Semi-automated preparation of fine-needle aspiration samples for rapid on-site evaluation

Fine needle aspiration (FNA) is a less invasive needle biopsy technique for sampling potential tumors in deep-seated body organs. This diagnostic procedure is used to investigate lumps and masses and involves inserting a hollow needle into the masses to sample suspicious cells and tissues for further analysis. During FNA interventions, rapid on-site evaluation (ROSE) is used for real-time assessment of the adequacy of the collected samples for further diagnostics. Sample adequacy is determined by the number of target cells used to determine tumor malignancy.

Besides reducing the number of FNA procedures and needle passes needed for appropriate samples, ROSE also improves the diagnostic yield of FNA significantly. It is worth noting that the properties of FNA samples, such as viscosity, widely vary between source organs and patients and depend on the medical procedure. They may also present clots or tissue fragments that might require further analysis and processing.

Sample preparation for ROSE is generally performed in the operating room by skilled and trained personnel such as cytopathologists. Nevertheless, most health centers are always reluctant to perform ROSE due to lack of highly trained cytopathologists or limited time, among other challenges. Thus, developing new and robust alternative methods is highly desirable. Although new methods based on artificial intelligence and telecytopathology have been investigated as potential alternatives, they require on-site sample preparation by trained personnel. Consequently, other alternative methods for on-site ROSE sample preparation, such as ThinPrep® instruments, suffer from various limitations that render them less effective.

Herein, researchers at KTH Royal Institute of Technology: Filipe Marques, Janosch Hauser, Emre Iseri, Professor Wouter van der Wijngaart and Professor Niclas Roxhed together with Dr. Igor Schliemann from Karolinska University Hospital proposed a new portable and semi-automated sample preparation device for ROSE. The device comprised a smearing tool and a capillary-drive chamber consisting of a water-soluble film to allow easy staining of the FNA samples. Finally, the applicability of the device was experimentally validated. Their research work is currently published in the journal, Lab on a Chip.

The research team showed that for both low and medium viscosity samples, the smearing tool provided consistent cell counts at relatively higher speeds above 5 cm s^‑1. The microfluidic staining sequence was consistent, taking about 46 ± 8 s to complete. The device had excellent repeatability with a maximum CV of 9% at 1.2, 2 and  cells ML^-1 and linearity with a determination coefficient . The device’s applicability for ROSE preparation was tested on different FNA model samples having different viscosity and origin, including pancreatic cells, thyroid, liver lymph nodes. It could detect cells and assess their sizes, backwound, shapes and malignancy. The device was also effective for preparing samples with tissue fragments.

Overall, the presented ROSE preparation device proved to be a robust, effective and versatile tool with great potential for disseminating ROSE and facilitating enhanced FNA yields. Furthermore, two possible modes for deploying this device were proposed. The first mode involved a medical doctor or nurse using the device to prepare ROSE and for sample imaging and analysis in the operating room. For the second mode, medical doctors or nurses would use the device to prepare and image the relevant glass slide sections and subsequently analyze the images using artificial intelligence or telecytopathology.

In summary, the design and testing of a semi-automated and portable device for performing ROSE of FNA samples was reported. The device is relatively simple to use and minimizes the number of equipment required in the operating room. Although specific training is required of the device operators, both deployment modes can help provide ROSE when cytopathologists are not readily available. In a statement to Medicine Innovates, Professor Niclas Roxhed envisioned that the proposed new device would lead to a wider implementation of ROSE with improved effectiveness and accuracy.