Editors: Donna E. Hansel, MD, PhD, division head of pathology and laboratory medicine, MD Anderson Cancer Center, Houston; James Solomon, MD, PhD, assistant professor, Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York; Erica Reinig, MD, assistant professor and medical director of molecular diagnostics, University of Wisconsin-Madison; Marcela Riveros Angel, MD, molecular genetic pathology fellow, Department of Pathology, OHSU; Andrés G. Madrigal, MD, PhD, assistant professor, clinical, Ohio State University Wexner Medical Center, Columbus; Maedeh Mohebnasab, MD, assistant professor of pathology, University of Pittsburgh; and Alicia Dillard, MD, clinical pathology chief resident, New York-Presbyterian/Weill Cornell Medical Center.

A novel assay for Parkinson disease and other synucleinopathies

September 2023—Neurodegenerative diseases are a broad group of disorders characterized by progressive loss of nerve cells in the central or peripheral nervous system. These diseases are often chronic and incurable, with symptoms ranging from cognitive decline to motor or sensory dysfunction. There are many types of neurodegenerative diseases, with various underlying etiologies. One group of diseases, the synucleinopathies, are associated with the misfolding and aggregation of the protein α-synuclein. This group includes disease entities such as Parkinson disease, dementia with Lewy bodies, and multiple-system atrophy. It is thought that, in the pathogenesis of synucleinopathies, small α-synuclein aggregates known as fibrils recruit soluble monomers to propagate their growth in a mechanism analogous to crystallization seeding. Once large enough, the aggregates deposit in neural tissues, forming lesions, such as Lewy bodies, that are visible under the microscope. How these α-synuclein fibril seeds spread through the nervous system is unclear, but they have been detected in the cerebrospinal fluid, skin, salivary gland, and olfactory mucosa of affected patients. Therefore, it is hypothesized that the pathogenesis of synucleionopathies may involve the spread of α-synuclein fibril seeds through blood. To detect α-synuclein seeds in serum, the authors developed an immunoprecipitation-based real-time quaking-induced conversion (IP/RT-QuIC) assay and evaluated its clinical sensitivity. The assay works based on α-synuclein fibrils seeding the growth of larger aggregates. In the assay, the patient’s serum is incubated with an anti-α-synuclein antibody bound to magnetic beads, which isolates the α-synuclein seeds from the serum. The isolated seeds are then mixed with recombinant α-synuclein monomers and a fluorescent reporter dye. The mixture undergoes periodic agitation, and as the seeds recruit α-synuclein monomers and form larger fibrils and aggregates, the fluorescence signal increases. Serum with higher concentrations of α-synuclein seeds undergo more rapid fibril formation, allowing for a quantitative output. The assay was tested on sera from a cohort of 270 patients with synucleinopathy, which was compared with sera from 128 controls and 72 patients with nonsynucleinopathy neurodegenerative diseases. The assay was highly accurate for distinguishing patients with Parkinson disease from controls, with a sensitivity of 94.6 percent, specificity of 92.1 percent, and area under the receiver operating characteristic curve (AUC) of 0.96. The AUC for distinguishing patients with multiple-system atrophy from controls was 0.64, and it was 0.90 for distinguishing patients with dementia with Lewy bodies from controls. The rate at which the aggregates formed in the assay correlated with the Unified Parkinson’s Disease Rating Scale part three and disease duration. In an analysis with a blinded external cohort, the assay was able to distinguish patients with Parkinson disease and patients with multiple-system atrophy from controls with AUCs of 0.86 and 0.80, respectively. The authors demonstrated that the amplified α-synuclein seeds formed in the IP/RT-QuIC assay retained their disease-specific structural characteristics and pathogenic properties. They concluded that this study details a potential novel clinical assay for identifying patients with synucleinopathies using blood samples.

Okuzumi A, Hatano T, Matsumoto G, et al. Propagative α-synuclein seeds as serum biomarkers for synucleinopathies. Nat Med. 2023;29(6):1448–1455.

Correspondence: Dr. Nobutaka Hattori at [email protected]

