The Optimal Technique for RMB
Fine-needle aspiration (FNA) vs core biopsy
There are currently two methods for obtaining biopsy tissue, FNA and core biopsy. With FNA multiple passes through the tumour obtain aspirated cellular elements, and, theoretically, provide a more extensive sampling of the tumour. With core biopsy on the other hand, a limited number of solid tissue samples are obtained. These tissue samples allow for assessment of tissue architecture but could be potentially affected by sampling errors.
FNA has been shown to be inferior to core biopsy in diagnostic yield. However, adding FNA to core biopsy has been suggested to increase the diagnostic accuracy of RMB . Also, novel processing methods have been proposed to increase the diagnostic accuracy of FNA, effectively transforming it into a microbiopsy .
The panel agrees that FNA should not be performed in lieu of core biopsy. However, performing FNA in addition to core biopsies may increase the diagnostic yield. Further studies are needed to determine the role of FNA.
Most centres use 18-G needles for RMB. Several studies investigated the effect of needle size on the diagnostic yield. In a direct ex vivo comparison between 14-, 18- and 20-G needle biopsies, smaller (20 G) needles had the highest rates of insufficient tissue for diagnosis; 14- and 18-G needles offered comparable results . Similarly, Hruby et al.  assessed 11 biopsy devices with needle sizes varying between 20 and 12 G in an animal model. Their findings suggest that 20-G needles had significantly lower performance, in terms of tissue yield measured as core length, compared with their larger counterparts; the authors found no significant difference in core length among the larger calibre devices (12–18 G). Consistent with these results, another study reported higher diagnostic values (sensitivity and negative predictive value) for needles 18 G or larger, albeit this did not reach statistical significance . In addition, in a cohort of laparoscopic cryoablations, needle size (18 vs 16 G) was not found to be associated with RMB diagnostic success .
The use of a coaxial biopsy cannula has been suggested to improve biopsy yield and increase standardisation of tissue samples  and may be viewed as an aid in preventing tumour seeding. However, larger studies are needed to determine whether the use of a sheath can result in improved performance of RMB.
The panel agreed that 18 G or larger needles should be used for core biopsy of renal masses to allow for optimal diagnostic yield.
Number of cores
To date, the optimal number of RMB cores to be taken has not been defined. It has been suggested that increasing the number of cores improves the diagnostic yield  in an ex vivo investigation. A study of 88 RMBs reported 96.6% sufficient material for diagnosis when taking two or more cores ; however, in 5.7% the biopsy results were inconclusive, resulting in a total of 90.9% diagnostic yield. Similarly, Wang et al.  reported on 110 RMBs with at least two cores sampled achieving 91% diagnostic yield in percutaneous biopsy. Rybicki et al.  achieved 90.4% diagnostic yield using three to five cores with 18–22-G needles. However, the number of cores taken was not associated with biopsy success during laparoscopic cryoablation in a study by Barwari et al. . Finally, a recent study by Menogue et al.  evaluated the results of 268 RMBs for a SRM. The authors found that increasing the number of cores taken improved the diagnostic success of RMB on univariable analysis. However, on multivariable analysis, when adjusted for the amount of tissue available for diagnosis, the number of cores was no longer a significant predictor of success .
Increasing the number of cores taken extends the amount of tissue available for diagnosis. The quality of the tissue obtain is paramount for the success of RMB. The panel unanimously agreed that currently a minimum of two cores should be sampled during RMB striving for optimal quality of tissue obtained to maximise RMB diagnostic yield.
There is currently no definition of the optimal sampling pattern during RMB. Hobbs et al.  have evaluated the value of central vs peripheral sampling in an ex vivo series of RMB and concluded that while additional cores improve the diagnostic yield regardless of sampling locations, two peripheral cores consistently resulted in improved diagnostic yield. The topic of sampling patterns may be of increased importance in larger masses with higher incidence of central necrosis that may result in non-diagnostic tissue .
The panel agreed that RMB sampling should aim to avoid areas of evident necrosis. The panel also highlighted that in SRMs, attempting a peripheral sampling may increase the chances of yielding little to no diagnostic tissue due to the small size of the lesion, and thus obtaining a high quality biopsy core should be the goal rather than adherence to a specific sampling pattern.
