management of sporadic renal angiomyolipomas: a …...1. introduction renal angiomyolipoma (aml) is...

EUO-204; No. of Pages 16

Management of Sporadic Renal Angiomyolipomas: A SystematicReview of Available Evidence to Guide Recommendationsfrom the European Association of Urology Renal CellCarcinoma Guidelines Panel

Sergio Fernandez-Pello a,*, Milan Hora b, Teele Kuusk c, Rana Tahbaz d, Saeed Dabestani e,Yasmin Abu-Ghanem f, Laurence Albiges g, Rachel H. Giles h, Fabian Hofmann i,Markus A. Kuczyk j, Thomas B. Lamk, Lorenzo Marconi l, Axel S. Merseburgerm,Thomas Powles n, Michael Staehler o, Alessandro Volpe p, Borje Ljungberg q, Axel Bex r,s,Karim Bensalah t

aDepartment of Urology, Cabueñes University Hospital, Gijón, Spain; bDepartment of Urology, University Hospital Plzen and Faculty of Medicine in Plzen, Charles

Universtity, Czech Republic; cDepartment of Urology, Royal Free Hospital, Pond Street, London, UK; dDepartment of Urology, University Hospital Hamburg

Eppendorf, Hamburg, Germany; eDepartment of Clinical Sciences Lund, Skåne University Hospital, Lund, Sweden; fDepartment of Urology, Chaim Sheba Medical

Center, Tel-Hashomer, Ramat-Gan, Israel; gDepartment of Cancer Medicine, Gustave Roussy, Université Paris-Saclay, Villejuif, France; hDepartment of Nephrology

and Hypertension, Patient Advocate International Kidney Cancer Coalition (IKCC), University Medical Center Utrecht, Utrecht, The Netherlands; iDepartment of

Urology, Sunderby Hospital, Sunderby, Sweden; jDepartment of Urology and Urologic Oncology, Hannover Medical School, Hannover, Germany; kAcademic

Urology Unit, University of Aberdeen, Aberdeen, UK; lDepartment of Urology, Coimbra University Hospital, Coimbra, Portugal; mDepartment of Urology,

University Hospital Schleswig-Holstein, Lübeck, Germany; nThe Royal Free NHS Trust and Barts Cancer Institute, Queen Mary University of London, London, UK;oDepartment of Urology, Ludwig-Maximilians University, Munich, Germany; pDivision of Urology, Maggiore della Carità Hospital, University of Eastern Piedmont,

Novara, Italy; qDepartment of Surgical and Perioperative Sciences, Urology and Andrology, Umeå University, Umeå, Sweden; r The Royal Free London NHS

Foundation Trust and UCL Division of Surgery and Interventional Science, London, UK; sDepartment of Urology, The Netherlands Cancer Institute, Antoni van

Leeuwenhoek Hospital, Amsterdam, The Netherlands; tDepartment of Urology, University of Rennes, Rennes, France

E U R O P E A N U R O L O G Y O N C O L O G Y X X X ( 2 0 1 9 ) X X X – X X X

ava i lable at www.sc iencedirect .com

journa l homepage: euoncology.europeanurology .com

Article info

Article history:

Accepted April 11, 2019

Associate Editor:Gianluca Giannarini

Keywords:

KidneyAngiomyolipomaActive surveillanceSelective arterial embolisationNephron-sparing surgerySystematic review

Abstract

Context: Little is known about the natural history of sporadic angiomyolipomas(AMLs); there is uncertainty regarding the indications of treatment and treatmentoptions.Objective: To evaluate the indications, effectiveness, harms, and follow-up ofdifferent management modalities for sporadic AML to provide guidance for clinicalpractice.Evidence acquisition: A systematic review of the literature was undertaken, incor-porating Medline, Embase, and the Cochrane Library (from 1 January 1990 to30 June 2017), in accordance with Preferred Reporting Items for SystematicReviews and Meta-Analyses guidelines. No restriction on study design was im-posed. Patients with sporadic AML were included. The main interventions includedactive surveillance, surgery (nephron-sparing surgery and radical nephrectomy),selective arterial embolisation, and percutaneous or laparoscopic thermal ablations

cro

. De. Te4@h

(radiofrequency, mi

* Corresponding author395, Gijon 33394, SpainE-mail address: spello8

Please cite this article in press as: Fernández-Pello S, et al. ManReview of Available Evidence to Guide Recommendations fromGuidelines Panel. Eur Urol Oncol (2019), https://doi.org/10.1016/j

https://doi.org/10.1016/j.euo.2019.04.0052588-9311/© 2019 European Association of Urology. Published by Elsevier B

waves, or cryoablation). The outcomes included indications

partment of Urology, Cabuenes University Hospital, Calle de los Pradosl. +34 985 185 000; Fax: +34 985 185 028.otmail.com (S. Fernandez-Pello).

agement of Sporadic Renal Angiomyolipomas: A Systematic the European Association of Urology Renal Cell Carcinoma.euo.2019.04.005

.V. All rights reserved.

https://doi.org/10.1016/j.euo.2019.04.005

mailto:[email protected]



for active treatment, AML growth rate, AML recurrence rate, risk of bleeding, post-treatment renal function, adverse events of treatments, and modalities of follow-up. Risk of bias assessment was performed using standard Cochrane methods.Evidence synthesis: Among 2704 articles identified, 43 were eligible for inclusion(zero randomised controlled trials, nine nonrandomised comparative retrospectivestudies, and 34 single-arm case series). Most studies were retrospective anduncontrolled, and had a moderate to high risk of bias.Conclusions: In active surveillance series, spontaneous bleeding was reported in 2%of patients and active treatment was undertaken in 5%. Active surveillance is themost chosen option in 48% of the cases, followed by surgery in 31% and selectivearterial embolisation in 17% of the cases. Selective arterial embolisation appeared toreduce AML volume but required secondary treatment in 30% of the cases. Surgery(particularly nephron-sparing surgery) was the most effective treatment in terms ofrecurrence and need for secondary procedures. Thermal ablation was an infrequentoption. The association between AML size and the risk of bleeding remained unclear;as such the traditional 4-cm cut-off should not per se trigger active treatment. Inspite of the limitations and uncertainties relating to the evidence base, the findingsmay be used to guide and inform clinical practice, until more robust data emerge.Patient summary: Sporadic angiomyolipoma (AML) is a benign tumour of thekidney consisting of a mixture of blood vessels, fat, and muscle. Large tumoursmay have a risk of spontaneous bleeding. However, the size beyond which thesetumours need to be treated remains unclear. Most small AMLs can be monitoredwithout any active treatment. For those who need treatment, options includesurgical removal of the tumour or stopping its blood supply (selective embolisation).Surgery has a lower recurrence rate and lower need for a repeat surgical procedure.© 2019 European Association of Urology. Published by Elsevier B.V. All rights reserved.

E U R O P E A N U R O L O G Y O N C O L O G Y X X X ( 2 0 1 9 ) X X X – X X X2


1. Introduction

Renal angiomyolipoma (AML) is a benign mesenchymaltumour composed of fat, smooth muscle, and blood vessels.AMLs belong to the group of perivascular epithelioid celltumours [1]. Most AMLs (approximately 80%) are sporadic;in the remaining 20%, AMLs are diagnosed in patients withtuberous sclerosis (TS) [2]. There are differences regardingthe clinical presentation and management of sporadic andTS AMLs. TS AMLs occur at a younger age, can grow fast, andare often bilateral and multilocular. Sporadic AMLs occurbetween 50 and 60 yr of age, are more common in women,are frequently unilateral, and have a slow growth rate[3,4]. AMLs can be diagnosed on imaging studies whenthere is evidence of fatty tissue inside the tumour. Biopsy israrely needed. However, imaging techniques cannot differ-entiate benign fat-poor AMLs from potentially malignantepithelioid AMLs [5,6].

