Novel One-Step Nucleic Acid Amplication (OSNA) Application in Prostate Cancer. What Can We Learn from its Usage in Other Cancer Entities? A Systematic Review

Background: Lymph node (LN) status is a key prognostic factor in the decision-making process of prostate cancer (PCa) management. Sectioning and haematoxylin and eosin (H&E) staining technique remain the gold standard for the evaluation of LN metastases despite some limitations, especially low sensitivity in detecting an accurate tumour burden within the LN, as well as a subjective and time-consuming result. One-step nucleic acid amplication (OSNA) quanties mRNA copies of cytokeratin 19 (CK19) in a fast, objective, automated, and reproducible way, raising a general interest to explore its utility for lymphatic metastasis identication in different malignancies. Methods: To present the latest evidence related to the detection of LN metastases in several tumours by using OSNA compared with the conventional H&E method, a systematic review of articles published since March 2021 was conducted using PubMed, Cochrane Library, and Web of Science databases. References from primary papers and review articles were checked to obtain further potential studies. Our procedure for evaluating records identied during the literature search followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses criteria. Results: Twenty ve studies were included. LN from six different groups of tumours: breast, gastrointestinal, gynecological, lung, head and neck and prostate cancers has been assessed. OSNA was compared with post-operative formalin-xed paran-embedded tissue sections with H&E staining as the reference standard. Contingency tables were created, and concordance rate, sensitivity, specicity and predictive values were reported. Seventeen studies analysed the discordant cases using different techniques. Conclusion: OSNA method has a high diagnostic accuracy for the detection of LN metastases in several CK19 expressing tumours. Available evidence encourages its usage in PCa patients to improve LN staging and prognosis.

New methods, such as serial section analysis (slices with a thickness of 1-2 mm), immunohistochemistry (IHC), and molecular tissue analysis using Reverse Transcription-Polymerase Chain Reaction (RT-PCR) for PSA have demonstrated a higher sensitivity to identifying low tumour burden in the nodes [11]. High costs, the time required for the analysis, and some limitations to standardization have hindered their routine application, though they remain relevant in clinical research.
In 2008, an innovative biomolecular technique called One-Step Nucleic Acid Ampli cation (OSNA) was introduced in Europe to assess LN metastases. OSNA is an automated system based on reverse transcription loop-mediated isothermal ampli cation method, able to quantify copies of cytokeratin 19 (CK19) mRNA. CK19 is a marker expressed by several solid tumours of epithelial origin, but not by healthy lymphatic tissue [12]. OSNA allows a quick and accurate analysis of the tumour burden of entire LN tissue in an objective, automated, and reproducible way [13][14][15]. It has been proven useful in different cancer entities, such as breast, colorectal, gastric, endometrial, cervical, lung, and head and neck cancer, achieving a high sensitivity and speci city in the detection of LN involvement, as well as a high concordance compared to comprehensive histopathological examination, in some cases even comparable to ultra-staging [16].
OSNA was rst applied in the intraoperative analysis of sentinel lymph node (SLN) in breast cancer, introducing an objective evaluation of the nodal tissue, as well as reducing the required time and effort by the laboratory personnel. More than 10 years ago, Tsujimoto et al. [15] demonstrated the correlation between OSNA and conventional histopathological analysis of the SLN in breast cancer and de ned the cut-off values for the distinction between macrometastases, micrometastases, and unaffected tissue. Since then, more than 200 studies have been published and the application range of OSNA was extended to other cancer entities [17].
The available scienti c and clinical evidence, together with the mentioned characteristics, has introduced OSNA in current national and European clinical guidelines as an alternative technique for the determination of lymphatic involvement in breast cancer through SLN analysis [18]. Moreover, data available from studies in colorectal cancer demonstrated that OSNA is a valid technique for the detection of lymphatic involvement also in this cancer entity [19]. Hence, OSNA is now included in the recommendations for the determination of biomarkers in colorectal carcinoma [20].
Interestingly, the quantitative outcome of the OSNA assay was identi ed as useful tool to predict, during surgery, non-SLN involvement in breast and gynecological cancer, thus supporting tailoring of surgical procedure [21]. In breast and colorectal cancer, OSNA was shown to provide also prognostic information [22].
Regarding urological tumours, based on previous studies that demonstrated the expression of CK19 in PCa tissue, Winter et al. showed that the OSNA method can detect CK19 mRNA in 100% of primary PCa tumours regardless of Gleason score and even more effectively than CK19 IHC expression, suggesting the valid application of this technique in LN evaluation [23]. In a very recent study, Engels et al. [24] demonstrated that OSNA can identify nodal metastases at an equivalent or, in cases of micrometastases, better rate than enhanced histological examination in PCa patients, con rming its promising use in intraoperative decision-making in personalized LN surgery.
To set up future clinical use of OSNA in PCa, the aim of this review is to analyse the available evidence of this technique in different tumours and propose short-term course of actions to translate the validated concepts and successes from the other malignancies to PCa.

