Authors’ reply

We express our gratitude to Denegri et al¹ for their careful analysis of our publication and for expanding the discourse on the clinical significance of the Kordian program.

The initial point raised in the letter concerns the estimation of untreated (native) low‑density lipoprotein cholesterol (LDL‑C) levels in our study population. According to the Kordian study protocol, the participants were not taking any chronic medications at the time of enrollment. As a result, the baseline cholesterol measurement taken during the primary care appointment accurately reflected the individual’s native (untreated) cholesterol level.

However, in those already treated with lipid‑lowering medications, the clinical diagnosis of familial hypercholesterolemia (FH) may indeed be more challenging. To apply the Dutch Lipid Clinic Network (DLCN) criteria and estimate the likelihood of FH in such patients, information on the native LDL‑C concentration is essential. In each case, we should aim to gather such data, for instance, by analyzing the patient’s available medical records. In the absence of such historical medical data, an alternative approach is to estimate the native LDL‑C concentration based on the LDL‑C level measured during ongoing lipid‑lowering therapy. To support this approach, dedicated calculators have been developed that estimate the untreated LDL‑C concentration from the current LDL‑C level, considering the type and intensity of the lipid‑lowering regimen at the time of measurement.² The use of such calculators inherently carries a risk of estimation error, which, based on our unpublished experience, most often results in overestimation of native LDL‑C levels. These observations are based on retrospective comparisons of calculator‑derived estimates with actual treatment efficacy data in patients with known native LDL‑C levels. This may be due to poor adherence to prescribed lipid‑lowering therapy. In such cases, the baseline LDL‑C value inferred from the reported pharmacotherapy must be overestimated. In our opinion, a key expectation associated with the expansion of digital health capabilities worldwide should be the creation of an integrated information system that consolidates data on laboratory investigations and administered treatments for each patient within a single platform, regardless of their original source. Beyond providing comprehensive and reliable clinical information, such a system would also ensure access to historical medical data, such as lipid profile measurements obtained before the initiation of pharmacotherapy.

Along with LDL‑C levels, other risk factors and chronic conditions were evaluated simultaneously, and the primary care provider (PCP) initiated appropriate treatment when needed. Experiences from the program suggest that patients with very high LDL‑C levels but no comorbidities are much more likely to have FH than those whose initial assessment points to additional untreated chronic conditions. Consequently, identifying FH in the latter group appears more challenging, and the proportion of identified mutations within this population is likely to be lower. Unfortunately, we cannot provide the number of individuals excluded from the process leading to the selection of 4018 people included in our initial Kordian cohort, as they would have been classified as a secondary cardiovascular prevention cohort. Recording this value was not included in the Kordian protocol.

As the authors of the letter correctly noted, 125 of the 378 patients evaluated at our center were ultimately referred for genetic testing. Initially, the cardiology specialists verified the original DLCN score given by the PCP and, in some cases, determined that the points had been overestimated (eg, for eyelid xanthomas). Additionally, the group not referred for genetic testing comprised individuals who during the Kordian program were diagnosed with comorbidities that, in our clinical assessment, could contribute to a secondary indrease in cholesterol levels. Consequently, their estimated probability of FH was considered low. These comorbidities included obesity (eg, with accompanying hyperlipidemia), suboptimally controlled diabetes, and chronic kidney disease, among others. Importantly, the use of pharmacotherapy itself was not a contraindication to genetic testing; the only requirement was that patients were not on lipid‑lowering therapy at the time of enrollment. If a PCP started long‑term medication treatment due to newly identified comorbidities and then referred the patient to a specialist center for further diagnostics, this approach was both recommended and fully aligned with the program’s principles and did not affect eligibility for genetic testing.

Regarding the cascade screening, this procedure is routinely implemented at our center for all individuals diagnosed with FH. After a detailed discussion of the diagnosis with each patient, we identify relatives who meet the eligibility criteria for cascade testing and are willing to undergo the procedure. Through the Kordian program, dedicated funding was secured to support genetic testing for 2 relatives of each genetically confirmed proband. In total, 87 individuals underwent cascade screening, and a pathogenic FH‑consistent variant was identified in 53 of them.

Concerning the long‑term management of patients diagnosed with FH through the Kordian program, we offered all individuals ongoing treatment at the University Hospital in Kraków. Most of them accepted this offer and remain under our care. This approach has enabled not only optimization and intensification of standard lipid‑lowering therapy³ but also systematic identification and enrollment of all eligible patients into the therapeutic proprotein convertase subtilisin / kexin type 9 modulator drug program.^4,5

In conclusion, we are grateful for the opportunity to discuss our research further and hope we have addressed all the key issues raised in the letter to the editor.