The main complication of AMLs is spontaneous bleedingin the retroperitoneum or into the collecting system, whichcan be life threatening [7]. The risk of bleeding is related tothe angiogenic component of the tumour that includesirregular blood vessels [7]. Little is known about the riskfactors of bleeding, but it is believed to increase withtumour size [2].

There are several options for AML management: (1)active surveillance (AS) based on regular imaging; (2)selective arterial embolisation (SAE) that can be used todevascularise AMLs, which may minimise further risk ofbleeding [8,9]; (3) surgical removal, preferably by


nephron-sparing surgery (NSS) [10], although totalnephrectomy may be necessary for large tumours [2];and (4) thermal ablations, which are less frequently used[6,7,11] but can be offered in selected patients.

There is uncertainty regarding how best to managepatients with AML, and how the different treatmentoptions compare with each other in terms of clinicaleffectiveness and harms. Moreover, very little is knownabout the natural evolution of AMLs and the risk ofbleeding. A 4-cm cut-off has traditionally been proposed totrigger active treatment. This 4-cm cut-off is neverthelessquite arbitrary and was based on the results of a literaturereview in 1986 [12]; however, some recent series suggestthat the risk of bleeding is increased only in much largertumours [13,14].

This systematic review (SR) was undertaken by theEuropean Association of Urology (EAU) Renal Cell Carcino-ma (RCC) Guidelines Panel to clarify the evidence availableon natural evolution, the risk of complications, andmanagement options of sporadic AMLs. The findings willbe used to formulate guideline recommendations regardingthe management of sporadic AML in our 2019 guidelines(Supplementary material).

2. Evidence acquisition

2.1. Search strategy and screening of studies

This review was undertaken by the EAU RCC GuidelinesPanel that includes urologists, oncologists, pathologists,



E U R O P E A N U R O L O G Y O N C O L O G Y X X X ( 2 0 1 9 ) X X X – X X X 3


radiologists, and patient representatives. The review wasperformed according to Preferred Reporting Items forSystematic Reviews and Meta-Analyses (PRISMA) guide-lines [15]. The search was conducted in accordance with theprinciples outlined in the Cochrane Handbook for System-atic Reviews of Interventions [16]. Studies were identifiedby searching electronic databases and relevant websites.Highly sensitive electronic searches were conducted toidentify published and on-going comparative studies ofAML treatment. Searches were limited to studies publishedbetween 1 January 1990 and 30 June 2017, with no languagerestriction. The search was complemented by additionalsources, including relevant SRs and reference lists ofincluded studies, which were hand searched to identifyadditional relevant studies.

The searched databases were MEDLINE (from 1946 topresent), MEDLINE In-Process (from 1946 to present),EMBASE (from 1974 to 11 May 2017), EBM Reviews—Cochrane Central Register of Controlled Trials (April 2017),and EBM Reviews—Cochrane Database of SystematicReviews (from 2005 to 10 May 2017). SRs and otherbackground information were identified by searching theCochrane Database of Systematic Reviews (the CochraneLibrary, 2015). Additionally, clinicaltrials.gov and the WorldHealth Organization (WHO) International Clinical TrialsRegistry were searched to identify on-going trials.

The search terms included the following: kidney/renalhemangiomyolipoma, and angiomyolipoma or angiomyo adj2lipoma or angio adj2 myolipoma or hemangiomyolipoma. Fulldetails of the protocol, search strategies, and websitesconsulted are described in the Supplementary material.

Two reviewers (S.F.P. and R.T.) screened all abstracts andfull-text articles independently. Abstract screening dis-agreement was resolved by discussion, and when noagreement was reached, the manuscript was considereddoubtful and subsequently included for full-text screening.In case of ambiguous studies at full-text screening, thedecision on final study inclusion was taken by reviewerswith the help of final input by one of the arbitrators (K.B.and M.H.).

2.2. Types of studies

Randomised controlled trials (RCTs), quasi-RCTs, andnonrandomised comparative studies, with a minimum of10 patients and no language restrictions, were eligible forinclusion. Studies with no comparative elements (ie, single-arm case series) with �10 patients were also included.Studies with mixed populations (ie, sporadic and TS AMLs)were included if the results were reported separately for thesporadic AML subgroup. Studies in which subgroups werenot reported separately would still be included if thesporadic AML subgroup represented �70% of the cohort.

2.3. Types of participants included

The study population consisted of patients diagnosed withsporadic AML with available follow-up (FU) after anytreatment, including surveillance. In an effort to improve


homogeneity, we considered exclusively series comprisingonly sporadic AMLs or with a minimum of 70% of sporadicAMLs.

2.4. Types of interventions included

The interventions included were the following: surgery(NSS and radical nephrectomy), SAE, percutaneous orlaparoscopic thermal ablations (radiofrequency, micro-waves, or cryoablation), and AS.

2.5. Types of outcome measures included

The outcomes were as follows:

1. Duration and frequency of AS2. Indications for active treatment3. Incidence of haemorrhage and active treatment during

AS4. Success rate of active treatments5. Incidence of retreatment after active treatment6. Renal function outcome after treatment when available7. Adverse events of treatments

2.6. Data extraction

A data extraction form was developed a priori to collectinformation on study design, participants, interventions,outcome measures, and risk of bias (RoB) or confounders.Two reviewers (S.F.P. and R.T.) independently extracted datarelated to prespecified outcomes.

2.7. Assessment of RoB and confounders

RoB in nonrandomised comparative studies was assessedusing a modified Cochrane tool. This is a pragmaticapproach from the methodological literature designed toreport RoB in nonrandomised studies [17–19]. A list of fourpotential confounders was defined a priori by the panelgroup: AML size, emergency cases, TS syndrome, and AML-related symptoms. Overall judgement regarding eachconfounder was based on whether it was measured ornot, it was balanced across groups, and any statisticaladjustment was made.

RoB in case series was assessed using a pragmatic approachbased on external validity (whether study participants wereselected consecutively, application of an a priori protocol, howattrition bias was dealt with, and whether outcomes wereproperly addressed and measured) [20,21].

2.8. Data analysis

Descriptive statistics were used to summarise baselinecharacteristics. Owing to the anticipated clinical heteroge-neity and the majority of studies being nonrandomised indesign, a narrative (qualitative) synthesis was planned. Aquantitative synthesis (ie, meta-analysis) could not beperformed because of the low level of evidence of thereported results.





3. Evidence synthesis

3.1. Quantity of evidence identified

A total of 2704 articles were identified by the literaturesearch. Of these, 186 studies were eligible for full-textscreening, and after removing duplicates, 163 full-textarticles were reviewed. Forty-three studies including2377 patients met the inclusion criteria and were evaluatedfor evidence synthesis. No RCT was included. There were nodirect comparative studies, only unbalanced comparativeresults of different treatments in the same series. Fig. 1represents the PRISMA flow diagram outlining the studyselection process.

Fig. 1 – AML systematic review PRISMA flow diagram. AML = angiomyolipoma;

for Systematic Reviews and Meta-Analyses; RCC = renal cell carcinoma; SAE = seTSC = tuberous sclerosis complex. a Small arms, risk of high heterogeneity on o


3.2. Characteristics of included studies

Synthesis of the evidence and description of the resultswere as follows:

1. One arm surveillance: n = 4 studies, 899 patients2. One arm ablation: n = 5 studies, 68 patients3. One arm surgery (surgery with or without previous SAE):

n = 13 studies, 583 patients4. One arm SAE: n = 12 studies, 321 patients5. Two arms comparative (surgery vs surveillance):

n = 3 studies, 111 patients6. Two arms comparative (surgery vs SAE): n = 4 studies,

248 patients

LAM = lymphangioleiomyomatosis; PRISMA = Preferred Reporting Itemslective arterial embolisation; Surg = surgery; Surv = active surveillance;utcomes report.