Use of vorasidenib to treat IDH1- and IDH2-mutant low-grade glioma

Gliomas, the most common malignant primary brain tumors in adults, are characterized and classified by their histologic and molecular features, one of the most important of which is mutational status in IDH1 or IDH2 genes. These genes encode the enzymes isocitrate dehydrogenase 1 and 2, which are used in the citric acid cycle to catalyze the decarboxylation of isocitrate into 2-oxoglutarate. However, somatic alterations in these genes, most frequently at the arginine codon 172 of IDH1, can result in a mutant enzyme that instead produces 2-hydroxyglutarate, an oncometabolite that accumulates in tumor cells and causes aberrant downstream effects that alter the epigenetic regulation of gene expression. For patients with IDH1– or IDH2-mutant gliomas that have a higher risk of progression (World Health Organization [WHO] grade 3 or high-risk WHO grade 2), the standard-of-care postoperative treatment is a combination of chemotherapy and radiation. However, these adjuvant treatments are not curative, and they carry many long-term risks, such as neurocognitive dysfunction and DNA hypermutation. Therefore, patients with low-risk WHO grade 2 tumors are usually monitored with serial imaging. The authors evaluated the investigational drug vorasidenib, a potent oral inhibitor of the mutant IDH1 and IDH2 enzymes that can cross the blood-brain barrier. During the period of active surveillance in the postoperative setting, patients with WHO grade 2 IDH-mutant gliomas were randomized to receive vorasidenib or a placebo. A total of 331 patients from 77 sites across 10 countries were enrolled in the trial. Baseline characteristics, such as median age and performance status, were similar between the vorasidenib and control groups. Patients receiving vorasidenib had significantly longer progression-free survival compared with those in the control group (median, 27.7 versus 11.1 months, P<.001). Patients receiving vorasidenib also had significant improvement in time to the next intervention compared with those in the control group (hazard ratio, 0.26, P<.001) as 85.6 percent of patients receiving vorasidenib received no other treatments after 18 months compared with only 47.4 percent of patients in the control group. Adverse events were primarily low grade. The most common serious side effect was an elevation in alanine aminotransferase or other liver enzymes. Overall, patients with IDH-mutant low-grade glioma who were receiving vorasidenib demonstrated a significant improvement in progression-free survival and time to the next intervention. Given that the standard of care in the postoperative setting is active surveillance, this therapy represents a significant treatment opportunity for patients during this critical period.

Mellinghoff IK, van den Bent MJ, Blumenthal DT, et al. Vorasidenib in IDH1- or IDH2-mutant low-grade glioma. N Engl J Med. 2023. doi:10.1056/NEJMoa2304194

Correspondence: Dr. Ingo K. Mellinghoff at [email protected]

Editor: Frederick L. Kiechle, MD, PhD

Submit your pathology-related question for reply by appropriate medical consultants. CAP TODAY will make every effort to answer all relevant questions. However, those questions that are not of general interest may not receive a reply. For your question to be considered, you must include your name and address; this information will be omitted if your question is published in CAP TODAY. Submit a question.

Q. Some recent clinical guidelines recommend lower therapeutic and toxic limits for digoxin than those provided in assay package inserts. What therapeutic ranges and toxic thresholds should laboratories use?

A. September 2023—The CAP General Chemistry and Therapeutic Drug Monitoring Survey program includes therapeutic drugs that have been in use for decades and for which numerous commercial assays are available. The serum/plasma concentrations of some drugs, such as acetaminophen, salicylate, and vancomycin, are frequently measured to assess for toxicity or as part of therapeutic drug monitoring, while other drugs, such as digoxin, are less frequently prescribed today and, therefore, less frequently assessed.

Once a mainstay for treating atrial fibrillation and heart failure, digoxin has largely been supplanted by newer medications. Because of its infrequent clinical use, health care providers and clinical laboratories may become less familiar with therapeutic drug monitoring of digoxin. However, it retains limited indications despite being widely recognized as a potentially toxic compound. More than 3,000 participants in the Survey program report digoxin measurements, on par with the number who report acetaminophen, salicylate, and vancomycin measurements.

While early studies suggested digoxin concentrations up to 2.0–2.5 ng/mL (2.6–3.2 nmol/L) were tolerable, recognition of toxicity risk at lower concentrations has led to support for lowering the upper therapeutic limit to 0.9–1.2 ng/mL (1.2–1.5 nmol/L).¹ Recent large trials and clinical guidelines also advocate lowering digoxin therapeutic ranges. These include the ARISTOTLE trial in atrial fibrillation² and the 2022 heart failure guidelines from the American Heart Association, American College of Cardiology, and Heart Failure Society of America.³

Despite concerns about risks for toxicity, recent studies support decreasing (e.g. to 0.5 ng/mL [0.6 nmol/L]) or eliminating (e.g. therapeutic range <1.0 ng/mL [<1.3 nmol/L]) the lower limit of the therapeutic range. In contrast, previous recommendations were often ≥0.8 ng/mL (≥1.0 nmol/L). Inappropriate lower limits could lead providers who are unfamiliar with digoxin management to raise doses to address perceived subtherapeutic concentrations.

Unfortunately, guidelines and clinical trials rarely address the assays used to measure digoxin or mention such issues as lack of assay standardization, evolution of methodologies, and variable influence of interfering substances.⁴ A 2013 survey of 60 laboratories found that most had upper therapeutic limits set between 2.0 and 2.5 ng/mL (2.6 and 3.2 nmol/L), reflecting older recommendations.¹ As there is little incentive for assay manufacturers to update product information, digoxin therapeutic ranges given in package inserts might not be based on recent studies or guidelines.