RMB for complex cystic lesions
RMB has been proposed for the evaluation of complex cystic lesions, specifically Bosniak IIF and III [13, 14]. Harisinghani et al.  reported on a series of 28 biopsies of complex cystic lesions; the authors were able to achieve sufficient material for diagnosis in all patients and, in surgically treated cases, RMB was consistently concordant with final pathology. Lang et al.  presented the results of 199 RMB of complex cystic lesions, with a 10% rate of insufficient tissue and 2.5% of misdiagnoses. Leveridge et al.  evaluated 345 RMBs and assessed parameters associated with a successful tissue diagnosis. They found that solid-appearing lesions were far more likely (odds ratio 13.9, CI 3.8–50.7) to yield a successful RMB than their cystic counterparts on multivariable analyses.
Biopsy of complex cystic lesions entails several difficulties. First, radiographic classification of these lesions must be accurate to obtain a precise a priori estimate of the biological nature of the lesion. Second, there is an objective difficulty with targeting the biopsy passes; when dealing with SRMs with a cystic component, the solid portions (septa, nodules) are even smaller in size and thus more difficult to sample precisely. Third, core biopsy of a cyst wall may result in its rupture and spillage of the material into the surrounding tissue. Having considered these issues, the panel agrees that there may be a role for core biopsy of select complex cystic lesions. However, the indication should consider likely lower diagnostic yield and increased technical difficulty and thus the decision should be made on an individual basis.
Several reports have suggested that diagnostic yield of RMB is dependent upon the size of the tumour. Lechevallier et al.  have reported a 37% biopsy failure rate in lesions of ≤3 cm compared with 9% failures for those >3 cm. Leveridge et al.  have found a positive association between tumour size and biopsy success on multivariable analysis, estimating a three-fold increase in the likelihood of RMB success with every 1 cm increase in lesion size. However, a study of 110 RMBs in renal masses of ≤4 cm by Wang et al.  reported no difference in diagnostic yield or accuracy between lesions of <2.5 cm and larger. Similarly, Barwari et al.  could not find an association between tumour size and biopsy yield in a series of 100 RMBs during laparoscopic renal cryoablation for lesions of <4.5 cm.
The panel agrees that while no tumour size limit should be set in the indications utilising RMB for SRMs, it should be considered during clinical decision-making that biopsy success in smaller masses could potentially be lower due to targeting difficulties.
Anatomically challenging scenarios
Currently, no anatomical characteristics of renal masses or kidney anatomical variants have been considered as contraindications for biopsy. Anatomical variants (e.g. a horseshoe or pelvic kidneys, etc.) may render the biopsy procedure technically more difficult and could potentially increase the risk of complications. The panel recommends considering these technically challenging cases on an individual basis in evaluating the risk of complications vs the potential benefit of RMB.
Morbidity of RMB
In a review of the literature, Volpe et al.  summarised the complications of RMB; they found rates of 0–2% significant complications requiring active treatment or hospital admission, with bleeding requiring treatment being the most common event [15, 17]. While morbidity of RMB appears low in modern series, it is likely that the incidence of complications is underestimated due to lack of routine, prospective follow-up protocols. To date, only a study by Leveridge et al.  has assessed the entire range of potential complications, reporting a 10.4% incidence. In their study, almost all of the recorded complications were minor (Clavien grade 1), with haematoma formation accounting for the vast majority of complications. It remains unclear whether there are clinical, radiographic or intraoperative parameters that may be associated with increased risk of complications. Upon review of the literature, only isolated reports of track seeding have been found in the early experience with percutaneous biopsy  whereas in modern series tumour seeding has not been reported yet .
RMB morbidity is low and severe complications appear to be uncommon. Specifically, the risk of tumour seeding along the biopsy track is minimal with use of modern coaxial biopsy techniques, albeit the risk may be underestimated due to lack of long-term follow-up and underreporting. The panel highlights the need for more detailed reports on complications of RMB and further research into potential risk factors. The morbidity profile of RMB is favourable and would not preclude the procedure when indicated.