7. Three arms (surgery, surveillance, and SAE):n = 2 studies, 147 patients

The length of FU was not equally reported across studies,but when available median FU varied from 4.5 to 96 mo(37 mo on average). The vast majority of studies weremonocentric. Twelve studies exclusively contained sporadicAMLs, 12 had >90% sporadic AMLs, 12 reported 70–90% ofsporadic AMLs, whilst seven did not clearly mention theproportions but met the inclusion criteria.

3.3. RoB assessment of the included studies

Overall, the RoB across studies was moderate to high. Allstudies had a retrospective and nonrandomised design,except for two prospective studies.

Table 1 presents the RoB summary and confounders fornonrandomized comparative studies. All studies had a highrisk of selection, performance, and detection biases. The risksof attrition and reporting bias were low, and confounders werereported in the majority of studies with a trend towards biggersize, emergency situations, and AML-related symptoms.

Table 1 – Risk of bias and confounder assessment for nonrandomised

Selection bias Performance bias De

Randomsequencegeneration

Allocationconcealment

Blindingof patients

Blindingof personnel

Bou

ass

Amano (2002)[29]

De Luca (1999)[30]

Delhorme (2017)[31]

Huyghe (2009)[32]

Koo (2010)[33]

Faddegon (2001)[34]

Lee (2009)[35]

Mues (2010) [4]

Seyam (2008)[36]

AML = angiomyolipoma; SAE = selective arterial embolisation.Green indicates low risk, red indicates high risk, and yellow indicates unclear risConfounders:1. Is the size equally distributed?2. Is the emergently treated AML equally distributed?3. Is the pure sporadic AML equally distributed?4. Are the symptoms equally distributed?


Table 2 represents the RoB summary for the 33 caseseries. In general, these studies were at a high RoBregarding the selection and detection of patients, and theywere at a low RoB in terms of FU/change of treatmentreporting and selective or proper outcome measurementreporting.

3.4. Comparison of intervention results

The summary of demographic findings of all includedstudies based on different treatment modalities are out-lined in Table 3. Patients were mostly in the 4th and 5thdecades of their life, and there was a clear predominance offemale gender (ratio of 4:1). AML size was reported in aninconsistent way, and there was a trend towards choosingAS for small asymptomatic AMLs and active treatment forlarger symptomatic AMLs. Many studies relied on the 4-cmcut-off to trigger active treatment.

3.4.1. Duration and frequency of AML FU

The FU results of different treatment modalities aresummarised in Table 4.

comparative studies (n = 9)

tectionbias

Attritionbias

Reportingbias

Confounding biases 1, 2, 3, and 4

lindedtcomeessment

Incompleteoutcomereporting

Selectiveoutcomereporting

Size Emergency PuresporadicAML

Symptoms

k.



Table 2 – Risk of bias assessment for case series (n = 33)

Selection bias Detection bias Attrition bias Reporting bias

Consecutivepatients?

A prioriprotocol?

Loss ofFU accountedfor? Change of

treatment accounted for?


Outcomeappropriatelymeasured?

SurveillanceBhatt (2016) [3]

Chan (2016) [22]

Mclean (2014) [26]

Ouzaid (2014) [14]

AblativeHan (2015)[37]

Castle (2012) [11]

Guo (2009)[38]

Makki (2016)[39]

Swanson (2012)[40]

SurgeryLane (2008) [27]

Boorjian (2007)[41]

Heidenreich (2002)[42]

Minervini (2007)[43]

Golan (2017)[44]

Liu (2016)[45]

Wang (2017) [28]

Yip (2000) [25]

Qin (2017) [23]

Lin (2016)[46]

Ng (2016)[47]



Please cite this article in press as: Fernández-Pello S, et al. Management of Sporadic Renal Angiomyolipomas: A SystematicReview of Available Evidence to Guide Recommendations from the European Association of Urology Renal Cell CarcinomaGuidelines Panel. Eur Urol Oncol (2019), https://doi.org/10.1016/j.euo.2019.04.005


Table 2 (Continued )

Selection bias Detection bias Attrition bias Reporting bias

Consecutivepatients?

A prioriprotocol?

Loss ofFU accountedfor? Change of

treatment accounted for?


Outcomeappropriatelymeasured?

Msezane (2010)[48]

SAEWang (2017) [28]

Bardin (2017)[49]

Chan (2011)[50]

Chick (2010)[51]

Duan (2016)[52]

Ramon (2009) [7]

Sheth (2016)[53]

Takebayashi (2009)[54]

Urbano (2017)[55]

Tso (2005)[56]

Tillou (2010)[57]

Chatziioannou (2012) [24]

FU = follow-up; SAE = selective arterial embolisation.Green indicates low risk, red indicates high risk, and yellow indicates unclear risk.



In AS patients, the length of FU varied from 21 to 49 mo.Imaging modalities were mostly ultrasound (US) andabdominal computed tomography (CT). There was nostandardised FU protocol throughout the studies. Onebased its FU on AML size [22]. All the others relied on anannual schedule, except for the 1st year when patients werecontrolled twice.

The mean FU after surgical treatment varied from 8 to96 mo. Likewise, the imaging FU was mostly based on USand CT. The frequency was increased during the 1st year,and then patients were followed annually.

In SAE series, the mean FU varied from 15 to 85 mo. Theimaging also relied on US and CT. Two studies alternated CTand magnetic resonance imaging. The frequency was


similar to surgery groups, with imaging control every 3–6mo during the 1st year and then annually.

The ablative series had a mean FU ranging from 6 to25 mo. The initial FU was more frequent with imaging at 3 d,1 mo, 6 mo, and yearly thereafter.

3.4.2. Indications for active treatment

The summary of indications for active treatment is shown inTable 5. Indications for active treatment in the majority ofthe studies were as follows: bleeding (spontaneous ruptureor haematuria), larger size (generally �4 cm), and presenceof symptoms (most commonly pain). In AS studies, thegrowth of AML was constantly considered as a criterion foractive treatment. On the contrary, AML growth was not



Table 3 – Demographic summary of included studies

Patients Age –average Gender % Sporadic AML Size average (cm) Size range Size other reports Asymptomatic Symptomatic

Active surveillance(1 arm)

899 49.6–59 155 M/599 F 92–96.7 1.124Min NRMax NR

NR 615 <4 cm97 >4 cm

498 69

Surgery(1 arm)

583 40.75–57 135 M/448 F 85–100 6.73Min 2

Max 13.3

0.3–24.0 33 <4 cm108 >4 cm

203 199

SAE(1 arm)

321 37–59 79 M/251 F 75–100 8.98Min 6.68Max 12.7

2.0–24.4 112 176

Ablative(1 arm)

68 49.2–67 12 M/33 F 100 2.5Min 1.5Max 3.4

1.0–5.5 5 9

Active surveillance(2 arms)Surgery(2 arms)

111 46.7–56 32 M/79 F 90–97.14 6.75Min 1.5Max 10

0.3–30.0 39 <5 cm14 >5 cm7 <5 cm13 >5 cm

59 29

SAE(2 arms)Surgery(2 arms)

248 45.2–53 28 M/111 F 86–90 6.12Min 3.70Max 9.40

1.2–14.3 5.5 cm SAE 122 55

4.4 cm Surg3-arm AS3-arm SAE3-arm Surg.

147 35–59 43 M/134 F 73–75 6.78Min 1.70Max 11.0

0.3–21.0 3 cm AS9.5 cm SAE3.8 cm Surg.

89 58

AML = angiomyolipoma; F = female; M = male; NR = not reported; SAE = selective arterial embolisation; Surg. = surgery.



considered to trigger active treatment, not even in two- andthree-arm comparative series. Cancer suspicion on imagingand patient’s preference were also considered an indicationfor active treatment in the majority of the studies.