All laboratories performing digoxin testing are encouraged to evaluate their test performance and patient demographics, in collaboration with cardiology and other relevant clinical practices, to determine the appropriate therapeutic range for their unique population.

1. Hauptman PJ, McCann P, Ramirez Romero JM, Mayo M. Reference laboratory values for digoxin following publication of Digitalis Investigation Group (DIG) trial data. JAMA Intern Med. 2013;173(16):1552–1554.
2. Lopes RD, Rordorf R, De Ferrari GM, et al; ARISTOTLE Committees and Investigators. Digoxin and mortality in patients with atrial fibrillation. J Am Coll Cardiol. 2018;71(10):1063–1074.
3. Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2022;79(17):e263–e421.
4. Dasgupta A. Significant improvement in digoxin immunoassays over four decades: newer assays are less affected by interferences. Ther Drug Monit. 2023;45(1):26–34.

Christine Snozek, PhD, D(ABCC)
Codirector, Clinical Chemistry and Support Services
Director, Point of Care and Central Processing
Mayo Clinic Arizona
Phoenix, Ariz.
Member, CAP Toxicology Committee

Matthew David Krasowski, MD, PhD
Clinical Professor of Pathology
University of Iowa Hospitals and Clinics
Iowa City, Iowa
Chair, CAP Toxicology Committee

Q. One of our providers noticed that two laboratories—one in New York and one in Florida—reported very different thyroid-stimulating hormone values for a patient and called our laboratory to determine which was correct. How should we handle such situations?

A. Differences among laboratories in reported thyroid-stimulating hormone (TSH) values and testing methods used on the same patient or even the same sample are not surprising. The most important factors that cause these differences are standardization and calibration bias, differences in specificity among methods, and differences in reference populations.

Standardization and calibration bias. In a CAP study, published in 2005, that used fresh frozen human serum, 17 TSH methods had mean differences of up to 0.48 mU/L at an average TSH level of 1.46 mU/L, or a difference as high as 39 percent when comparing the method with the highest values to that with the lowest values.¹ A 2010 report from the IFCC Working Group for Standardization of Thyroid Function Tests similarly showed a broad range of differences among 16 immunoassays for TSH, using a panel of 40 human samples.² In the IFCC study, the most discrepant TSH testing methods differed by an average of 39 percent, similar to the findings reported by the CAP in 2005.¹ The IFCC Working Group noted that in view of these findings, harmonization of TSH testing methods may be beneficial, particularly in light of contemporary clinical practice guidelines proposing to lower the TSH clinical decision limit.³

Despite ongoing standardization efforts, recent data from the CAP Accuracy-Based Programs (figure) demonstrate that when using pooled human serum specimens to compare testing methods, differences in standardization have been consistent and persistent.

Differences in specificity among methods. A 2013 article by Faix and Thienpont described the current state of the art and challenges in measuring TSH.⁴ In particular, the authors highlighted the role of molecular heterogeneity in TSH measurement. They reported that the pituitary gland releases a heterogeneous mixture of TSH glycoforms, an array of unique molecules with various carbohydrate side chains. This high degree of molecular heterogeneity, especially with respect to glycosylation, contributes to a broad range of measurable epitopes. Since immunoassays employ a variety of monoclonal antibodies, each type varying in its ability to detect TSH epitopes, the amount of TSH measured in a clinical sample by a given testing method depends not only on the testing method’s standardization but also on the selection of monoclonal antibodies in the manufacturer’s reagent kit. This variability can potentially lead to different TSH results across methods used on the same samples, even with well-standardized TSH testing methods. In addition to variability in specificity due to differences in selectivity for TSH epitopes, discrepant TSH values may be attributable to interference from autoantibodies or macro-TSH.⁵

Differences in reference populations. TSH reference intervals also vary among testing methods.⁶ The TSH reference interval defined by each manufacturer is sensitive to the manufacturer’s selection of individuals included in the reference population. Including certain subgroups may skew the high end of the distribution, making it non-Gaussian. Including older adults, in whom higher TSH levels are often observed, is one example. People with obesity compose another subgroup that typically has higher TSH values, unrelated to thyroid function.

Until TSH test methods are harmonized, laboratories should exercise caution when comparing results from different TSH testing methods. And when practicable, the same testing method should be used each time when monitoring a patient known to have thyroid disease.

