PD‐L1 and beyond: Immuno‐oncology in cytopathology

Abstract Over the past decade, immunotherapy has emerged as one of the most promising cancer treatments. Several monoclonal antibodies targeting the programmed death 1 (PD‐1)/ programmed death ligand‐1 (PD‐L1) pathway have been integrated into standard‐of‐care treatments for a wide range of cancer types. Although all the available PD‐L1 immunohistochemistry (IHC) assays have been developed on formalin‐fixed histological specimens, a growing body of research has recently suggested the feasibility of PD‐L1 testing on cytological samples. Although promising results have been reported, several important issues still need to be addressed. Among these are pre‐analytical issues, cyto‐hystological correlation, and inter‐observer agreement. This review will briefly summarise the knowledge gaps and future directions of cytopathology in the immuno‐oncology scenario.

cancer (NSCLC), are highly challenging if not impossible to obtain.
Consequently, cytopathologists have no choice but to resort to cytological samples for both morphological characterisation and predictive testing. It is in this context that molecular cytopathology has emerged as a major player in diagnostic and predictive pathology.
Indeed, the growing popularity of molecular cytopathology stems from the fact that most molecular tests are highly versatile and can, therefore, be applied to a wide range of cytological preparations.
However, the feasibility of PD-L1 IHC evaluation on cytological specimens still warrants thorough investigation. In fact, as of today, the commercially available PD-L1 assays have never been validated on cytological samples. 18 Nonetheless, since both immunostaining and predictive testing are routinely performed in cytopathology practice, pathologists have been exploring the feasibility and reliability of assessing PD-L1 expression in cytological samples.
In this review, we will briefly summarise the knowledge gaps and future directions of cytopathology in the immuno-oncology scenario.

| PRE-ANALY TI C ISSUE S: DOE S THE SAMPLE T YPE MAT TER?
Several types of cytological samples are used in routine practice. However, being characterized by distinct pre-analytical issues, each specimen should be considered as a separate entity. In particular, the common reluctance to use cytological samples for PD-L1 evaluation primarily stems from the notion that alcohol-based fixatives might compromise IHC staining. 6,19,20 Consequently, since PD-L1 IHC procedures have been validated only on FFPE samples, formalin-fixed cell block (CB) preparations are generally recommended. However, not all CBs are processed in the same way.
Indeed, CB preparatory techniques may vary significantly depending on several factors, that is, the choice of the fluid medium used for the FNA needle rinse (formalin, saline or alcohol-based fixatives followed by formalin post-fixation), the fixation time, and the method of concentration. 19,[21][22][23][24]  Evidence that the type of fixative does not compromise PD-L1 staining is also confirmed by studies assessing the feasibility of using "traditional" non-formalin fixed cytological preparations, including direct smears or liquid-based cytology specimens (LBC) for PD-L1 IHC testing. [27][28][29][30] Indeed, although some studies have indicated that    It is worth noting that the literature has shown that the adequacy rate of cytological samples is generally higher than 80% 21,30,[32][33][34][35][36][37][38][39][40][41][42] and only occasionally lower than 70% 44 (Table 2). These data are remarkable if one considers the great difficulty of obtaining a sufficient number of tumour cells in small biopsies. 38 It is also of note that rapid on-site evaluation (ROSE) can be used to assess specimen adequacy and possibly improve CB quality in terms of tumour cellularity. 45 However, conclusive evidence regarding the use of ROSE on downstream ancillary testing outcomes is still lacking. 46 Altogether, these studies clearly indicate that cytological materials constitute a reliable source for PD-L1 evaluation in NSCLC patients.

| INTEROBS ERVER AG REEMENT
Cytopathologists should take into account interobserver variability rates before deeming cytological specimens suitable for PD-L1 as-

| G UIDELINE S
The literature has clearly established that cytology specimens It is widely known that predicting ICI therapy outcome on the basis of a single biomarker, such as PD-L1, is far from perfect. Finally, in addition to tumour cells, the tumour microenvironment (TME) and its dynamic reshaping have emerged as major players in cancer progression and treatment outcomes. The importance of this samples. 68 This new approach could break new ground in the evaluation of immunological dynamics by exploiting the ability of cytopathologists to perform serial cytological tumour sampling. 69 In conclusion, this review clearly indicates that cytological samples constitute a reliable source for PDL-1 IHC analysis (Figure 2), as evidenced by the remarkable specimen adequacy and concordance rate seen between cytological and histological specimens.
Moreover, the fact that that cytological fixatives do not compromise PD-L1 staining further attests to the utility of cytological specimens for PD-L1 testing in routine clinical practice. However, there are few challenges which still need to be addressed. In particular, training programs should be provided to ensure adequacy assessment and proper sample management, and preparation protocols must be validated and standardized across individual laboratories. Moreover, the value of dedicated expertise in PD-L1 interpretation in cytological samples cannot be underestimated.
Finally, although much of the new evidence regarding TMB and TME is still preliminary, we are confident that cytological samples will have great utility in precision immuno-oncology.

ACK N OWLED G M ENTS
We thank Paola Merolla for English language editing.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.