The 4-cm cut-off was considered to trigger activetreatment in 21%, 41%, and 82% of AS, surgery, and SAEpatients, respectively. In AS series, the 4-cm cut-off wasretained only in one study (187 out of 899 patients). Insurgical series, the 4-cm cut-off was taken into account insix of 13 studies and concerned 238/583 patients. In SAEseries, the 4-cm of cut-off was applied in eight of ninestudies (264/321 patients).

3.4.3. Outcomes in active treatment groups

Summary of outcomes in active treatment groups areoutlined in Table 6.

In all surgical series (except for two without treatmentmodality specifications), there was clear predominance ofNSS. Two NSS series reported the use of SAE before surgery.In the majority of the series, there was no recurrence, butthree publications reported a recurrence rate between 4%and 15%. Improvement or complete remission of symptomswas seen in all five series that reported it. Secondarytreatment after surgery was rare (5/583, 0.85%). Renalfunction measurement and reporting were inconsistent(the results are summarised in Table 6).

In SAE series, there was a predominant use of polyvinylalcohol and coil particles to perform the procedures. Therecurrence rate varied from 4% to 39%. Two series reportedneither recurrence nor tumour growth. Symptom outcomeswere reported in eight series including 273 patients andimprovement varied from 41% to 100%. Eighty-four patients(30.7%) required secondary AML treatment: 39 for unspeci-fied reasons, 18 for haemorrhage, 24 for AML growth, and


three because of symptoms. When a secondary treatmentwas required, SAE was used in 69 procedures and surgery in18 procedures. Renal function changes were reported in fourstudies with various measurements, but renal functionalteration did not seem clinically relevant (Table 6).

Ablative series comprised patients treated by micro-wave, radiofrequency, and cryotherapy. There was noreported recurrence. Resolution of symptoms was completein the single study that reported it. Secondary treatmentswere inconsistently reported, and only one patient requiredanother thermal ablation for a reason that was not specified.There was no change in renal function after treatment(Table 6).

3.4.4. Outcomes in AS groups

Findings regarding the outcomes are summarised in Table 7.In one-arm AS series, there were a total of 97 patients

(11%) with growing AMLs. AML growth was reported in twoways across studies: (1) proportion of growing AMLs(number of growing AMLs/number of total AMLs) and (2)annual growth rate (cm/yr). Global proportions of growingAMLs ranged from 3% to 20%. Two studies reported annualgrowth rates of 0.02 and 0.01 cm/yr. Other studies reportedgrowth rates according to AML size: AMLs <2 cm had agrowth rate of 0.07 cm/yr, AMLs >2– < 4 cm had a growthrate of 0.1 cm/yr, and AMLs >4 cm had a growth rate of0.92 cm/yr. Twenty patients (2.2%) had spontaneous bleed-ing or haematuria, and 51 patients (5.7%) required activetreatment (77% SAE, 19% surgery, and 4% radiofrequencyablation).

3.4.5. Adverse events in active treatment groups

Adverse events of active treatment according to the Clavien-Dindo classification are highlighted in Table 8.



Table 4 – Duration, frequency, and imaging modalities performed throughout the different studies

FU Available Frequency Modality Duration, mo ((range))

Active surv. 1 armBhatt (2016) [3] Yes Not reported Not reported 43 (14–144)Chan (2016) [22] Yes <2 cm: no FU; 2–3 cm: US 5 yr; 3–

4 cm: US 2 yr; >4 cm: consider surgeryUS 30 (8.66–51.34)

MacLean (2014) [26] Yes 12 mo US and CT 21.8 (6–85.3)Ouzaid (2014) [14] Yes 6 mo, 6 mo, yearly CT 49 (9–89)Surgery 1 armLane (2008) [27] Yes Not reported CT 40.8 (0–288)Boorjian (2007) Yes Not reported Not reported 96 (1–372)Heidenreich (2002) Yes 3 mo after surgery CT 58 (3–114)Minervini (2007) Yes 3 mo after surgery, 6 mo, 6 mo, 6 mo,

6 mo, yearlyCT/US alternative 56 (10–120)

Golan (2017) Yes Not reported Not reported 8 (1–15)Lui (2016) Yes 3 mo after surgery Not reported 23.1 (17.7–18.5)Wang (2017) [28] a Yes 3 mo, 6 mo CT 8.2 (4–20.5)Yip (2000) [25] Yes Not reported CT 26 (1–80)Qin (2017) [23] a Yes 3 mo, 3 mo, 3 mo, 3 mo, yearly CT Not reportedLin (2016) Yes 3 mo, 6 mo, 1 yr, yearly US, if abnormality CT 40 (30.5–61.5)Ng (2016) Not reported – – –

Msezane (2010) Yes Not reported Not reported 29 (SD 20)Zhang (2016) Yes Not reported Not reported 19.5SAE 1 armWang (2017) [28] Yes Not reported Not reported 35.9Bardin (2017) Yes 3 mo after SAE, yearly CT/MRI 20.5 (0.5–56)Chan (2017) Yes 3 mo after SAE, 3 mo, yearly CT 85.2 (15.6–242.4)Chick (2010) Yes Not reported CT/US 44.2 (12–116)Duan (2016) Yes 1 wk after SAE, 6 mo, 12 mo, 24 mo CT 50.2 (24–72)Ramon (2009) [7] Yes 3 mo after SAE, 3 mo, yearly CT 58 (3–148)Sheth (2016) Yes Not reported CT 54 (2–266)Takebayashi (2009) Yes 3 mo after SAE, 6 mo, 12 mo, yearly CT 26.4 (14.64–38.6)Urbano (2017) Yes 6 mo, 12 mo, 24 mo CT/MRI 36.7 (5–124)Tso (2005) Yes Not reported CT 48 (2–84)Tillou (2010) Yes Not reported Not reported 53.2 (5–101)Chatziioannou (2012) [24] Yes Not reported CT/US 15 (6–15)Ablative 1 armHan (2015) Yes 3 d, 3 d, 1 mo, 3 mo, 3 mo, 3 mo, 3 mo,

6 mo, each 6 moCT/CE-US/MRI 10 (6–36)

Castle (2012) [11] Yes 1 mo after ablation, 6 mo, 12 mo, yearly CT 21.2 (1.3–71.7)Guo (2009) Yes 1 mo after ablation, 6 mo CT/CE-US 14.6 (6–23)Makki (2016) Yes Not reported CT/MRI 25 (12–33)Swanson (2012) Yes Not reported MRI 6 (3–46)Surg. vs AS 2 armAmano (2002) Not reported – – –

De Luca (1999) Yes 2 yr US (22–164)Delhorme (2017) Yes 6–12 mo first 5 yr, then yearly CT 43 (1–205)Surg. vs SAE 2 armHuyghe (2009) Not reported – MRI –

Koo (2010) Not reported – – 64.8 (11–97)Faddegon (2001) Not reported 6 mo after surgery – –

Lee (2009) Not reported – – 21.53 (SD 22.58)Surg., AS, SAE 3 armMues (2010) [4] Yes 12 mo CT/MRI 54.8(0.2–211.7)Seyam Yes – – 39.3 (33.9–44.7)

AS = active surveillance; CT = computed tomography; CE-US = contrast-enhanced ultrasound; FU = follow-up; MRI = magnetic resonance imaging;SAE = selective arterial embolisation; SD = standard deviation; Surg. = surgery; surv. = surveillance; US = ultrasound.a Includes an SAE group in addition to surgery.



In surgical series, there were no Clavien 4–5 complica-tions. Minor complications occurred in 19% of the cases. Thehospital stay was inconsistently reported, but on averagevaried between 2 and 9 d.