1. Steele BW, Wang E, Klee GG, et al. Analytic bias of thyroid function tests: analysis of a College of American Pathologists fresh frozen serum pool by 3900 clinical laboratories. Arch Pathol Lab Med. 2005;129(3):310–317.
2. Thienpont LM, Van Uytfanghe K, Beastall G, et al. Report of the IFCC Working Group for Standardization of Thyroid Function Tests; part 1: thyroid-stimulating hormone. Clin Chem. 2010;56(6):902–911.
3. Demers LM, Spencer CA, eds. Laboratory Medicine Practice Guidelines: Laboratory Support for the Diagnosis and Monitoring of Thyroid Disease. National Academy of Clinical Biochemistry; 2002.
4. Faix JD, Thienpont LM. Thyroid-stimulating hormone: why efforts to harmonize testing are critical to patient care. Clinical Laboratory News. May 1, 2013. https://www.aacc.org/cln/articles/2013/may/tsh-harmonization
5. Hattori N, Ishihara T, Shimatsu A. Variability in the detection of macro TSH in different immunoassay systems. Eur J Endocrinol. 2016;174(1): 9–15.
6. Barth JH, Luvai A, Jassam N, et al. Comparison of method-related reference intervals for thyroid hormones: studies from a prospective reference population and a literature review. Ann Clin Biochem. 2018;55(1):107–112.

Neil Greenberg, PhD, DABCC
Principal Consultant
Neil Greenberg Consulting Services LLC
Rochester, NY
Member, IFCC Working Group on Commutability in Metrological Traceability
Member, CAP Accuracy-Based Programs Committee

Editors: Raymond D. Aller, MD, & Dennis Winsten

Need for speed only one factor in selecting a digital scanner

September 2023—The musician Frank Zappa said, “One size does not fit all,” a declaration that counters the claims of many clothing manufacturers and holds true for a variety of products, including, one could argue, digital scanners.

Matthew Hanna, MD, director of digital pathology informatics at Memorial Sloan Kettering Cancer Center, agrees. “There isn’t a one-size-fits-all scanner—at least that’s what we felt,” he says.

Digital scanners vary in features and functionality, and it’s important to select the one that’s right for your lab, Dr. Hanna says. For example, a pathology lab that scans glass slides before the pathologist reads them will have different speed and throughput needs than a lab that scans and archives slides that have already been read by a pathologist. Furthermore, different types of pathology slides may necessitate different imaging modes, Z-stacking capabilities, or other scanner features, he adds. Therefore, a laboratory should consider numerous factors when shopping for a digital scanner, says Dr. Hanna, who spoke on this topic at the 2023 annual meeting of the United States and Canadian Academy of Pathology. Following are some points he raised.

Imaging mode. Brightfield imaging is used for most clinical workflows, yet a digital scanner with fluorescence-based scanning modes serves multiple purposes. For example, some dermatopathologists employ fluorescence-based markers to confirm diagnoses, and clinical research often involves fluorescence imaging. Furthermore, it may be worthwhile to purchase a scanner with a darkfield fluorescence scanning mode because the fluorescence stain in glass slides tends to fade over time with exposure to light. So this type of functionality provides “an immortalized image of these darkfield slides or FISH slides, and you can pull the slides back up years later for research,” Dr. Hanna says.

Slide throughput and continuous load. Know the volume of slides your laboratory needs to scan in a set amount of time and whether or not one scanner can handle that workload, Dr. Hanna advises. “If your lab is outputting 120 slides an hour, and the scanner you looked at can scan 30 slides an hour, you need four scanners to keep up with workflow,” he says. However, laboratories should take into account what Dr. Hanna calls “dwell time,” the period of time when glass slides are being held for a courier or another step in the distribution process. If the lab produces 120 glass slides an hour during the day shift and its slides aren’t distributed until the next morning, the lab could purchase a scanner with a lower throughput, which may be more cost-effective, and scan slides during overnight hours—and it would still be prepared for the morning pickup, he says.

Laboratories should also consider the benefits of newer continuous-load scanners, Dr. Hanna says. If 15 racks of slides were being processed on an old scanner, he explains, and racks one through three were finished, the scanner would have to be stopped or paused to remove the completed racks. The scanner would then restart the process from the beginning or pick up where it left off. The newer continuous-load scanners, on the other hand, can scan without interruption as racks are loaded and offloaded.

Slide size. For some pathology cases, such as prostate resections and select sarcoma cases, labs put specimens on 2- × 3-inch whole mount slides instead of the standard 1- × 3-inch slides. Laboratories that want to use whole mount slides must not only ensure that a digital scanner can accommodate larger slide sizes but factor the larger size into their estimations of throughput and storage costs, Dr. Hanna says. A 2- × 3-inch slide will take twice as long to scan and will require twice the storage space of a typical slide.

