Search

Introduction In the selection of non-small cell lung cancer (NSCLC) patients for immunotherapy, specimen processed as cell blocks (CBs) may be the only available material to assess PD-L1 expression. Therefore, optimal CB preparation becomes paramount. In this context, here we assessed whether inadequate fixation time might be one of the pre-analytical factors affecting PD-L1 expression. Methods Ex vivo CBs from placental (n = 3) and NSCLC (n = 8) resection specimens were obtained. PD-L1 staining was performed on CBs prepared at increasing fixation times (12 hours, 48 hours, 72 hours, 96 hours, 168 hours and 504 hours) using the companion diagnostic SP263 Assay and a validated 22C3 laboratory developed test (LDT). Staining intensity and percentage of positive cells were evaluated. Results All placental CBs showed moderate to strong PD-L1 positivity in most cells, regardless of the fixation time. Likewise, the percentage of SP263-stained NSCLC cells was similar at all fixation times except for one case, which showed less intense SP263 staining at 168 hours. Conversely, in 5/8 cases, the 22C3 LDT percentage of positive cells and staining intensity decreased at 168 hours and 504 hours. Conclusions Our results show that fixation time influences the performance of 22C3 LDT on CBs. Thus, we recommend that the fixation time of cytological materials be carefully checked, especially when PD-L1 testing is delayed until the oncology request. Indeed, delays in tissue processing and paraffin embedding may lead to sub-optimal performance of PD-L1 staining on CBs.

Background Thyroid fine-needle aspirates (FNAs) with undetermined morphology can be outsourced to centralized laboratories for comprehensive molecular profiling. When a local, rapid screening rules out easily detectable BRAF and NRAS mutations outsourcing is minimized, leading to cost savings. The fully automated Idylla technology, that does not require trained staff, is an emerging option. However, Idylla platform has only been validated to process formalin fixed paraffin embedded (FFPE) sections. Here we investigate whether also the FNA needle rinse could be genotyped by the same cytopathologist who performs the FNA, a procedure that can be termed rapid on site molecular evaluation (ROME). Methods To validate this approach, the Idylla BRAF and NRAS Test was performed on the rinses from 25 simulated (bench-top) FNAs, in a first part of the study. Genotyping data were compared with those obtained on matched histological FFPE blocks. The second part of the study was carried out on 25 prospectively collected routine FNAs to assess the performance of the Idylla BRAF and NRAS assay against a gold standard real time polymerase chain reaction method. Results Idylla NRAS-BRAF Mutation Test was performed on needle rinse as well as histological FFPE blocks. A sensitivity of 88.9%, a specificity of 100.0% were obtained comparing the Idylla NRAS-BRAF Mutation Test on needle rinse to the reference method. Conclusions The FNA needle rinse can be directly genotyped. This obviates the need of cell block preparation, making possible a rapid combined morphological and molecular evaluation. Since DNA extraction is no longer necessary, the cytopathologist can perform ROME him/herself.