In SAE series, there was only one grade 5 complica-tion. Minor complications occurred in 179 (56%)patients. The majority of the complications were SAEsyndromes. Sixteen patients required surgery or


secondary embolisation because of haemorrhage. Hos-pital stay was inconsistently reported and variedbetween 1.5 and 8.5 d.

No major complication was reported in ablative series(68 patients). Grade 1, 2, and 3 minor complicationsoccurred in 12 (17.64%), three (4.41%), and one (1.47%)patients, respectively. There were no reported data regard-ing the length of hospital stay.



Table 5 – Indication analysis for active treatment and size cut-off

Intervention group General indications for active treatment Size cut-off Size cut-off (cm)

Active surveillance Size growth: yes. 1 study 4 cm: 1 study (187/899)(n = 4 studies; 899 patients) Bleeding: spontaneous rupture or haematuria (187/899 pts)

Size: >4 cmLocation: noPain: yesCancer suspicion on imaging: yesPatient preference: yes

Surgery Size growth: no 9 Studies 4 cm: 6 studies (238/583)(n = 13 studies; 583 patients) Bleeding: spontaneous rupture or haematuria (325/583 pts) 5 cm: 1 study (28/583)

Size: >4, >5, >7, >10 cm 7 cm: 1 study (36/583)Location: no 10 cm: 1 study (23/538)Pain: yesCancer suspicion on imaging: yesPatient preference: yes

SAE Size growth: no 9 Studies 4 cm: 8 studies (264/321)(n = 12 studies, 321 patients) Bleeding: spontaneous rupture or haematuria (274/321 pts) 8 cm: 1 study (10/321)

Size: >4, >8 cmLocation: noPain: noCancer suspicion on imaging: noPatient preference: no

Ablative therapies Size growth: not reported 0 studies –

(n = 5 studies, 68 patients) Bleeding: not reportedSize: not reportedLocation: not reportedPain: yes (9 patients)Cancer suspicion on imaging: not reportedPatient preference: not reported

Surgery vs AS Size growth: no 1 study (23/111 pts) <2, <4 cm: AS(n = 3 studies, 111 patients) Bleeding: yes >4 cm: surgery

Size: yes, but not specifiedLocation: noPain: yesCancer suspicion on imaging: yesPatient preference: no

Surgery vs SAE Size growth: no 1 study (129/213 pts) 4 cm: 1 study (129/213 pts)(n = 3 studies, 213 patients)a Bleeding: yes

Size: 4 cmLocation: noPain: yesCancer suspicion on imaging: yesPatient preference: yes

3-arm AS vs surgery vs SAE Size growth: no 1 study (87/177 pts) 4 cm: 1 study (87/177)(n = 4 studies, 177 patients) Bleeding: yes

Size: yes, but not specifiedLocation: noPain: yesCancer suspicion on imaging: yesPatient preference: no

AS = active surveillance; pts = patients; SAE = selective arterial embolisation.a Only three of five studies in surgery versus SAE report data about “treatment indications”.



3.5. Discussion

3.5.1. Principal findings

The main indications for active treatment of AML were asfollows: (1) increased tumour size, (2) presence ofsymptoms (bleeding and pain), and (3) suspicion of canceron imaging. The 4-cm cut-off to trigger active treatment hasbeen followed for decades [12], mainly based on theassumption that larger tumours have an increased risk ofbleeding. However, there is very little evidence to supportthe 4-cm threshold. In our SR, the 4-cm cut-off was followedin 18 out of 43 studies, which involved 1016 patients


representing 44% of the total AML population. Some studiesconsidered a larger size cut-off ranging from 5 to 10 cm [26–28]. In view of the results of AS series, where a 4-cm cut-offwas considered in only 187 of 899 patients (21%), there is avalid argument that active treatment should not be decidedsolely on the 4-cm cut-off [13,14]. The average size of theactively treated AMLs (that did not bleed) in the largest ASseries was 8 cm, ranging from 0.8 to 29 cm [3]. The AML sizeseems to matter, but the limit upon which active treatmentshould be proposed cannot be defined well based on theavailable evidence. Other factors such as AML growth andpatient’s preference should be taken into account [3,12].



Table 6 – General outcomes in active treatment modalities

Activetreatmentmodality

Growth/relapse Pain/symptom resolution 2nd Treatmentrequirement

2ndTreatmentmodality

Renal functionevolution

SurgeryLane (2008) [27] a Not specified 0% relapse Not reported 0, no 2nd

treatment– Not reported

Boorjian (2007) NSS 4% relapse (2 pts) 100% 1, afterhaemorrhage

SAE Preop Cr 1 andpostop 1.1

Heidenreich (2002) Not specified 0% relapse 100% 0, no 2ndtreatment

– Preop Cr 0.9 andpostop 1.2

Minervini (2007) NSS 0% relapse Not reported 0, no 2ndtreatment

– Preop Cr 0.95 andpostop 0.99

Golan (2017) NSS 0% relapse 100% 1, afterhaemorrhage

SAE Preop Cr 0.85 andpostop NR

Liu (2016) NSS 0% relapse Not reported 0, no 2ndtreatment

– Preop eGFR 42.5/postop 33.5

Wang (2017) [28] NSS � SAE 0% relapse Not reported 0, no 2ndtreatment

– NSS: preop 69.8/postop 84.2NSS/SAE: preop 71.1/postop 70.1

Yip (2000) [25] NSS, RN 4% relapse 100% 1, RCC suspicion onFU

Surgery Not reported

Qin (2017) [23] NSS � SAE 0% relapse 100% 0, no 2ndtreatment

– NSS: <33.05% eGFRpostopNSS/SAE: <15.08%eGFR postop

Lin (2016) NSS 0% relapse Not reported 0, no 2ndtreatment

– Preop eGFR 102.4/postop 99.8

Ng (2016) NSS 15% relapse (3 pts) Not reported Not reported – <11% eGFR postopMsezane (2010) NSS 0% relapse Not reported 1, contralateral

AMLNSS Preop eGFR 99.2/

postop 841, afterhaemorrhage

SAE

Zhang (2016) NSS 0% relapse Not reported 0, no 2ndtreatment

– Preop Cr 92/postop92;Preop eGFR 32/postop 32

SAEWang (2017) [28] PVO, coils 0% 94% 39 (34 “planned”;

5 “unplanned” forgrowth)

39 SAE Not reported

5 (4 haemorrhage,1 growth)

Bardin (2017) 10 agents 13% (3 pts) 91% 8 (7 haemorrhage,1 growth)

4 SAE,1 surgery

Preop eGFR 78/postop 77

5 (2 symptoms,3 growth)

4 SAE,4 surgery

Chan (2011) PVO, coils 6% (1 pts) 81% 2 (1 haemorrhage,1 SAE syndrome)

2 SAE,3 surgery

Not reported

17 (1 haemorrhage,16 growth)

2 surgery

Chick (2010) PVO 17% (6 pts) 83% Not reported Not reported15 SAE,2 surgery

Duan (2016) PVO, coils 4% (1 pts) 96% Not reported – Preop eGFR 58.5/postop 73.5

Ramon (2009) [7] PVO 39% (16 pts) 97.5% 2 (2 haemorrhage) – Preop Cr 0.89/postop0.87

3 (3 growth)Sheth (2016) Microspheres,

coils, gelatin0% relapse Not reported 2 SAE Not reported

Alcohol Not reported 1 SAE,2 surgery

Takebayashi (2009) Not reported Not reported 3 Surgery 4 patients > Cr 0.1–0.2 mg/dl;1 patient > Cr 1–2 mg/dl

EVO 3 (3 haemorrhage) Not reportedUrbano (2017) Lipids and OH 0% relapse 100% 2 SAE,

1 surgeryNot reported

Tso (2005) 23% (3 pts) 41%





Table 6 (Continued )

