Abnormal lung CT imaging is one of the most important sequelae of COVID-19 at convalescence, especially for those with severe and critical illnesses. In the previous clinical study, we confirmed QM could effectively improve the abnormal GGOs, fiber-strip, and grid-like shadows of lung CT in convalescent COVID-19 patients. Through this study, we also found that the use of QM is directly related to the good or poor prognosis of convalescent COVID-19 patients. Therefore, those results indicate that QM represents a potential therapy with an effective and safe profile for the treatment of COVID-19 with lung injury during convalescence.
The trajectory of COVID-19 is notably marked by metabolic reprogramming, presenting unique opportunities for both therapeutic intervention and diagnostic advancements. Embracing this perspective, the integration of metabolomic and lipidomic emerges as a highly promising approach for crafting non-invasive diagnostic tools and tailoring personalized treatments for COVID-19 patients. In our research, we discovered that specific metabolites and lipid molecules – Pro Ser Ser Val, PC36:1(18:0_18:1), and BMP36:3(18:2_18:1) – exhibit a strong association with recovery in COVID-19 patients. This finding underscores the potential of these biomarkers in guiding and optimizing treatment strategies for this disease.
Known for its role in protein synthesis and structure, proline may be involved in the response to viral infections. It is a component of collagen, which is crucial for lung tissue repair and integrity(Eid et al., 2023). Given that COVID-19 often results in lung damage, proline's role in collagen formation might be significant in the healing process or in the development of fibrosis(Ergin Tuncay et al., 2022). Serine is involved in various metabolic pathways and is essential for the synthesis of certain proteins and lipids. Its role in immune function and potential antiviral properties could be relevant in the context of COVID-19, especially in modulating the body's immune response to the virus(Rahbar Saadat et al., 2021). As one of the branched-chain amino acids, valine is crucial for muscle metabolism, energy production, and immune function(Wang et al., 2023). In the context of COVID-19, its role in supporting immune function and potentially mitigating muscle wasting (which can be a complication in severe cases) could be important. The specific combination of these amino acids forming the peptide Pro Ser Ser Val might have unique properties or interactions that are relevant in the context of COVID-19, particularly in terms of recovery and the management of long-term effects like pulmonary fibrosis. Our research revealed a strong correlation between Pro Ser Ser Val and the prognosis of convalescent lung damage (pulmonary fibrosis in certain cases).
Viruses induce complex changes in the lipidome of host cells and use host metabolic resources to aggravate infection(Bley et al., 2020). Viruses can use the host's lipid membrane to mask and escape the attack of the host immune system(Izquierdo-Useros et al., 2010). The SARS-CoV-2 is no exception. For example, studies have found that the plasma lipid spectrum of COVID-19 patients has significantly changed compared with that of healthy people(Dasgupta et al., 2023). Bis(monoacylglycero)phosphate (BMP), an uncommon endolysosomal phospholipid that controls the destiny of endocytosed molecules, is required for the maintenance of lipid storage capacity and transit in the body, as well as the fine-tuning of the plasma membrane cholesterol domain(Akgoc et al., 2015). Many studies have found an association between BMPs and COVID-19. For example, a study found that the levels of BMPs in the plasma of patients with mild COVID-19 were lower than those of healthy people, and suggesting that this reduction was the result of SARS-Cov-2-infected alveolar macrophages(Song et al., 2020). In addition, BMP may interfere with SARS-CoV-2 entry and subsequent internalization by regulating the cholesterol-rich membrane domain at the plasma membrane, or it may interfere with SARS-CoV-2 replication and maturation by controlling the flow of lipids through the endocytic pathway(Luquain-Costaz et al., 2020). Meanwhile, lysosomes critically rely on BMPs to stimulate lipid catabolism, cholesterol homeostasis, and lysosomal function(Medoh et al., 2023). SARS-Cov-2, on the other hand, is capable of hijacking lysosomes to become a vehicle for self-transmission, causing the inactivation of lytic enzymes in lysosomes and disrupting the antibody presentation pathway(Ghosh et al., 2020).
Phosphatidylcholines (PCs) are important components of lung surfactants and cell membranes (Wirtz, 1991). Lung surfactant is mainly secreted by type II alveolar epithelial cells, which is an important substance to maintain lung function and prevent lung infection(Schmitz and Muller, 1991). Some studies have directly found that PCs related to viral infection or pulmonary fibrosis, such as dimyristyl phosphatidylcholine and palmitomyristyl phosphatidylcholine, were found to significantly inhibit the infecting ability of wild-type and D614G mutant SARS-CoV-2 in HEK293T-ACE2 and Vero-E6 cells(Du et al., 2022). In mice infected with respiratory syncytial virus, PCs were significantly increased(Shan et al., 2018). In a bleomycin-induced mouse model of pulmonary fibrosis, levels of 26 PCs tripled(Pascoe et al., 2021). In our study, PCs and BMPs were also found to be closely related to the prognosis of lung injury (pulmonary fibrosis in some patients) during the recovery period of COVID-19. Meanwhile, a previous study showed that BMP and PCs in the platelet lipidome of COVID-19 patients have significant changes, which is similar to our findings(Schuurman et al., 2022). Therefore, we speculate that PCs and BMP may be important metabolic markers related to the prognosis of lung injury after SARS-CoV-2 infection.
Our study represents a pioneering effort in utilizing a novel nomogram to assess chest CT image recovery in COVID-19 patients, demonstrating impressive diagnostic accuracy. However, it's important to acknowledge certain limitations in our research. Firstly, the data used were derived from a clinical retrospective analysis. Due to constraints in data availability, not all relevant clinical data were incorporated into the study. Secondly, the validation of our model was internal. For enhancing the model's future predictive accuracy, external validation is essential. Thirdly, the relatively small sample size of this study might introduce potential biases, possibly affecting the repeatability and reliability of our findings. To address these concerns, future studies should consider expanding the sample size and possibly conducting multi-center research to further validate and extend the application of our findings.
In summary, our retrospective clinical study indicates a significant correlation between the use of QM and the prognosis of lung injury during COVID-19 recovery. Additionally, the study identified potential metabolic markers, including Pro Ser Ser Val, BMPs, and PCs, which are instrumental in assessing the prognosis of lung injury in post-COVID-19 convalescence. Moreover, the integration of transcriptomic and metabolomic data into a nomogram has demonstrated robust predictive capabilities for the prognosis of COVID-19, highlighting its potential utility in clinical settings. This approach underscores the importance of comprehensive metabolic profiling in understanding and predicting the recovery trajectory of COVID-19 patients, particularly those suffering from lung injury.