Typically either ultrasonography (US) or CT guidance is used for RMB; however, MRI could also be used for this purpose . There is no data to suggest superiority of one image guidance method over another . While many lesions can be identified by US, lesions of the upper pole may be obscured by aerated lung, and obesity can limit US penetrance of anterior lesions. Furthermore, a reasonable window for biopsy needle passage is not always present due to rib configuration, particularly for upper pole lesions. However, US has the advantage of visualisation of vascular structures, low cost, accessibility and avoidance of ionizing radiation . CT guidance provides excellent visualisation of extrarenal structures adjacent to the renal mass, e.g. bowel and ureter . During CT guidance, the absence of i.v. contrast often prevents visualisation of endophytic lesions. The addition of i.v. contrast during targeting allows a limited window of up to several minutes, which may not be sufficient for adequate needle positioning. Furthermore, respiratory motion can render CT guidance difficult and inaccurate in patients with poor respiratory compliance, especially in SRMs. Overall, US guidance may be preferable to CT and MRI in view of avoiding ionizing radiation and lower cost, respectively.
The panel identifies the need for further research to evaluate the different imaging guidance methods for RMB. Furthermore, the panel suggests choosing the image-guidance method based on availability, expertise and individual patient and tumour characteristics with preference for US guidance to limit radiation exposure and cost.
RMB approach (subcostal, supracostal) is largely guided by the anatomical characteristics and location of the mass. There is no data to suggest an effect of the approach on diagnostic yield of RMB. However, the supracostal approach could be associated with increased risk of pneumothorax, splenic and liver puncture, particularly with upper pole lesions in a cephalad oriented kidney, albeit clinically significant complications after RMB remain a rare occurrence. A lateral approach may also result in a higher risk of colonic, splenic or liver puncture depending on individual patient organ configuration.
The panel agrees that the choice of RMB approach should be determined by the best available strategy to obtain high quality material for tissue diagnosis given the anatomical characteristics of the mass. There is no indication of superiority of a supra- or subcostal approach. Pre-procedural imaging should be carefully reviewed to identify the location of adjacent structures that should be avoided during biopsy.
Accuracy and Reliability of RMB
One of the main concerns about RMB is the rate of non-diagnostic samples. These could be samples with insufficient tissue for diagnosis, samples representing normal kidney parenchyma or other situations whereby the cause of the renal mass cannot be described by pathological diagnosis of the biopsy. Rates of non-diagnostic samples range between 0 and 47% in the different series . Non-diagnostic samples appear to be more frequent in smaller lesions and those with cystic appearance on imaging [15, 16].
There is a need for standardisation of the nomenclature on this topic. The panel agrees that the category of non-diagnostic biopsies should include those with insufficient tissue for analysis, as well as biopsies that upon pathological analysis do not yield a diagnosis that explains the aetiology of the renal mass (i.e. normal kidney parenchyma, insufficient cellularity, necrosis etc.).
In addition, there is a consensus that tissue quality is paramount for accurate diagnosis and strategies to ensure adequacy of the sample need to be put in place. This could be done by visual confirmation, i.e. examining the tissue sample at the time of biopsy. It was felt that visual inspection of the samples harvested is a sufficient and cost-effective strategy to improve diagnostic yield whereas frozen section analyses or pathological confirmation of sample adequacy are not required.
Repeat RMB for non-diagnostic samples
Even with visual verification of tissue sample adequacy during RMB, non-diagnostic samples will still represent a clinical problem. There is evidence that repeat biopsy after an initial non-diagnostic result may achieve similar diagnostic rates as the initial procedure . Moreover, delaying active treatment for SRMs has been shown to be safe and has not compromised future treatment options or oncological control .
We recommend repeating RMB in cases with an initial non-diagnostic outcome. However, the optimal timing for repeat RMB remains to be defined. It may be reasonable to consider an alternative imaging method for repeat biopsy guidance. It is important to highlight again the importance of obtaining adequate material for diagnosis in one procedure and reducing the need for subsequent re-biopsies.
Diagnostic accuracy for malignancy
The diagnostic accuracy of RMB in differentiating malignant from benign lesions is high and ranges between 86 and 100%, with specificity approaching 100% in most series . Of note, repeat biopsy after an initial non-diagnostic one reportedly achieves similar success rates .
