Our new lipoprotein phenotyping method (1) enables the classification of all dyslipidemias except for Type III, using standard lipid panel test results, (2) demonstrates an association between some of the lipoprotein phenotypes with ASCVD risk factors, such as diabetes and metabolic syndrome, and (3) revealed for several of the phenotypes a strong association with ASCVD events and hence could be used as risk enhancer conditions. Moreover, new lipoprotein phenotypes could potentially be automated and reported by clinical laboratories on all patients with a standard lipid panel. This may be especially valuable in identifying patients with high-risk mixed dyslipidemias, who are often missed or under-treated[36]. The overall genetic, prognostic, diagnostic and therapeutic implications for the different lipoproteins phenotypes and how this information can help guide clinical management are described below and in Table 4.
When the Fredrickson classification was first developed, one early hypothesis was that the different lipoprotein phenotypes were due to distinct monogenic mutations, but this is not the case for most phenotypes (Table 4). Even in the case of Familial Hypercholesterolemia (FH) with the Type IIa phenotype for which there are known causal monogenic mutations in the LDL-receptor gene or related genes, most patients with the Type IIa have polygenic hypercholesterolemia [37]. In addition, the lipoprotein phenotype unlike the genotype is not invariant and can change in response to drug therapy and environmental factors, such as diet and physical activity. Nevertheless, establishment of the lipoprotein phenotype may help focus the management of hyperlipidemia [38]. For example, although secondary causes, such as obesity and diabetes, are in general associated with dyslipidemias, several of the secondary causes listed in Table 4 are more specific for certain phenotypes. As described in Table 4, secondary causes should always be considered when first evaluating a patient. Secondary causes can frequently be uncovered by a careful patient history and physical exam and when addressed can often resolve the dyslipidemia.
In terms of primary causes, the more stringent criteria that we applied for identifying Type IIa and some of the other phenotypes than the Sniderman classification should enable a more targeted identification of patients with monogenic mutations. Once the more common secondary causes are excluded, genetic testing of Type IIa patients to identify specific pathogenic mutations (LDLR, APOB, LDLRAP1, PCSK9) can be considered, depending on the LDL-C level (Table 4) and can help inform family cascade screening[23]. Patients with extreme elevations of TG (Type I and V) can also have specific pathogenic mutations (e.g. LPL, APOC2, GPIHBP1, LMF1, APOA5)[39]. Type V patients, however, who typically present in adulthood with dyslipidemia, are more likely to have multifactorial chylomicronemia with one or more secondary causes, such as uncontrolled diabetes[40]. Polygenic risk scores for both hypercholesterolemia[37] and hypertriglyceridemia[41] have also been described, but their clinical utility is still being investigated. As we do in Table 4 and as recommend by others[42], the designation “Familial” is perhaps best reserved for those cases in which a primary monogenic cause or genotype has been established by DNA sequencing or biochemical testing. Otherwise, the phenotypes can either be referred to by their traditional Roman numeral designations as we did or alternatively by the more generic descriptive terms shown in Table 4 in order to avoid confusion between genotype and phenotype.
Another value to the lipoprotein phenotyping is related to their prognosis and need for referral to a medical specialist. In the cases of Type I and V, referral to a lipid specialist and nutritionist is highly recommended for preventing pancreatitis[43]. It has been shown that patients with markedly elevated TG often have poor follow-up [44]. Identifying such patients by lipoprotein phenotyping, followed by specific and appropriate treatment can reduce the incidence of acute pancreatitis [45] In case of Type I, referral to a nutritionist is important because adherence to a strict low-fat diet is challenging but often necessary as pharmacotherapy often does not add much benefit when lipoprotein lipase activity is very low or nonexistent [34]. Because of the high incidence of ASCVD in patients with Type IIa, IIb, and III phenotypes [12, 33], and their frequent need for combination pharmacotherapy[46], many of these patients should be seen by a lipid specialist, particularly if they do not show a good response to initial therapy. Type VI patients should not only be referred to a lipid specialist but also to a nutritionist for instructions on a low-fat diet and for high-dose fat soluble vitamin supplementation. They may also need additional referrals to an ophthalmologist and neurologist to monitor and or treat the other clinical manifestations of their disorder.
Lipoprotein phenotyping also has diagnostic testing implications. Because small LDL, IDL, and remnant lipoproteins, which are all considered pro-atherogenic, were found by NMR to be enriched in Type IIb, IV, and V, more in depth advance lipid testing, including the measurement of apoB, may be useful for these phenotypes (Table 4). Numerous studies have shown that when TG is elevated that LDL-C, particularly when calculated, underestimates ASCVD risk and that NonHDL-C, apoB, or LDL-P may be better suited for risk assessment[16, 17, 47, 48]. Although no single pathogenic mutation has been associated with Type IIb[36], apoB may be useful to screen Type IIb families with a strong history of ASCVD, because it has been shown to be a better risk predictor in this population than LDL-C[16]. ApoB should also be ordered in all Type VI patients to confirm their diagnosis and whenever Type III is suspected.