Search strategy
To retrieve all relevant papers published before the end of March 2021, three databases including PubMed, Cochrane Library, and Web of Science were searched by two independent reviewers combining the following Medical Subject Headings: onestep nucleic acid ampli cation, OSNA, lymph nodes, lymph node metastases, cytokeratin 19, CK19. References from primary papers and review articles were checked to obtain further potential studies. Our procedure for evaluating records identi ed during the literature search followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) criteria [25]. Disagreements were resolved through discussion.

Eligible criteria
We de ned study eligibility using the PICO strategy (patient population, intervention, comparison, and outcomes) [26]. A study was considered relevant to this review according to the following criteria: 1) Adult patients with con rmed cancer, eligible for surgical treatment and undergoing SLN biopsy (SLNB) or regional lymphadenectomy; 2) patients did not undergo any neoadjuvant treatment; 3) the main objective was to compare OSNA using fresh LN with postoperative standard formalinxed para n-embedded (FFPE)-H&E analysis; 4) LN were dissected and analysed using both OSNA and the standard technique at the same time; 5) the pathological examination method was fully described; 6) results were reported per node (minimum 100 nodes); 7) su cient data was available to calculate true-positive, false-positive, false-negative and truenegative values. We limited these criteria to English original studies. Review articles, meta-analysis, conference abstracts, and letters were excluded.

Study selection
The ow diagram of study selection process was outlined in Fig. 1. A total of 244 potentially relevant studies were identi ed using the searching terms described. 89 duplicated studies were initially excluded. After screening titles and abstracts, 102 papers were removed. From the remaining 52 studies, 28 were excluded after full text review because the comparison was made with intraoperative frozen section or touch imprint cytology as a reference method, less than 100 nodes were included, analysis was performed per patient, or insu cient data was available to form 2x2 tables.
Finally, 25 studies met all the requirements to be considered in the systematic review.