There is strong evidence that RMB can achieve very high accuracy in identifying malignancy. It is important to highlight that the interpretation of a biopsy sample can only be as good as the tissue sample provided and therefore the importance of an adequate and representative tissue sample cannot be overstated.
Diagnostic accuracy for histological subtyping
Histological subtyping of RCC provides additional prognostic value. The rates of accurate histological subtyping in a recent literature review were 86–98%  and inter-observer agreement is reportedly high . While in most cases correct histological subtyping can be performed on a core biopsy sample, some instances remain problematic. For example, hybrid tumours combining areas of different histological architecture represent a problem for accurate definition of the malignant subtype depending on the area(s) of the tumour captured in the biopsy sample. Similarly, at times it may be difficult to distinguish low-grade chromophobe RCC from oncocytoma even with the aid of immunohistochemical markers.
While the overall accuracy of histological subtyping is very good, the sampling nature of RMB may not adequately capture the entire morphology of a tumour, which may lead, in a small minority of cases, to inaccurate histological subtyping of RCC. The panel recommends that difficult diagnoses should be made in conjunction with an expert uropathologist.
Diagnostic accuracy for grading
Grading of RCC is in itself a debated issue with questions raised about the techniques of grading, the applicability of grading to histologies other than clear cell and prognostic significance of grade [24-27]. Having considered those issues affecting RCC grading, the reported accuracy of grading on RMB ranges between 46% and 76% whereby a simplified two-tier Fuhrman classification achieves better results (64–76%) .
The results in the literature indicate suboptimal performance of RMB for grading. The panel discussion raised several important issues to consider. First, due to the sampling nature of biopsy and non-homogeneity of grade in the tumour, identifying an area of low-grade cancer does not necessarily exclude the presence of higher grade morphology in the remainder of the lesion that may not have been sampled. This results in a high positive predictive value for high grade and lower value for low grade. These phenomena may have accounted for the findings of Harris et al.  suggesting under-grading of RCC on biopsy compared with final pathology.
Second, there is evidence of considerable inter-observer variability in grading RCC even on final pathology specimens  and this is certainly reflected in a lower accuracy of RMB grading. The Fuhrman grading system was the one predominantly used in the studies evaluating RMB accuracy and no data exist on reproducibility of other proposed systems on biopsy material, albeit in most, inter-observer agreement has been suboptimal thus far .
In summary, the panel agrees that grading of RCC on biopsy material can be performed, considering the quality of the submitted specimens and the sampling nature of the biopsy. Identification of high-grade components on biopsy carries a very high positive predictive value. Moreover, the grading accuracy may be improved by using a two-tier system that reduces inter-observer variability inherent to RCC grading. The panel does not find that at present there is indication to suggest grading systems other than those currently in use for final pathology specimens. In non-clear cell pathology, in view of current debates over the validity and prognostic significance of nuclear and nucleolar grading systems [25, 29-31], we recommend reaching a local agreement between the pathologist and the urologist on whether to and how to apply grading systems to those histologies.
Diagnosis of oncocytic lesions
While in most cases an accurate histological subtyping can be accomplished on biopsy material, in select and infrequent cases, the diagnosis can be difficult. The special case of such difficulty is the ‘oncocytic neoplasm’ that includes in the differential diagnoses: oncocytoma, a hybrid oncocytoma-chromophobe RCC lesion, low-grade chromophobe RCC, and small or tangential biopsies of papillary type 2 (eosinophilic) RCC. The diagnostic problem with oncocytic lesions is rather limited for several reasons. First, inconclusive cases by light microscopy are frequently resolved with application of additional staining techniques, such as colloidal iron and widely available immunohistochemical markers. Second, an adequate tissue sample would markedly increase the likelihood of identifying a papillary type 2 RCC. Third, hybrid tumours are typically associated with known syndrome conditions, e.g. Birt–Hogg–Dube', and their sporadic appearance is not common. Finally, the natural history of hybrid oncocytoma/chromophobe tumours, low-grade chromophobe tumours and oncocytomas is characterised by low metastatic potential , especially in SRMs .