In case of the NH phenotype, we observed in NHANES that more than 75% of these patients had an apoB ≥ 130 mg/dL, whereas less than 3% of NM phenotype patients did so. This suggests that ordering apoB as a risk enhancer test is less likely to be informative for the NM phenotype, and its main impact in the NH phenotype may be in identifying those patients that have lower not higher risk because of a lower-than-expected apoB value.
Because the lipid cut-points utilized by our phenotyping method are largely based on current lipid guidelines, lipoprotein phenotyping can be an aid in cardiovascular risk assessment and for improving compliance with guidelines. According to 2018-Multisociety Guidelines[1], patients with an LDL-C ≥ 190 mg/dL, which is equivalent to Type IIa in the new classification system, are considered to have sufficient risk to warrant high-intensity statin treatment without requiring calculation of their 10-year ASCVD risk. Thus, the proposed lipoprotein phenotyping method can be useful by automatically identifying and highlighting these high-risk patients. A similar argument can also be made for aggressive statin treatment of all Type IIb patients. Presently, many Type IIb patients may be under-treated unless their nonHDL-C is also considered because their LDL-C is often less than 190 mg/dL[36], which occurred about half the time in our study population. Consistent with what has been previously described[36], we found in ARIC that Type IIb patients had overall an equal if not greater ASCVD risk than Type IIa [36, 48]. In addition, almost all of Type IIb patients in our study had an apoB ≥ 130 mg/dL (96%), a known ASCVD risk enhancer. Furthermore, they are also at increased risk because of their high TG (≥ 175 mg/dL), another ASCVD risk enhancer. Based on our survival curve analysis, Type V and IVb patients could also be considered high risk patients and considered as risk enhancer conditions for those at intermediate risk by the 10-year ASCVD risk calculation. One group of patients for which ASCVD risk calculation is currently deemed unnecessary because of their low risk are primary prevention patients with an LDL-C ≤ 70 mg/dL Based on our classification system, this would include all NL patients.
The classification of patients into different lipoprotein phenotypes can also help direct the choice of therapy (Table 4). For Type IIa and NH, the primary lipid abnormality is elevated LDL-C, so statins should be the drug of first choice, if lifestyle modifications prove insufficient. In regard to TG lowering for pancreatitis prevention, fibrates and high dose omega-3 fatty acid supplementation are useful [49]. These medications, however, should always be used in conjunction with management of secondary causes of hypertriglyceridemia. Lifestyle modifications should also be recommended and can often be very effective for Type V [25, 26, 50]. For many patients with Type I and some V patients with persistent marked hypertriglyceridemia or multiple episodes of pancreatitis, a very low-fat diet is often an essential part of therapy [46, 50]. Gene therapy for LPL deficiency was approved in Europe, but it is no longer available [51]. Volanesorsen, an anti-sense oligonucleotide against ApoC-III, can lower TG even in patients with Type I from LPL deficiency[52], and is approved in Europe but not in the US[53]. For Type I patients presenting with acute pancreatitis from APOC2 gene mutations, infusion of fresh frozen plasma from normal donors containing apoC-II can be considered for rapidly lowering TG [54].
In treating more moderate forms of hypertriglyceridemia seen in Type IIb, III, and IV, reduction of dietary sugar and ethanol intake, as well as weight reduction and increased physical activity, are often very effective and recommended by current guidelines[50]. If these lifestyle measures do not sufficiently lower ASCVD risk, statin therapy should also be added, and it may have a slight benefit in also lowering TG. Statin monotherapy, however, may be insufficient, especially in cases of persistent elevations of LDL-P, small, dense LDL particles or remnant particles [33]. In these cases, full-dose omega-3 fatty acid supplementation in the form of Icosapent ethyl should be considered as it has been shown in the REDUCE-IT trial to reduce ASCVD risk on top of statins in patients with mild to moderate hypertriglyceridemia[13, 55]. Fibrates or nicotinic acid could also be considered[46], but ASCVD benefit has not yet been demonstrated for these drugs in randomized clinical trials when used with statins[14, 33, 46, 56, 57]. With respect to fibrates as add-on to statins, there is only evidence from post-hoc analysis demonstrating that fenofibrates are possibly effective[58, 59]. Although it is clear from genetic studies that many of the causes of hypertriglyceridemia increase the risk for ASCVD[58, 60, 61]; it is not known at this time if lowering TG, in general, will translate into reduced ASCVD events[58, 62]. Finally, for Type VI patients with ABL and HHBL, it is critical to treat them with high dose Vitamin A and E supplementation and sometimes other fat-soluble vitamins and essential fatty acids in order to prevent blindness, neuropathy and other clinical problems[27]. A very low-fat diet can also alleviate many of the severe gastrointestinal symptoms from fat malabsorption that almost all these patients develop.