Scan speeds. The scan speeds quoted by many digital scanner vendors are based on scanning a 15- × 15-mm image, but that standard does not accurately represent the size of a typical image, according to Dr. Hanna. In fact, an evaluation of specimen size in scanned glass slides, conducted at Memorial Sloan Kettering, found that scanned images were typically 2.8 times that size. “This is why anytime one of the vendors tells me their scan speeds, I immediately triple them,” Dr. Hanna says.

Magnification and resolution. The standard 40× magnification that pathologists are accustomed to using with microscopes equates to a resolution of 0.25 microns per pixel in a digital scanner, Dr. Hanna says. And some digital scanner vendors use 20× or 40× magnification measures in marketing material in an attempt to translate microns per pixel into terms pathologists understand—but these terms are used inconsistently across vendors, he notes. Some vendors describe a resolution of 0.25 microns per pixel as 20× magnification and others describe it as 40×. For that reason, it’s better to focus solely on the micron-per-pixel resolution metric when evaluating digital scanners, Dr. Hanna says.

Scan area selection. Some digital scanners scan a rectangular area around a specimen, including the white space in the background. Others scan multiple areas of interest, but they only capture those areas that meet a certain level of contrast to the background. Scanning areas of interest can shorten scanning times, Dr. Hanna says, but the downside is that the scanner might be more likely to miss small pieces of tissue when there is low contrast between the specimen and background.

Multilayer support. Many digital scanners scan specimens in just one focal plane, or Z-plane, but for some specimens that tend to be thicker, pathologists prefer to scan in multiple focal planes to get a more three-dimensional view, Dr. Hanna says. Digital scanners with Z-stacking capabilities offer pathologists a multi-dimensional view that is comparable to that provided by a microscope. “For cytology slides and hematology smears, a lot of pathologists like the ability to scan in multiple Z-planes because that means they are able to have at least some of the fine focus that would otherwise be lost if you were scanning in just one focal plane,” he explains.

Re-scan rates. It’s important to consider re-scan rates when evaluating a scanner’s throughput, Dr. Hanna says. If a lab needs to scan 100 slides, for example, and its scanner has a failure rate of one percent, the lab should calculate total slide throughput capacity as 101 slides. The amount of time that re-scans take can vary based on the type of scanner, he adds. Some scanners recognize when scans are not sufficiently focused and can automatically re-scan them quickly, while others require that humans intervene to diagnose problems. Laboratories should put processes in place for investigating the causes of scan failure, Dr. Hanna adds.

Whole slide image file formats. Some digital scanners have proprietary file formats, so a scanner from vendor A and a viewer from vendor B may not be compatible. While efforts are underway to develop a universal imaging standard, labs should carefully evaluate whether the file format a digital scanner uses allows that device to be used with other vendors’ products, Dr. Hanna says.

Lab space and weight load. To streamline clinical workflows by minimizing the distance glass slides need to travel to be scanned, digital scanners should be placed in the lab or as close to it as possible, Dr. Hanna says. However, he cautions, it’s important to consider the equipment’s weight when choosing a location. A heavier digital scanner may need to sit on reinforced floors, he notes.

Rack interoperability. Putting coverslips on glass slides is often the last step before sending the slides to the digital scanner. And if the rack used in the coverslipping step is not compatible with the scanner, each slide will have to be manually transferred to another rack for scanning, Dr. Hanna explains. “Ensuring interoperability between the slide racks that your lab’s coverslipper uses and the whole slide scanner is key to avoiding inefficient workflows.”

Barcode formats. Most digital scanners can read the major barcode formats, but labs affix barcodes to glass slides in numerous ways, including by printing them on stickers or etching them into the glass slides. Regardless of the format, laboratories need to ensure that barcodes are legible and located where the scanner can read them, Dr. Hanna says.

Laboratories that perform third-party consultation testing of slides may encounter glass slides with multiple barcodes—from different institutions as well as their own. Some scanners can be configured to only recognize barcodes from the institution in which the scanner is located, but if a scanner does not have that setting, the lab should ensure that its own barcode is visible and other institutions’ barcodes are obscured, he adds.

Evaluation models. Most vendors have models of their digital scanners that they will loan to prospective clients for up to three months upon request, Dr. Hanna says. Laboratories can use an evaluation model to determine what digital scanning features they need, observe how the digital scanner manages lab workflow, and assess whether the scanner is user friendly.

“We never buy scanners without evaluating them first in our lab,” Dr. Hanna says, “and I always recommend to others to ask for the evaluation unit.”

—Renee Caruthers

Tips for creating glass slides for digital scanning

• Don’t put too large of a piece of tissue on a slide. The tissue should not extend to the edge of the glass. If it does, cut it in half and make two slides.
• Place pieces of tissue close together on the slide to minimize the scan time and lower digital storage costs.
• Know the margins of the scanner’s tissue-detection area. Center the tissue sample as much as possible so that it is fully within the scanner’s view.
• Avoid wet or overhanging coverslips. Coverslips that hang over the edge of a slide can get caught, damaging the slide or even the scanner. Wet coverslips can shift inside the scanner.

