• Covid-19
• Antirheumatic Agents
• Inflammation

We read with great interest the letter by Kawada in which the studies assessing colchicine as a treatment for COVID-19 were reviewed.1 Given its known inhibitory effect on NACHT-LRRPYD-containing protein 3 inflammasome2 and its possible antiviral properties,2 colchicine was hypothesised as a candidate therapy for COVID-19 since the very beginning of the pandemic.3 4 To the best of our knowledge, our observational retrospective study5 reported on the largest number of patients with COVID-19 treated with colchicine (122 consecutive patients), and they were compared with a control group treated with the Standard of Care (SoC, 140 consecutive patients).5 Kawada underlined the effectiveness of colchicine in reducing mortality and the need for non-invasive ventilation, as shown by different studies.5–7 After the hospital discharge, we observed additional deaths in patients treated with colchicine and in those treated with SoC (p=0.25). The long-term analysis of our cohort confirmed the original description of improved survival for patients treated with colchicine (mortality rate at 270 days: colchicine 20% vs SoC 44%; p=0.0001). We would like to point out the statement by Kawada reporting that ferritin and arterial oxygen tension (or pressure) (PaO₂)/fractional inspired oxygen (FiO₂) level at entry were not risk factor for death in our cohort. The Cox regression analysis showed that both parameters were associated with mortality.5 Notably, the 122 patients treated with colchicine had baseline worse inflammatory features, such as higher levels of ferritin and of C reactive protein (CRP) and worse PaO₂/FiO₂,8 if compared with patients treated with SoC. Ferritin is an acute phase reactant enhanced by the IL-1beta–IL-6 axis, which is the same inflammatory pathway targeted by colchicine. Therefore, it might be speculated that ferritin could serve as a predictive biomarker for colchicine response in COVID-19. To test this hypothesis, we subgrouped patients according to the baseline serum levels of ferritin (ferritin: ≥300 ng/mL vs <300 ng/mL). We observed a significantly lower mortality rate in patients treated with colchicine, as compared with SoC only in the subgroup of patients with ferritin level ≥300 ng/mL (17% vs 40%; p<0.01; table 1). On the other hand, among the small number of patients with lower levels of ferritin, difference in mortality was not significant (11% vs 29%; p=0.39; table 1). The role of PaO₂/FiO₂ in predicting mortality in Acute Respiratory Distress Syndrome (ARDS) is well known.8 According to the PaO₂/FiO₂ classification for mild–severe ARDS (PaO₂/FiO₂: ≤300 mm Hg/%), we stratified patients in four subgroups as reported in table 1. Interestingly, the patients with moderate–severe ARDS treated with colchicine displayed the greatest benefits as compared with patients treated with SoC (table 1). Nevertheless, among patients treated with colchicine, mortality was lower in those without ARDS or with mild ARDS (PaO₂/FiO₂: >200 mm Hg/%) than in those with moderate–severe ARDS (4% vs 20%; p=0.018). Therefore, even if colchicine was effective also in the late phase of disease, these data might suggest that it should be started as soon as possible, possibly at disease onset, as already proposed by others.9 Kawada also stated that colchicine may not prevent the cytokine storm. To address this point, we focused on the alterations in proinflammatory markers, such as neutrophil count, CRP and ferritin levels, after colchicine treatment. We found a significant reduction of CRP levels after colchicine treatment and not in the SoC group.10 Interestingly, the reduction in CRP levels after 6 days of treatment significantly correlated, although weakly, with the amelioration in PaO₂/FiO₂ (r=−0.37, p=0.0009). In conclusion, all these observations suggest that colchicine may be effective also in patients with a more severe pulmonary disease in COVID-19. We have shown that its anti-inflammatory action was correlated with the improvement of alveolar capillaries exchange. In consideration of the low cost and safety of the drug, demonstrated in long-term trials for other diseases and the pandemic still being active worldwide, we strongly advise on putting efforts in the ongoing randomised clinical trials for the use of colchicine in COVID-19.11