We submit that oncocytic appearing neoplasms may be, in a minority of cases, difficult to characterise on light microscopy. Use of additional staining and immunohistochemistry reduces the diagnostic dilemma to rare occasions. Adequate tissue sampling by core biopsy can provide the expert uropathologist sufficient material for these select cases to reach an agreement in classification. Moreover, from the clinical standpoint, the differences in classification of these tumours may have but a limited impact on clinical decision-making.
Indications for RMB
Biopsy for ablative techniques
Obtaining information about the histology and grade of the lesion being treated is of paramount importance, both for devising an adequate surveillance strategy after treatment and for evaluating the outcomes of cancer control for the given management approach.
In the special case of ablative techniques, if biopsy of the treated mass was not obtained before the procedure, tissue samples can be obtained immediately before or after ablation. The theoretical advantages of biopsy after ablation include reduced potential for bleeding and seeding once the tumour has been destroyed. However, with ablative techniques there is concern that alterations in the tissue induced by ablation would render pathological interpretation challenging. In radiofrequency ablation for instance, disruption of tissue architecture and induction of coagulative necrosis and inflammatory response make histological interpretation of post-ablation biopsy difficult. Despite this, Margulis et al.  have reported that these predictable histological changes after radiofrequency ablation did not impact their ability to establish a histological diagnosis. However, the authors did not analyse the effect of post-ablation biopsy on their ability to grade the cancerous lesions. Cryoablation on the other hand, preserves tissue architecture by inducing rapid tissue freezing. The cryoablation literature suggests that biopsy before ablation may have a higher diagnostic yield and better accuracy of pathological diagnosis than after ablation . In another study, obtaining biopsy tissue after one cycle of cryoablation did not significantly alter tissue architecture allowing for accurate assessment of histotype and grade . Similar conclusions were reached in a preclinical model while obtaining tissue cores after double-freeze cycles . To date, the purported advantages of post-ablation biopsy remain controversial and are not likely to outweigh the benefits of adequate tissue sampling.
We strongly recommend performing a biopsy and obtaining a diagnosis for every patient treated with ablative approaches.
In view of the importance of obtaining sufficient and adequate tissue for diagnosis, and considering the paucity of evident advantages of post-ablation biopsy, we favour performing a biopsy before tumour ablation. However, if additional tissue is required to render a diagnosis, it is reasonable to submit immediate post-ablation biopsies.
Indications for RMB in SRMs
The steadily climbing incidence of RCC is mainly due to increased detection of localised RCC, characterised by excellent oncological outcomes after treatment . While the incidence of RCC is increasing, mortality rates fail to decline accordingly, so the question of over diagnosis and the associated overtreatment is inevitable . In addition to the increasing detection of localised RCC, it has been shown that, in a contemporary patient population, SRMs may reveal benign pathology in 36% of cases . The incidence of adverse pathological features, indicating aggressive biological potential of SRMs, is overall low and is related to the size of the lesion. Indeed, in renal masses of <2 cm that are pathologically confirmed RCC, 90% or more are low (Fuhrman 1–2) grade . These data, along with a dramatic increase in the incidence of renal masses and the recognition of limited malignant potential of a large portion of those SRMs, combined with an ever aging population accumulating medical comorbidities, emphasise that knowledge of renal mass pathology can be beneficial in devising management strategies.
AS of renal masses may be an appropriate approach for benign renal masses, and potentially for select RCCs with characteristics indicating low malignant potential. Conversely, an expedient extirpative or ablative approach would probably be favoured for those renal masses presenting pathological characteristics of potentially aggressive nature, especially in healthy individuals. The treatment decision should balance tumour characteristics with patient factors, i.e. comorbidities, that would impart an overall benefit of the proposed therapy .
A survey among urologists has shown that the use of RMB in characterisation of the renal mass remains limited at present, with only a minority of clinicians using biopsy in their practice . Cost-effectiveness models have been developed to further evaluate the potential impact of RMB. A recent study  constructed a model considering immediate surgery vs initial biopsy; even assuming that 90% of biopsied patients would undergo surgery due to biopsy results, initial sampling of the renal mass proved to be cost-effective to guide treatment decisions. Similarly, Heilbrun et al.  evaluated RMB cost-effectiveness in a three-way scenario of AS vs immediate surgery vs biopsy-guided decision-making in masses of <2 cm and concluded that the role of RMB should be expanded.