Activetreatmentmodality

Growth/relapse Pain/symptom resolution 2nd Treatmentrequirement

2ndTreatmentmodality

Renal functionevolution

Microspheres,coils

Not reported

Tillou (2010) 4 agents Not reported 100%Not reported

Chatziioannou 2002) [24] Not reported Not reportedAblativeHan (2015) MWA 0 100% 0 – Not reportedCastle (2012) [11] RFA Not reported Not reported Not reported – Not reportedGuo (2009) RFA 0 Not reported Not reported – Not reportedMakki (2016) Cryotherapy Not reported Not reported Not reported – Not reportedSwanson (2012) RFA,

cryotherapy0 Not reported Not reported 1 Ablative Preop eGFR 77/

postop eGFR 75

AML = angiomyolipoma; Cr = serum creatinine; eGFR = estimated glomerular filtration rate; EVO = ethylene vinyl alcohol; FU = follow-up; MWA = microwaveablation; NSS = nephron-sparing surgery; preop = preoperative; postop = postoperative; pts = patients; PVO = polyvinyl alcohol; RCC = renal cell carcinoma;RFA = radiofrequency ablation; RN = radical nephrectomy; SAE = selective arterial embolisation.a No specification about the modality of surgery (radical or nephron sparing) or approach.

Table 7 – Success and outcomes in active surveillance groups

Surveillance 1 arm4 studies = 899pts

Growth Growth rate(cm/yr)

Spontaneousbleeding/haematuria

Activetreatment(patients)

Active treatment modality(treatments)

Bhatt (2016) [3] 9% (40 pts) 0.021 12 pts 5.59% (25) SAE 22 (73.3%), surgery 4 (13.3%),RFA 2 (6.7%); mtor 2 (6.7%)

Chan (2016) [22] 19.78% (37 pts) 0.013 1 pts 3.2% (6) SAE 6 (100%)Maclean (2014) [26] 11.85% (16 pts) <2 cm (0.07) 3 pts 2.22% (3) SAE 2; surgery 1

2–4 cm (0.1)Ouzaid (2014) [14] 3% (4 pts) >4 cm (0.92) 4 pts 13.07% (17) SAE 11 (64.7%), partial neph 5

(29.4%), mtor 1 (5.9%)Not reported

97/899 20/899 51/899 SAE 41, surgery 10, RFA 2, mtor 3

mtor = mammalian target of rapamycin; neph = nephrectomy; pts = patients; RFA = radiofrequency ablation; SAE = selective arterial embolisation.



Some of the most relevant findings in AS series were therate of AML growth and the risk of spontaneous bleeding. InAS studies, AML size was smaller (only 13.6% of AMLs were>4 cm) than the size reported in treated patients. We foundthat 11% of the patients experienced AML growth and only2% a spontaneous rupture or haematuria. The relationshipbetween AML size and the probability of rupture remainsunclear based on our findings. However, the probability ofspontaneous rupture in small AML appears to be very low[26]. Active treatment in AS series was decided in 6% of thepatients. Patients were most frequently treated by SAE(77%) rather than by surgery (19%). As such, AS of smallAMLs appears to be a safe option. More data are neededregarding AS in larger AMLs (>4 cm), as they accounted for14% of AS patients [3].

Owing to the heterogeneity of the data, surgical and SAEseries were hardly comparable. The results of ablativetherapies were scarcely reported in terms of recurrence,secondary treatments, and complications. Tumour growth orrecurrence occurred in 4–15% of patients treated by surgeryand in 6–39% of patients treated by SAE. However, the


definition of recurrence was different throughout SAE studies.Secondary treatment rate was 1% after surgery and 31% afterSAE. The preferred modality for retreatment was SAE (60%after surgery vs 80% after SAE) [27,28]. Based on these data,surgical treatment seems to perform better in terms ofrecurrence and requirement for secondary treatment. Bothsurgery and SAE were comparable in terms of majorcomplications. One has to acknowledge that the literatureregarding SAE and ablative treatments of AMLs is limitedcompared with surgical treatment. Therefore, although theanalysis of data seems to favour surgery in terms of outcomesand additional treatment, it might just be related topublication bias (with surgical series being much morefrequently reported). Therefore, the message we deliver hasto be interpreted with caution. There was no well-defined FUscheme. In most series, FU relied on US and CT.

3.5.2. Implications for clinical practice

Despite clinical heterogeneity and uncertainty in theevidence base, the following are potentially importantfindings that might be highlighted.



Table 8 – Adverse events in active treatment modalities according to Clavien-Dindo system and hospital length stay

CD1 CD1explanation

CD2 CD2explanation

CD3 CD3explanation

CD4 CD4explanation

CD5 CD5explanation

Hospstay day av.

(range)

SurgeryLane (2008) [27] – e – – – – –

Boorjian (2007) 5 5—Ileus 14 1—Infection 4 1—Haemorrh-SAE – – 6 [2–13]13—Transfusion 3—Urinoma-Stent

Heidenreich (2002) – 1—Urinoma 0 3 – – –

1—Ileus 1—Infection 2–3aMinervini (2007) – 1—Atrial fibr. 2 1—Transfusion 1 1–3b – – 6 [5–11]

2—InfectionGolan (2017) 3 1—Urinoma 4 2—Transfusion 1 1—Infection-drain – – 2 [2,3]Liu (2016) – 1—Ileus 1 1—Transfusion – 1—Haemorrh-SAE – – –

Wang (2017) [28] 1 4 1—Infection – – – –

2—Pain-SAE sd 1—AV fistula-SAEYip (2000) [25] – 1 1—Transfusion 1 – – 9 [6–21]

1—Pseudoan-SAEQin (2017) [23] – 4 1—CVA transient – – – 5 (�1.3)Lin (2016) 1 4 4—-Transfusion – – – 8 (�1.4)Ng (2016) – 3 1—Ileus – – –– 3.3 [2–8]

3—TransfusionMsezane (2010) – 1 1 – – 2.3 (SD 2.6)Zhang (2016) NR NR 1—Infection NR NR NR 7 (SD 2)

3—Transfusion1—Transfusion

SAEWang (2017) [28] – 69 68—SAE sd – – – NR

1—InfectionBardin (2017) – 15 1—Infection 6 2—Infection-drain – 1 1—CVA brain NR

14—SAE sd 3—Haemorrh-SAEChan (2011) – 11 6 1—Haemorrh-Surg – – 5–7 [5–20]

11—SAE sdChick (2010) – 12 – 3—Haemorrh-SAE – – NR

1—Infection 3—Haemorrh-SurgDuan (2016) – 15 11—SAE sd 2 – – NR

15—SAE sd 1—SAE sd-SurgRamon (2009) [7] – 5 1 1—Haemorrh-Surg – – NR

5—SAE sd 1—Haemorrh-SurgSheth (2016) – 6 2 – – NR

6—SAE sd 2—Infection-drainTakebayashi (2009) NR NR NR NR NR [2–4]

5—SAE sd 2—Haemorrh-SAEUrbano (2017) 0 5 0 1—Haemorrh-Surg 0 0 1.5 [1–3]

6—SAE sdTso (2005) 0 6 0 0 0 NR

10—SAE sdTillou (2010) NR 11 1—Infection 0 0 0 8.5 [1–28]

4—SAE sdChatziioannou (2012) [24] – 4 3 – – 6 [2–8]AblativeHan (2015) 11 11—Pain 1 1—Infection-AB 1 1—Colonic injury-Surg – – NRCastle (2012) [11] NR 1—Skin enfisema NR 2—Transfusion NR NR NR NRGuo (2009) NR NR NR NR NR NRMakki (2016) 1 2 – – – NRSwanson (2012) NR NR NR NR NR NRSurg vs Surv3 studies with inconsistent report of adverse events, except one which reports: C1 + C2 = 15; C3 + C4 + C5 = 0Surg vs SAE4 studies with inconsistent report of adverse events, except one with more blood loss for SAE and better hospital stay for SAE (16 vs 19 pts)

AB = antibiotic; CD1 = Clavien-Dindo grade 1; CD2 = Clavien-Dindo grade 2; CD3 = Clavien-Dindo grade 3; CD4 = Clavien-Dindo grade 4; CD5 = Clavien-Dindograde 5; CVA = cerebrovascular accident; fibr. = fibrillation; Hosp = hospital; NR = nor reported; SAE = selective arterial embolisation; Haemorrh = haemorrhage;sd = syndrome; SD = standard deviation; Surg = surgery; Surv = surveillance.