—Matthew Hanna, MD

Upcoming DPA annual meeting

The Digital Pathology Association is hosting its annual meeting, Pathology Visions 2023, from Oct. 29–31, in Orlando, Fla.

Under the theme of “Igniting digital pathology transformation,” the meeting will address real-world, practical applications of digital pathology and artificial intelligence today and into the future. It offers more than 50 expert presenters, over 70 poster presentations, and nearly 50 vendors showcasing the latest innovations.

For more information, visit https://digitalpathologyassociation.org/.

Hc1 joins forces with oncology network to enhance lab staffing

The performance-analytics and operations-management technology provider Hc1 is partnering with the American Oncology Network to develop solutions that connect laboratories to data-driven staffing recommendations in real time.

The collaboration is intended to enable AON’s alliance of physicians and other health care professionals to optimize their laboratory staffing through artificial intelligence and machine-learning models using Hc1’s Workforce Optimization solution, which is under development.

“This partnership enables us to develop better processes with real-time actionable insights into our data and imparts us with forward-thinking recommendations based on a detailed analysis to optimize staffing and expenses,” said Curtiss McNair, vice president of laboratory services for AON, in a press statement.

The Workforce Optimization system is slated to be installed and tested this fall at AON’s central laboratory in Ft. Myers, Fla.

AON provides protocols for managing administrative procedures and ancillary services, including pathology, for its affiliates across 18 states.

Hc1, 317-219-4646

Duke Health and Microsoft collaborate to advance AI

Duke Health has entered a five-year-long partnership with Microsoft to responsibly and ethically harness the potential of generative artificial intelligence and cloud technology, in part by developing the Duke Health AI Innovation Lab and Center of Excellence.

Microsoft will provide Duke with state-of-the-art training to foster a cloud-savvy information technology workforce and construct a secure cloud environment to simplify and modernize IT operations. Duke will use the Microsoft Azure cloud to streamline clinical care, promote health equity, and expand research and education.

Duke Health and Microsoft will also develop AI-based solutions to fast-track innovation and use Microsoft’s Azure OpenAI Service to augment health care experiences for providers and patients through such means as automating administrative tasks to reduce workloads and expanding personalized patient education.

QuidelOrtho partners with BYG4lab to strengthen portfolio

The in vitro diagnostics technology provider QuidelOrtho has entered a software-development partnership with BYG4lab, a provider of data-management solutions for the laboratory, to enhance QuidelOrtho’s data-management offerings across its portfolio of diagnostic systems.

Through the agreement, the companies will jointly develop proprietary tools that allow autoverification to become more routine and available to labs of all sizes.

“Our partnership with BYG4lab reaches across the business, from clinical labs to transfusion medicine to point-of-care, and it allows QuidelOrtho to rapidly integrate affordable, cutting-edge and time-saving informatics solutions,” said Douglas Bryant, president and chief executive officer of QuidelOrtho, in a company press release.

The collaboration expands on an earlier commercial partnership between the companies.

QuidelOrtho, 800-874-1517

Sysmex extends long-standing alliance with Roche Diagnostics

Sysmex has expanded its 25-year-long global business partnership with Roche Diagnostics. The revised agreement renews the companies’ nonexclusive total laboratory solution collaboration that allows customers to purchase products for clinical chemistry, immunochemistry, and hematology testing from one vendor. The companies will also jointly explore ways to tackle social issues.

“Sysmex has agreed with Roche to expand the scope of their collaboration to include not only their products and sales and services but also the creation of a circular resource value chain in the in vitro diagnostics domain to deliver greater value to customers in laboratories from an eco-social perspective,” according to a press release from Sysmex.

Under the agreement, Roche will continue to distribute Sysmex’ hematology products and share management resources.

Sysmex, 847-996-4500

HNL Lab Medicine contracts with Proscia and Leica

The clinical diagnostics laboratory HNL Lab Medicine recently announced that it will use Proscia’s Concentriq Dx digital pathology software and Leica Biosystems’ high-throughput scanner hardware to establish an advanced digital pathology practice.

Allentown, Pa.-based HNL Lab Medicine is a full-service medical laboratory that operates more than 50 patient service centers in Pennsylvania.

Proscia, 215-608-5411

Dr. Aller practices clinical informatics in Southern California. He can be reached at [email protected]. Dennis Winsten is founder of Dennis Winsten & Associates, Healthcare Systems Consultants. He can be reached at [email protected].