There is consensus by the panel that the role of RMB in renal mass characterisation should be expanded. The panel agrees that RMB should be recommended to patients with SRMs in consideration of treatment options, as pathological characterisation of the renal mass would allow for a more precise definition of the biological potential of the neoplasm and thus would provide important data elements to the clinician and the patient about the optimal management strategy. We think that RMB should be used in solid masses whenever imaging or clinical characteristics cannot suggest a specific categorisation of pathology (e.g. angiomyolipoma by imaging or in the setting of tuberous sclerosis complex, RCC in von Hippel–Lindau syndrome [VHL], etc.) and the management options include both conservative and invasive strategies. The panel feels that RMB should be recommended to patients considering AS as a management strategy. Conversely, when the management options are limited (e.g. conservative management is not considered or possible active treatment is not contemplated), biopsy should be avoided, as the results would not alter the management strategy. We also submit that predictive models estimating the probabilities of the mass to harbour malignant and potentially aggressive pathology could aid in focusing the use of RMB in these patients.
Routine vs selective use of RMB
The main goal of obtaining more information about the biological risk of the renal mass is to estimate the benefits and the potential harms associated with the different management options. This decision-making process has to include patient factors, e.g. comorbidities, competing risks, systems of belief and personal preferences, to be weighed against the tumour characteristics and its potential to adversely affect not only life expectancy but also quality of life for the individual.
The panel agrees that while the role of RMB should be expanded, its routine use in every patient with a renal mass should not be encouraged.
We agree that in patients with syndromes associated with specific histology of renal tumours and those with clear imaging evidence of specific pathology, renal mass biopsy can be avoided, as its contribution to the management plan is likely to be minimal.
RMB in patients with a history of malignancy
Secondary renal metastases from other primary sites are uncommon. Indeed, in patients with a known non-renal malignancy of low stage, incidentally discovered renal masses are most often represented by de novo renal tumours with pathological characteristics not dissimilar from sporadic RCC . In addition, the absence of other suspected metastatic deposits of the primary malignancy suggests a non-metastatic nature of the renal mass . In patients without evidence of progression of a non-renal malignancy, the metastatic nature of the renal mass is inordinately uncommon .
For a history of primary renal tumours, both synchronous and metachronous renal lesions have shown the potential for diverse histology in the sporadic setting [46, 47]. Another report on patients previously treated for oncocytoma has shown that the risk of metachronous renal lesions representing RCC is similar to the general population .
In view of the published evidence, the panel advocates RMB in the setting of a known, non-renal malignancy without evidence of progression. Similarly, with the exception of syndrome conditions suggesting monomorphic renal tumours (e.g. VHL, tuberous sclerosis, etc.) and clear evidence of local recurrence after RCC treatment, RMB should be indicated in patients with a prior history of RCC. In synchronous unilateral or bilateral multiple renal tumours, we do not advocate relying on the histology of one renal mass but rather recommend RMB of all visible tumours, that could be accomplished in a single or stages procedure.
RMB for candidates considering AS
AS for SRMs has gained acceptance as a viable management strategy with encouraging results .
The panel highlights that AS with the possibility of delayed treatment should not be confused with watchful waiting whereby no intention of active treatment exists. We think that biopsy in the setting of AS has multiple roles. First, RMB results could indicate suitability for AS in an individual patient. Second, biopsy results could guide towards risk-stratified surveillance schemes. For example, patients with benign lesions could embark on a less intense imaging schedule, if any, as opposed to those with malignant pathology. Finally, obtaining pathological data will enable accurate comparisons between AS series as well as detailed delineation of the natural history of the various histological entities. The panel agrees that RMB has a pivotal role in AS protocols, both in selection of candidates and in risk-stratification for surveillance strategies. Conversely, in patients on watchful waiting whereby patient characteristics preclude active treatment, biopsy should be avoided.