1. The growth rate of sporadic AMLs is low.2. The probability of spontaneous rupture is low.3. The need for active treatment is infrequent.4. There is no high-level evidence to suggest any cut-off size

for active treatment, and the 4-cm cut-off should bereconsidered.

5. NSS seems to be the most effective treatment option inorder to avoid recurrence and secondary treatments.

3.5.3. Strengths and limitations of the review

This review was undertaken in a rigorous manner inaccordance with recognised standards by a multidisciplin-ary panel of clinical, methodological, and patient experts(EAU RCC Guidelines Panel) according to PRISMA guidelines.Overall, the certainty of the evidence obtained from this SRis low, due to the nature of available studies that wereretrospective and mostly noncomparative, had high clinicalheterogeneity, and had moderate to high RoB. This nature ofthe available evidence made a meta-analysis inappropriate.

4. Conclusions

Our SR found that AMLs are mostly managed by AS (48% ofthe cases) and that most of them do not grow (89% of cases).Even if they grow, their size increases very slowly andbleeding is a rare event (2.2%). There is no clear relationshipbetween AML size and the occurrence of bleeding, althoughlarger AMLs seem to be associated with an increased risk ofbleeding. Nevertheless, the 4-cm cut-off that has tradition-ally been used as an indication for active treatment mayhave to be reconsidered and should not be used in isolationto trigger active treatment; other factors such as patient age,rate of growth, and preference should also be taken intoaccount. According to our systematically reviewed data, themost frequently reported active treatment is surgery (31%),particularly NSS, followed by SAE (17%). Both have similarmorbidity, but there are more recurrences and need forsecondary treatment in SAE patients (0.85% in surgery vs31% in SAE). Despite the limitations of the evidence, thefindings of this SR may be used to guide and inform clinicalpractice, until more robust data emerge.

Author contributions: Sergio Fernández-Pello had full access to all thedata in the study and takes responsibility for the integrity of the data andthe accuracy of the data analysis.

Study concept and design: Bensalah, Hora, Dabestani, Fernández-Pello.Acquisition of data: Fernández-Pello, Tahbaz.Analysis and interpretation of data: Fernández-Pello, Lam, Marconi, Hora,Bensalah.Drafting of the manuscript: Fernández-Pello, Kuusk, Bensalah, Hora.Critical revision of the manuscript for important intellectual content:

Fernández-Pello, Kuusk, Hora, Tahbaz, Dabestani, Abu-Ghanem, Albiges,Giles, Hofmann, Kuczyk, Lam, Marconi, Merseburger, Powles, Staehler,Volpe, Ljungberg, Bex, Bensalah.Statistical analysis: None.Obtaining funding: None.Administrative, technical, or material support: None.Supervision: Bensalah, Ljungberg, Bex.Other: None.


Financial disclosures: Sergio Fernández-Pello certifies that all conflicts ofinterest, including specific financial interests and relationships and affilia-tions relevant to the subject matter or materials discussed in the manuscript(eg, employment/affiliation, grants or funding, consultancies, honoraria,stock ownership or options, expert testimony, royalties, or patents filed,received, or pending), are the following: Professor Dr. Axel Bex has receivedcompany speaker honoraria from Pfizer; has participated in trials for PfizerEurope; has participated in advisory boards for GlaxoSmithKline andNovartis; is a company consultant for Pfizer and Novartis; and has receivedgrants/research support from Pfizer. Professor Dr. Thomas Powles is acompany consultant for Novartis, Pfizer, and GlaxoSmithKline; has receivedcompany speaker honoraria from Novartis, Pfizer, GlaxoSmithKline, andGenentech; has participated in trials for GlaxoSmithKline, Pfizer, BMS,Genentech, and Genetech; and has received grants/research support fromGlaxoSmithKline, Pfizer, and Novartis. Professor Dr. Börje Ljungberg hasreceived company speaker honoraria from GlaxoSmithKline, Roche, Pfizer,and Novartis; has participated in trials for GlaxoSmithKline, Medivation,Pfizer, and Janssen R&D; and has been on advisory boards for Pfizer andGlaxoSmithKline. Professor Dr. Laurence Albiges has received consulting/advisory fees from BMS, Pfizer, Novartis, Sanofi, Amgen, Bristol-Myers Squibb,Bayer, and Cerulean; and research funding from Pfizer and Novartis. ProfessorDr. Karim Bensalah has received grants/research support from Pfizer, andhonoraria or consultation fees from Intuitive Surgical. Professor Dr. MichaelStaehler is a companyconsultant for Pfizer, Novartis, GlaxoSmithKline, Roche,Astellas, and Bayer; has received company speaker honoraria from Pfizer,Novartis, GlaxoSmithKline, Roche, Astellas, Bayer, and Aveo; has participatedin trials for Pfizer, Novartis, GlaxoSmithKline, Roche, Bayer, Aveo, Wilex, andImmatics; has received fellowships and travel grants from Pfizer, Novartis,GlaxoSmithKline, Roche, and Bayer; has received grants/research supportfrom Pfizer, Novartis, GlaxoSmithKline, Roche, Bayer, and Aveo; and took partin the S-TRAC trial as an investigator and is an author of the S-TRACpublication. Professor Dr. Milan Hora has received company speakerhonoraria from Covidien, Olympus, Janssen, and Astellas; has participatedin trials for Janssen; and has received grants/research support from Ipsen.Professor Dr. Markus A. Kuczyk is a stock shareholder of Bayer Healthcare,Astellas, Storz, Pfizer, Wyeth, and Novartis; is a company consultant for KarlStorz, Coloplast, AstraZeneca, Astellas, Storz, and Hexal; has receivedcompany speaker honoraria from Pfizer, Astellas, Bayer, GlaxoSmithKline,Pierre Fabre, Janssen Cilag, and Hexal; has participated in trials for the ProtecTStudy, Millenium Study C21004, Millenium Study C21005, Astellas, Ipsen, andJanssen; and has received grants/research support from Wyeth and Pfizer. Dr.Thomas B. Lam is a company consultant for and has received companyspeakerhonoraria from Pfizer, GlaxoSmithKline, Astellas, and Ipsen. ProfessorDr. Axel S. Merseburger is a company consultant for Ipsen Pharma, Bayer,Astellas, Janssen Cilag, Novartis, and Pfizer; has received company speakerhonoraria from Ipsen Pharma, Wyeth, Astellas, Novartis, Pfizer, and SEP; hasparticipated in trials for AstraZeneca, Bayer, Pfizer, TEVA, Novartis, andAstellas; has received grants/research support from Wyeth; and hasparticipated in a company-sponsored speakers bureau for TEVA, Janssen,Pfizer, Astellas, Ferring, and Novartis. Dr. Teele Kussk has received companyspeaker honoraria from Ipsen. Professor Dr. Rachel H. Giles, Professor Dr.Alessandro Volpe, Dr. Saeed Dabestani, Dr. Fabian Hofmann, Dr. LorenzoMarconi, Dr. Sergio Fernández-Pello, Dr. Rana Tahbaz, and Dr. Yasmin Abu-Ghanem have nothing to disclose.