Disruptive technologies at the point of care

September 2023—A wrist-worn high-sensitivity cardiac troponin I monitor was one of the wearable devices and health monitors highlighted in a session on emerging technologies for point-of-care testing at the Association for Diagnostics and Laboratory Medicine meeting in July.

James Nichols, PhD, D(ABCC), of Vanderbilt University Medical Center, in his talk on disruptive technologies, cited a study published this year in which a transdermal infrared spectrophotometric sensor was shown to be clinically feasible for rapid, bloodless prediction of elevated hs-cTnI levels in patients with acute coronary syndromes (Sengupta S, et al. Eur Heart J Digit Health. 2023;4[3]:145–154).

For the study, 238 hospitalized patients with ACS at five sites in India were enrolled. The final diagnosis of myocardial infarction (with or without ST elevation) and unstable angina was adjudicated using ECG, cardiac troponin testing, echocardiography, and coronary angiography. A transdermal infrared spectrophotometric sensor-derived deep-learning model was trained (three sites) and externally validated with hs-cTnI (one site) and echocardiography and angiography (two sites). Overall, an AUC-ROC of 90 percent and higher was observed.

Continuous glucose monitors were among the other technologies Dr. Nichols spoke of, including some of the questions CGM raises: Where should nursing staff enter CGM data? In the lab area of the electronic health record, much like manual POC test results? Or in nursing notes? What data is important for clinicians to trend? Is there a difference in CGM data preferences for clinicians between inpatient and outpatient use? How should accuracy and reliability be documented?

A standard for the Integration of Continuous Glucose Monitor Data into the Electronic Health Record, known as iCoDE, was released in 2022 (diabetestechnology.org/icode/).

“Wearable devices and health monitors are expanding, and we as laboratorians can’t ignore this,” said Dr. Nichols, medical director of clinical chemistry and POC testing and professor of pathology, microbiology, and immunology. “It is health data, and it may not be regulated under CLIA, but it clearly is laboratory data that’s being integrated for use in diagnosis and management of the patient. And we need to figure out how to get it into the medical record and how to keep it separate from traditional laboratory analyses.”

A third device he highlighted is an artificial intelligence application studied at Mayo Clinic, where researchers applied AI to smartwatch ECG recordings to identify patients with left ventricular dysfunction (Attia ZI, et al. Nat Med. 2022;28[12]:2497–2503). The authors concluded that consumer watch ECGs can be used to identify patients with cardiac dysfunction.

The International Federation of Clinical Chemistry and Laboratory Medicine has a committee on mobile health and bioengineering in laboratory medicine. “It is looking at the engineering, evaluation, and validation of sensors and wearables and the connectivity of that data with the patient’s electronic medical record,” Dr. Nichols said, “as well as the use of artificial intelligence algorithms.”

—Sherrie Rice

Next-generation sequencing

September 2023—I read with interest “In anatomic pathology labs, a balancing act” (August 2023). Some of the roundtable participants highlighted an area of next-generation-sequencing–based diagnostics that is a blind spot for pathologists, molecular biology tool manufacturers, and laboratory information system vendors—namely how to reduce the fractional cost of performing NGS-based analysis. On the topic of gene panels, the participants offered that the workflows are complex, reimbursement is relatively low, and startup costs are high—all true statements. However, I was struck that they evaluated the cost structure only in the setting of tissue oncology, with the implication being that the fully loaded cost of the diagnostic must be borne by the degree and level of oncology-based sequencing.

I would suggest that NGS-based diagnostics be looked at in a larger context and that the diagnostic barriers be broken down so that a larger number of NGS-based diagnostics can be evaluated in a facility. For example, noninvasive prenatal testing can be a significant number of samples for a health care system’s population. Additionally, amplicon-based sequencing as a replacement for array-based genotype scales better and more efficiently. Both of these types of diagnostics have the further benefit of being more statistical analyses rather than complex interpretive analyses, so they could be more easily incorporated into a typical health system laboratory. Limiting the cost structure of NGS-based diagnostics to only oncology diagnostics is the equivalent of basing a chemistry analyzer purchase on the number of low-volume assays while ignoring high-volume assays.

Because of the capacity of NGS instruments, diagnostic workflows are driven by the need to maximize the number of sequences performed during a sequencing run. Once a run starts, the unused capacity becomes the equivalent of a commercial airplane taking off with unsold seats—the costs and potential profits are amortized over a smaller number of participants. Functionally, maximizing sequencing usage means that almost everything in an NGS-based workflow is batched, and health systems will not be able to compete with large reference laboratories based on cost because the economics of the health system batch size won’t support it. Unless these health systems, tool manufacturers, and LIS vendors consider an expanded and less siloed view of what NGS-based diagnostics can accomplish within their client hospitals, cost justifying the infrastructure necessary to provide NGS diagnostics will continue to be an issue.

