Over the last two decades, the introduction of tandem mass spectrometry (MS/MS) for the simultaneous analysis of acylcarnitine and amino acid species in dried blood spots (DBS) [1
] has greatly expanded the number of target conditions [2
]. Currently, the US Secretary of Health and Human Services recommends screening for a panel of 34 conditions [3
], more than half of them can be detected only by MS/MS methodology. Traditionally, the interpretation of an abnormal MS/MS result for a given analyte has been based on the comparison to cutoff values (high, low, or both) chosen by one of alternative methodologies (percentiles, multiple standard deviations added to the mean value, threshold of disease ranges) [4
]. Once a result has been flagged as potentially abnormal, a more in depth evaluation is usually performed through the calculation of ratios to determine its significance according to established protocols, and eventually it triggers referral to follow up. Although this approach has been and continues to be used extensively, it soon became obvious that false positive test results [5
] and also the risk of false negative events [6
] were too common to be resolved without improved strategies. An obvious requirement is that any sequential intervention should not require additional patient contact, and therefore the only available choice has to be more testing of the same blood spot used by the first tier method. Indeed the development of second tier tests (2TT) became a major focus of our laboratory shortly after becoming a provider of newborn screening services in 2004. It was evident that the two most common causes of false positives by MS/MS were tyrosine and propionylcarnitine and they were prioritized accordingly [8
]. Further method development led to the incorporation of succinylacetone into the first tier analysis [10
There are many disorders of propionate, cobalamin, and methionine metabolism and their differential diagnosis at the biochemical level is complex (Table 1
). Enzymatic deficiencies of propionyl-CoA carboxylase and methylmalonyl-CoA mutase are associated, respectively, with Propionic acidemia and several complementation groups of Methylmalonic acidemia that have in common an elevation of propionylcarnitine (C3) in DBS. C3 is also elevated in inherited and acquired conditions leading to deficiencies of vitamin B12 (cobalamin, Cbl), often related to maternal conditions (e.g., pernicious anemia, short gut syndrome) or dietary practices (e.g., vegetarian diet, macrobiotic diet). Inherited conditions of Cbl absorption and transport are caused by mutations in the genes encoding Cbl binding factors, trans-membrane transporters, and receptors [11
]. In addition to Homocystinuria, isolated elevation of methionine (Met) is the biochemical marker of several inherited defects of the conversion of methionine to homocysteine. High Met can also be secondary to dietary interventions (total parenteral nutrition (TPN), catabolic state, etc.), or reflect significant liver disease [12
]. Finally, low Met concentration is the first tier indicator of three disorders of remethylation [13
], and also the trigger of the same 2TT intervention [14
]. Overall, at least 20 conditions leading to abnormal concentration of C3 and/or Met in various combinations have been described (Table 1
), arguably representing the most complex differential diagnosis to be confronted in the post-analytical interpretation of newborn screening of inborn errors of metabolism by MS/MS. Although treatment strategies for these disorders differ according to the specific group or particular entity, some general principles can be applied: prompt hospitalization, low protein diet, carnitine supplementation, and additional pharmacological interventions. On the other hand, precautionary implementation of these aggressive measures in cases later resolved as false positive events can be highly traumatic for patients and their families [15
], and also quite expensive.
In addition to pioneering 2TTs for markers measured by MS/MS, our laboratory has also been leading the development and implementation of the collaborative project Region 4 Stork (R4S) in pursuit of performance improvement [16
]. The initial phase of R4S focused on clinical validation of high and low target ranges for cutoff values [5
]. Once it became apparent that such an approach was not widely adopted, the focus shifted to multivariate pattern recognition software that generates post-analytical interpretive tools. The novelty of these tools rested on replacing analyte cutoff values with the degree of overlap, calculated by comparison of data converted to Multiples of the Median (MoM), between reference and condition-specific disease ranges; and also on the provision of an interpretation based upon an objective measure of likelihood of disease [17
]. By the time R4S was sunset in September 2018, it had been deployed in 69 countries by 269 laboratories, more than 1200 users were active and data were collected from 20,938 individual cases affected with one of 94 inherited and acquired conditions. The cumulative (2011–2018) utilization of the tools exceeded 411 million calculated scores. R4S has been replaced by Collaborative Laboratory Integrated Reports (CLIR; https://clir.mayo.edu
), a second generation software with the notable addition of covariate-adjusted reference and disease ranges applicable to DBS collected up to one year age (R4S was validated only for use in the first 10 days of life), paving the way to the concept of precision newborn screening [19
], which is the achievement of a <0.01% false positive rate per test without additional patient contact and before molecular testing. In 2019 alone, CLIR tools were utilized 238 million times worldwide, more than 650,000 times per day.
After 15 years of clinical utilization, there is strong evidence that the biochemical 2TTs for disorders of propionate, methionine, and cobalamin metabolism have contributed to performance improvement of newborn screening by MS/MS, a reality validated by its replication worldwide with a combination of modifications and enhancements [27
]. Like in the case of succinylacetone [10
], it is now possible to analyze tHcy as part of the first tier screening when warranted by a high local incidence of Homocystinuria [33
]. Evidence of similar clinical utility has been observed for additional first tier MS/MS markers with poor specificity [34
], and other groups of conditions as well [36
], so the trend of developing new tests is likely to continue, starting with one new application also reported in this special issue [39
]. Furthermore, to specifically resolve the intrinsic ambiguity observed in this cohort of isolated elevation of tHcy, we are in the process of adding methionine, cysteine, and cystathionine to the 2TT panel.
2TTs should not be expected to be a complete substitute for molecular testing, rather they are complementary components of the newborn screening system. Yet, higher maturity and dependability should be achieved in the interpretation of variants beyond the too frequent classification as “uncertain significance” [40
]. To that extent, perhaps greater consideration and evidentiary weight could be given to biochemical evidence so it could become an accelerator toward the delivery of greater benefits for large number of patients. Significant financial savings, achievable by this approach and illustrated by placing individual savings as calculated above in the context of many instances occurring on a daily basis nationwide, could prevent significant health care expenditures without even accounting for the financial and emotional burden on families that are forced to cope with the prospect of their newborn being affected with a serious and chronic medical condition.
The implementation of a full menu of biochemical 2TTs by every program, perhaps with the exception of those with a number of births >150,000 per year (national programs with a single testing laboratory and a few US states) is hindered by capacity constraints and needed resources. However, a regionalization effort with individual programs, each becoming the provider of one or more 2TTs for a given geographical area, offered as an in-kind “trade” rather than a service, could lead to full coverage and efficient utilization. Alternatively, reliance on a referral laboratory either academic or commercial is a valid alternative, a model that has being gaining momentum especially among US programs in the process of adding lysosomal disorders to their mandated panels.