Comment on: Systemic lupus erythematosus patients have unique changes in serum metabolic profiles across age associated with cardiometabolic risk: reply

Dear Editor, We read with interest the correspondence by Chen et al. [1], Jury et al. [2], where several concerns were raised by the authors that we would like to address.

We acknowledge that this was a cohort study with correlation analyses which cannot infer causality between the identified metabolomic signatures and cardiovascular outcomes in SLE. However, there are significant challenges when investigating metabolites causative of cardiovascular endpoints in interventional clinical trials due to the need for long-term follow-up and large cohorts. Therefore, we only suggest that atherosclerotic outcomes could be improved by therapeutic management of metabolites and acknowledge that they are unlikely the only drivers. We agree that validation and longitudinal analyses are required to establish signature stability therapeutic impact.

With respect to our findings, Chen et al. provide examples of Mendelian randomization studies that show a lack of causal relationship of HDL on SLE [1]. While these studies are of interest to our research, they use a different metabolite analysis methodology, limited to basic lipid measures routinely taken in a clinical setting. In comparison, our study used a Nuclear Magnetic Resonance (NMR)-based platform, allowing us to identify changes in multiple small and medium-HDL subclasses across age in SLE, reporting the lipoprotein profile in more granular detail. Lipoprotein size is important when considering cardiovascular risk [3], as discussed further in our manuscript.

Chen et al. highlight the role of omega-3 supplementation in reducing disease activity in SLE and the link with nutritional intervention as therapeutic strategy [1]. We emphasize that our results are not challenging therapeutic strategies recommended by evidence-based medicine for SLE, including antimalarial drug use. We did not investigate dietary impacts in our paper; therefore, we cannot make conclusive recommendations regarding dietary interventions for cardiovascular risk management in SLE. However, there is general enthusiasm from patients towards dietary strategies for SLE management, as we have published [4, 5]. Although attractive to patients, research in other conditions highlights long-term compliance challenges with dietary restrictions; therefore, we advocate for combined treatment strategies for management of comorbidity risk in SLE across age [6]. Of note, we demonstrated that the omega-6 : omega-3 ratio was uniquely increased in SLE patients between 26–49 years of age, of whom had higher overall disease activity. Thus, as suggested, age should be a potential consideration in future randomized controlled trials of nutritional intervention.

Concerns surrounding our age grouping have been raised. With respect to different age-related research questions, our metabolomics data is open access to enable future discovery. For our cross-sectional analysis study, patients/HCs were split by age into three groups based on a compromise between sample distribution balance (to maintain statistical power) and age brackets: ≤25 years (n = 62/43), 26–49 years (n = 50/46) and ≥50 years (n = 52/31). The proposal to split the groups based on the arbitrary age cut-offs for childhood vs adult-onset SLE (<18 vs ≥18 years) or based on the ‘fertile age’ vs ‘prone to cardiovascular events’ periods (18–35 vs 35–44 years), poses additional challenges as both age and disease duration are likely to significantly influence both the metabolome and cardiovascular risk. In this respect, people younger than 18 will likely have a shorter disease duration (5–6 years, as JSLE peaks at puberty) compared with the other groups, potentially leading to skewed conclusions through unbalanced disease durations. In addition, the Framingham study (referenced by Chen et al. [1]) reported that women aged 44–50 have a 50-fold increased risk of myocardial infarction compared with controls [7], a population missed by the proposed ‘prone to cardiovascular events’ age range, justifying further our age range choice of 26–49 years.

In support of our ≥50 age cut-off (Group-3), a large study comparing age at disease onset categories of SLE patients identified a distinct late-onset SLE group (onset >50 years) [8]. In addition, we recognise the association of menopause with cardiovascular risk, and previous studies found that the mean age at menopause in SLE was 50 [9], similar to the general population. Finally, capping our cohort at 55 years, as suggested by Chen et al. [1], would have excluded 21/25% of our cohort for SLE/HC. Drawing from previous literature and the above-mentioned considerations, we support that our patient groups provided the best compromise between statistical power and biologic considerations. We also believe that we provided a balanced analysis by treating age as a linear/continuous variable in subsequent analyses to support our conclusions. We provided evidence that ApoA1 (HDL) and GlycA levels change across age in a linear fashion; thus, we expect that different age groupings would not alter our overall conclusions.

The originality and strength of our work centres around using the largest NMR metabolomic cohort study to date in SLE to investigate age-associated metabolic profiles in patients, particularly in the context of treatments, disease activity and comorbidities. We look forward to the validation of these findings in future longitudinal studies.

Data availability

No new data were analysed or generated in support of this article.

Funding

This work was supported by a Versus Arthritis Career Development Fellowship (22856), as well as grants from the NIHR UCLH Biomedical Research Centre grant BRC772/III/EJ/101350, BRC773/III/CC/101350, Lupus UK and The Rosetrees Trust (M409), and was performed within the Centre for Adolescent Rheumatology Versus Arthritis at UCL. UCLH and GOSH supported by grants from Versus Arthritis (21593 and 20164), GOSCC and the NIHR-Biomedical Research Centres at both GOSH and UCLH. The views expressed are those of the authors and not necessarily those of the NHS, NIHR or the Department of Health.

Disclosure statement: The authors have declared no conflicts of interest.