle the possible association of HCQ with mortality through an unsupervised machine learning analysis ( 10 , 11 ). Using hierarchical clustering, we could identify two clusters of COVID-19 patients based on their socio-demographic and clinical characteristics: one, including younger patients with lower circulating inflammation levels and better renal function, and the other composed of generally older and more co-morbid subjects, more prevalently men and smokers.…”

“…Using hierarchical clustering, we could identify two clusters of COVID-19 patients based on their socio-demographic and clinical characteristics: one, including younger patients with lower circulating inflammation levels and better renal function, and the other composed of generally older and more co-morbid subjects, more prevalently men and smokers. Interestingly, HCQ appeared to be associated with reduced mortality only in the low (hazard ratio 0.46, 95%CI: 0.39 to 0.54) but not in the high-risk cluster (hazard ratio 0.89, 95%CI: 0.65 to 1.22; P for difference P<0.001), suggesting a selective beneficial effect of HCQ in low risk COVID-19 patients ( 11 ). This was in line with the findings of the IDENTIFY study, a multicentre US clinical trial ( 12 ); in that study, HCQ treatment was associated with higher survival in the treated harm, especially in those patients that were predicted to benefit most, based on a supervised machine learning algorithm applied to their characteristics.…”