Quality assessment
Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) was used as an evidence-based quality assessment tool [27]. QUADAS-2 comprised four domains: patient selection, index test, reference standard, and ow and timing. The risk of bias of each study was evaluated by two independent reviewers as low "+", high "-" or unclear "?" risk.
The QUADAS-2 results summarized in Table 1 suggest a low risk of bias and a moderate to high overall quality of all 25 included studies.  [49] and thyroid cancers [50]-and 6) PCa [24]. All studies were prospectively designed.
OSNA was considered as index test and a threshold of 250 copies of CK19 mRNA per µL was xed to differentiate between negative (<250 copies/µL) and positive (≥250 copies/µL) results. OSNA was compared with post-operative FFPE tissue sections with H&E staining as the reference standard. 11 studies also included also CK19 IHC analysis in addition to H&E staining and OSNA. A LN was cut into at least two parts (depending on LN size) and divided between OSNA assay and pathology. Contingency tables were created, and concordance rate was reported. 17 studies analysed the discordant cases (OSNA + / H&E -; OSNA -/ H&E+) using different techniques.
Detailed characteristics are shown in Table 2.
Sensitivity, speci city, positive predictive value (PPV), negative predictive value (NPV), and concordance are listed in Table 3. Discordant cases are included in the reported results.   [10], as well as a subjective and time-consuming result. To overcome these limitations, OSNA assay has been developed as a fast, objective, automated, and reproducible way to examine the whole LN, raising a general interest to explore its utility for lymphatic metastases identi cation in different tumours.
OSNA gives a quantitative result of CK19 mRNA copies, which is present in several simple epithelia but is not expressed in healthy lymphatic tissue [12]. CK19 was initially proposed as a marker for the detection of LN metastases in breast cancer, where it is found in up to 98% of cases [51]. In 2007, Tsujimoto et al. [15] determined 250 copies/µl as the optimum cut-off point to de ne a positive axillary LN in breast cancer population. Nonetheless, it is known that the number of positive LN and the size of metastases are signi cant prognostic factors in most tumours. Therefore, it was also established a second cut-off point of 5,000 copies/µl to distinguish between micro and macrometastases [15]. Subsequent studies have con rmed these values and all the results re ected in this review are based on them.
In 2013 V. Peg et al. [21] de ned the concept of total tumour load (TTL) as the total CK19 mRNA copies of all positive SLNs. TTL serves as a predictive and prognosis value, providing more accurate staging than pathological ndings. Accordingly, different OSNA studies in breast SLN have set cut-off values in order to predict the axillary LN status; some of which (10,000-15,000 copies) are already included in clinical guidelines [52,53]. In 2017, Rakislova et al. [54] explored its utility to predict recurrences in colorectal carcinoma, and a recent study con rmed that a TTL ≥ 6,000 copies/µl was associated with worse disease free survival in those patients [55].
The analysis of SLN in breast cancer patients is still its main clinical application, but over the years OSNA has raised interest of the pathology community for a more accurate LN staging in other cancer entities. In the last decade, several reports comparing OSNA with histopathological examination have been published, but after a systematic review of the available literature, to date only two studies related to PCa have been found [23,24].
All the articles included in this review compare OSNA assay with postoperative H&E staining in the same LN. There is a general concordance between OSNA and standard H&E of over 85%. No full information about discordant cases is available, but we have found not only different explanations for them but also heterogeneity in its analysis. Main justi cations for the discordant cases are tumour allocation bias (TAB), and more rarely low or no tumour CK19 expression and contamination by other epithelial cells [46].
As CK19 is the single molecular marker used in OSNA assay, low tumour CK19 expression may result in a false-negative OSNA case. Different CK19 expression levels have been described for other malignancies such as colorectal (94.1%) [36], gastric (98.6%) [39], gynecological (98%) [43], lung (96%) [48], HNSCC (91.1%) [49] or PCa (100%) [23]. Moreover, certain tumour subtypes are more likely to the lack of CK19 expression, as observed in metaplastic and lobular breast carcinomas [56]. Interestingly, Goda et al. [49] performed a CK19 IHC in primary HNSCC as a rst step when analyzing discordant results, detecting no expression of CK19 in 75% of those cases. Aiming to reduce false-negative cases in breast, thyroid and lung carcinomas, Vegué et al. [30], del Carmen et al. [50], and Escalante Pérez et al. [48] veri ed the presence of CK19 in primary tumour by IHC before LN analysis. For a more accurate interpretation of the results, we encourage future researchers to include CK19 expression in primary tumours as patient inclusion criteria.
One of the main advantages of OSNA is its capacity to analyse the whole LN, leading to a more precise diagnosis than conventional histology. However, to properly compare both techniques on the same node, it is mandatory to split it, leading to a possible misdetection of metastasis by one of the methods, which is called TAB. Trying to justify the discrepancies by the need for sectioning, 17 studies have reported different strategies: second-round OSNA analysis, CK19 IHC, exhaustion of the para n-embedded SLN slices, CK19 Western blot, or CK19 qRT-PCR. Most of the metastases from the discrepancies were con rmed thanks to these strategies.
Regarding PCa, CK19 has been found not only in neoplastic tissue but also in basal and luminal cells of normal, dysplastic and benign hyperplastic tissues, although complete data concerning CK19-RNA levels is still missing [57].  [54]. In fact, Engels et al. [24] suggested that such approach could be feasible to analysed PCa LNs as a faster and economic alternative. Moreover, it could be interesting to de ne a TTL value to predict the risk of early biochemical recurrence and the need of adjuvant treatment.
When compared to histopathological examination, OSNA offers the advantage to obtain objective and quantitative data about tumour load of the whole LN in a fast and effortless way, avoiding interobserver variability. Multicentric prospective studies with large cohorts of patients may investigate the the bene ts of OSNA and its impact as a potential prognostic factor for PCa. In contrast, a potential disadvantage of examining the whole LN with OSNA is that there is no tissue left for subsequent histopathological examination following complete homogenization. Nonetheless, RNA-based molecular tests are possible using OSNA lysate, thus allowing any follow-up molecular testing. Moreover, fresh lymphatic tissue requires a thorough dissection to avoid missing nodes, which must be completely separated from fat tissue by a trained pathologist. It is also important to note that in the case of a coexisting neoplasm, OSNA cannot de ne which primary tumour the metastases come from and is unable to distinguish LN tumour cells from other benign epithelial inclusions. Therefore, it is imperative to exclude other cancers with the same lymphatic drainage and avoid contaminations.
Despite these concerns, the results reported show a high speci city, concordance rate, and NPV of OSNA assay when compared with the H&E method. Speci cally, in breast cancer patients, that high NPV provides enough evidence to become the gold standard for SLN evaluation. As slicing is required for the analysis per node, a concordance of 100% cannot be achieved due to TAB. A high concordance rate of over 85% suggests that OSNA could be an alternative technique to histopathological examination in terms of its ability to detect lymph node metastases. Future investigations may bring light to the clinical impact of OSNA in PCa as well as to its potential predictive and prognostic roles.

Conclusion
OSNA is a suitable tool to standardize LN evaluation in most CK19 expressing tumours due to the possibility to analyse the whole LN in a fast, objective, automated, and reproducible way. OSNA assay has demonstrated a high diagnostic accuracy for the detection of LN tumour burden also in PCa, but more studies are needed to con rm its validation. Moreover, intraoperative use of OSNA may be a great opportunity to set up sentinel-guided LN dissection in PCa.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and material
All the data and materials analysed are included in the main paper. Further data are available from the corresponding author on reasonable request.
Competing interests Figure 1 The ow diagram of study selection process