John C. Spinosa, MD, PhD
Chief Medical Advisor
Latana Consulting Group
San Diego, Calif.

Send letters to editor to [email protected].

“Igniting Digital Pathology Transformation”—the theme for Pathology Visions 2023. The Digital Pathology Association (DPA) invites you to attend this year’s annual meeting taking place October 29–31 in Orlando, Fla.

Pathology Visions brings together hundreds of pioneers and peers across the field of digital pathology to address real-world, practical applications of digital pathology and AI today, as well as the future. Attendees have access to more than 50 expert presenters, over 70 poster presentations, nearly 50 vendors showcasing the latest innovations, countless opportunities for discussion and engagement, and much more. If you are a clinician, scientist, technician, or trainee #PathVisions23 is the place for you.

“I am excited to see digital pathology fully launch and get implemented into mainstream practice. Digital pathology is for everyone because it encompasses so many things and can be successfully applied to clinical and nonclinical applications. Excitement in this field persists because there is always innovation and new creative applications. I am particularly motivated to attend the upcoming Pathology Visions meeting this year where I will get the opportunity to team up with all of you. I look forward to seeing all those worthwhile devices, emerging products, and sharing our success stories and contemporary ideas. Together we can continue to ignite this field and make a difference.” —Dr. Liron Pantanowitz, DPA president, and chair, Department of Pathology, UPMC.

Your industry is changing. Are you keeping up? For more information on the DPA and #PathVisions23, please visit: https://digitalpathologyassociation.org/

Sept. 12, 2023—Primera Technology has launched its next-generation Signature EVO slide printer and Signature EVO cassette printer. Integrated PTLab software allows users to print text, graphics, logos, and linear or 2D barcodes directly onto slides and cassettes. Both printers offer black and color printing and feature a backlit LED screen and Ethernet connectivity. The slide printer stores 100 slides and prints up to 10 slides per minute. Its SlideSeparator technology uses a puff of highly compressed air to separate slides. This mechanism, the company says, protects slides from environmental factors such as static electricity and humidity, which can cause slides to adhere to one another, resulting in misfeeds or broken slides. The cassette printer holds 40 cassettes in each of the four hoppers and is available as a standalone manual printer or as a fully automated system.

UnitedHealthcare is requiring, effective Aug. 1, that genetic test claims associated with about 250 CPT codes include a Z-Code issued by the MolDx Diagnostic Exchange (DEX) registry, operated by Palmetto GBA, a Medicare administrative contractor. To help clinical labs meet their needs for swift Z-Code registration of their genetic tests, the Dark Report and Dark Daily are presenting a free, 90-minute webinar titled “Essential Guide to Obtaining Z-Codes for Molecular and Genetic Tests” on Thursday, June 29, at 1 pm EDT.

The keynote speaker is Gabriel Bien-Willner, MD, PhD, chief medical officer of Palmetto GBA and medical director of the Molecular Diagnostic Services Program (MolDx). He will provide an up-to-the-minute picture of the DEX MolDx Z-Code registry, including the application and documentation processes needed to register a molecular test and obtain a unique Z-Code for that test.

Second to speak is Valerie Collier, MS, CGC, genetic counselor at ARUP Laboratories. She has been directly involved in helping ARUP Laboratories and its clients obtain Z-Codes for their new molecular and genetic tests. Collier’s insights and recommendations are based on four years of submitting applications to DEX MolDx, responding to the documentation requirements, and receiving Z-Codes for these new genetic tests.

Next to present is Kyle Fetter, chief operating officer of XiFin. Since the inception of the MolDx and Z-Code programs more than a decade ago, Fetter and his team have worked with client laboratories to help them register their molecular and genetic tests and obtain Z-Codes. Fetter’s team also interacts regularly with other private health insurers and is tracking their plans to follow UHC’s lead and issue their own requirements that molecular test claims include Z-Codes.

Full details on the speakers, content, and how to register for this free webinar are at https://www.darkdaily.com. XiFin provided an educational grant in support of this webinar.

For additional information, contact Amanda Curtis, 512-264-7103

Sept. 6, 2023—Paris-based Tribun Health announced the completion of its series B funding, securing an investment of €15 million. This financing round is led by Fonds Patient Autonome, managed by Bpifrance, and LBO France, along with new investors, including Owkin and Vivalto, with whom the company has also signed commercial agreements. Tribun says the capital will be used to accelerate the development and commercialization of its AI-powered digital pathology portfolio, facilitate the expansion of sales and marketing activities in Europe and North America, and serve as funding for its recent purchase of Keen Eye.

Enhance traceability in your laboratory with crisp, clear print on the surface of the cassette. Learn more.