Results
For the 25 states that utilized IRT fixed cutoffs, the distribution shown in Figure 1 was wide with the mean value being 63.7 and the median 60 ng/ml. For 12 states that employ a 2-specimen IRT/IRT/DNA algorithm (AZ, CO, DE, ID, MD, NV, NM, OR, TX, UT, WA, and WY) the IRT cutoff question (Table 1) was restricted to the initial dried blood spot (DBS) specimen since states rely on this value for decisions about proceeding further in the NBS algorithm. Even amongst states that use a fixed IRT cutoff, there is variation in when In some states, multiple samples are obtained at different points in time or the cutoff is dependent on the age of the baby at the time of sample collection. For our analysis, we used the initial cutoff value reported by the state.
For the 26 regions that utilized IRT floating cutoffs, the range is from the 95th to the 98.8th percentile with a mean of 95.9th percentile and median at the 96th percentile. Texas uses a unique hybrid model of both floating and fixed cutoffs; the initial IRT has a floating cutoff and if it is above the 95th percentile, a fixed cutoff is applied to a second sample. For the purposes of this analysis, we considered Texas to be a state with a floating cutoff. West Virginia reported switching from a floating cutoff of the 90thpercentile to the 95th percentile in 2021 due to numerous false positives. The latter value was used for analysis in this publication.
There was tremendous variation in how states do CFTR variant analysis. Two states test for F508del only, and the other states and regions use panels that test between 23 to 365 CFTR variants. Four states (CA, NC, NY, and WI) incorporate CFTR sequencing into their algorithm. If the IRT is elevated, CFTR variant panel analysis is performed, and if 1 variant is detected CFTR sequencing is performed, either by the Sanger method (CA) or next generation sequencing (NC, NY, and WI).
Some states are clustered together, in which one state performs the analyses for one or more other states, and some states outsource theirCFTR variant analysis to an outside laboratory (PerkinElmer Genomics, Pittsburgh, PA). These are described in Table 2. Although states that utilize outsourcing all utilize the same laboratory, they each set their own IRT cutoffs and select their own CFTR variant panel.
We noted a geographic distribution of which states that utilized fixed vs. floating IRT cutoffs, as shown in Figure 3. In general, western states tended to use fixed cutoff values, especially before Colorado (with Wyoming) switched to a floating cutoff near the end of 2021. In contrast, most of the eastern states began with a floating cutoff strategy and have retained it. The midwest and south are mixed with about half using each approach.
The free text portion of the survey provided valuable insight and feedback. Some respondents sought advice on newer methodologies such as next generation sequencing. A request was also submitted by one state’s CF clinical leader for a virtual conference that would include the NBS lab leaders to discuss recent issues and clarify CFTR panel options. As a consequence of follow-up emails and telephone calls to respondents, communications ensued that provided learning experiences. In one case, a connection was made between a CF center director and NBS lab leader which stimulated a potential partnership. We also recognized from some comments that keeping up with evolving CF NBS algorithms was a challenge because CF clinical leaders may not be adequately informed when the state NBS lab changes the CF NBS algorithm. This situation is especially challenging when “border babies” are involved, i.e., when a woman residing in one state where she and her family receive healthcare has her infant delivered in an adjacent state where the NBS specimen is collected and analyzed.