Funding/Support and role of the sponsor: None.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at doi:10.1016/j.euo.2019.04.005


http://dx.doi.org/10.1016/j.euo.2019.04.005




References

[1] Martignoni G, Amin MB. Angiomyolipoma. World Health Organiza-tion classification of tumours. In: Eble JN, Sauter G, Epstein JI,Sesterhenn IA, editors. Pathology and genetics of tumours of theurinary system and male genital organs. Lyon, France: IARC Press;2004. p. 65–7.

[2] Nelson CP, Sanda MG. Contemporary diagnosis and management ofrenal angiomyolipoma. J Urol 2002;168(4 Pt 1):1315–25.

[3] Bhatt JR, Richard PO, Kim NS, et al. Natural history of renal angio-myolipoma (AML): most patients with large AMLs >4 cm can beoffered active surveillance as an initial management strategy. EurUrol 2016;70:85–90.

[4] Mues AC, Palacios JM, Haramis G, et al. Contemporary experience inthe management of angiomyolipoma. J Endourol 2010;24:1883–6.

[5] Eble JN, Sauter G, Epstein JI, Sesterhenn (Eds.) IA. World HealthOrganization Classification of TumoursPathology and genetics oftumours of the urinary system and male genital organs. Lyon: IARCPress; 2004.

[6] Nese N, Martignoni G, Fletcher CD, et al. Pure epithelioid PEComas(so-called epithelioid angiomyolipoma) of the kidney: a clinico-pathologic study of 41 cases: detailed assessment of morphologyand risk stratification. Am J Surg Pathol 2011;35:161–76.

[7] Ramon J, Rimon U, Garniek A, et al. Renal angiomyolipoma: long-term results following selective arterial embolization. Eur Urol2009;55:1155–61.

[8] Hocquelet A, Cornelis F, Le Bras Y, et al. Long-term results ofpreventive embolization of renal angiomyolipomas: evaluation ofpredictive factors of volume decrease. Eur Radiol 2014;24:1785–93.

[9] Murray TE, Doyle F, Lee M. Transarterial embolization of angiomyo-lipoma: a systematic review. J Urol 2015;194:635–9.

[10] Staehler M, Sauter M, Helck A, et al. Nephron-sparing resection ofangiomyolipoma after sirolimus pretreatment in patients withtuberous sclerosis. Int Urol Nephrol 2012;44:1657–61.

[11] Castle SM, Gorbatiy V, Ekwenna O, Young E, Leveillee RJ. Radio-frequency ablation (RFA) therapy for renal angiomyolipoma (AML):an alternative to angio-embolization and nephron-sparing surgery.BJU Int 2012;109:384–7.

[12] Oesterling JE, Fishman EK, Goldman SM, Marshall FF. The manage-ment of renal angiomyolipoma. J Urol 1986;135:1121–4.

[13] Kuusk T, Biancari F, Lane B, et al. Treatment of renal angiomyoli-poma: pooled analysis of individual patient data. BMC Urol2015;15:123.

[14] Ouzaid I, Autorino R, Fatica R, et al. Active surveillance for renalangiomyolipoma: outcomes and factors predictive of delayed in-tervention. BJU Int 2014;114:412–7.

[15] Moher D, Altman DG, Liberati A, Tetzlaff J. PRISMA statement.Epidemiology 2011;22:128, [author reply 128].

[16] Higgins JPT, Green S. Cochrane handbook for systematic reviews ofinterventions version 5.1.0. The Cochrane Collaboration; 2011 .http://handbook.cochrane.org

[17] van den Bergh RCN, van Casteren NJ, van den Broeck T, et al. Role ofhormonal treatment in prostate cancer patients with nonmetastaticdisease recurrence after local curative treatment: a systematicreview. Eur Urol 2016;69:802–20.

[18] Deeks JJ, Dinnes J, D’Amico R, et al. Evaluating non-randomisedintervention studies. Health Technol Assess 2003;7, iii–x, 1–173.

[19] Reeves BC, Higgins JPT, Ramsay C, Shea B, Tugwell P, Wells GA. Anintroduction to methodological issues when including non-ran-domised studies in systematic reviews on the effects of interven-tions. Res Synth Methods 2013;4:1–11.

[20] Dalziel K, Round A, Stein K, Garside R, Castelnuovo E, Payne L. Do thefindings of case series studies vary significantly according to methodo-logical characteristics? Health Technol Assess 2005;9, iii–iv, 1–146.


[21] Viswanathan M, Ansari M, Berkman N, et al. Methods guide forcomparative effectiveness reviews. Assessing the risk of bias ofindividual studies in systematic reviews of health care interven-tions. Agency for Healthcare Research and Quality web site. 2017.http://www.effectivehealthcare.ahrq.gov/ehc/products/322/998/MethodsGuideforCERs_Viswanathan_IndividualStudies.pdf.

[22] Chan KE, Chedgy E, Turner KJ. Contemporary management ofsporadic angiomyolipomas of the kidney, does size matter? Sur-veillance guidelines and review of management. J Endourol2016;30:A181.

[23] Qin C, Wang Y, Li P, et al. Super-selective artery embolization beforelaparoscopic partial nephrectomy in treating renal angiomyoli-poma. Urol Int 2017;99:277–82.

[24] Chatziioannou A, Gargas D, Malagari K, et al. Transcatheter arterialembolization as therapy of renal angiomyolipomas: the evolutionin 15 years of experience. Eur J Radiol 2012;81:2308–12.

[25] Yip SK, Tan PH, Cheng WS, Li MK, Foo KT. Surgical management ofangiomyolipoma: nephron-sparing surgery for symptomatic tu-mour. Scand J Urol Nephrol 2000;34:32–5.

[26] Maclean DFW, Sultana R, Radwan R, McKnight L, Khastgir J. Is thefollow-up of small renal angiomyolipomas a necessary precaution?Clin Radiol 2014;69:822–6.

[27] Lane BR, Aydin H, Danforth TL, et al. Clinical correlates of renalangiomyolipoma subtypes in 209 patients: classic, fat poor, tuber-ous sclerosis associated and epithelioid. J Urol 2008;180:836–43.

[28] Wang C, Yang M, Tong X, et al. Transarterial embolization for renalangiomyolipomas: a single centre experience in 79 patients. J IntMed Res 2017;45:706–13.

[29] Amano T, Miwa S, Imao T, Takashima H, Takemae K. Comparativestudy of operation and surveillance for patients with renal angio-myolipoma. Hinyokika Kiyo 2002;48(2):67–70.

[30] De Luca S, Terrone C, Rossetti RS. Management of renal angiomyo-lipoma: a report of 53 cases. BJU Int 1999;83(3):215–8.

[31] Delhorme JB, Fontana A, Levy A, Terrier P, Fiore M, Tzanis D, et al.Renal angiomyolipomas: At least two diseases. A series of patientstreated at two European institutions.. Eur J Surg Oncol 2017 Apr;43(4):831–6.

[32] Huyghe E, Delchier MC, Mottier ML, Malavaud M, Otal P, Soulie M,et al. Management of renal angiomyolipoma: medical and costeffectiveness comparison between selective embolization and sur-gery.. Eur Urol Suppl 2009;8(4):203.

[33] Koo KC, Kim WT, Ham WS, Lee JS, Ju HJ, Choi YD, et al. Trends ofpresentation and clinical outcome of treated renal angiomyoli-poma. Yonsei Med J 2010 Sep;51(5):728–34.

[34] Koo KC, Kim WT, Ham WS, Lee JS, Ju HJ, Choi YD, et al. Trends ofpresentation and clinical outcome of treated renal angiomyoli-poma. Yonsei Med J 2010 Sep;51(5):728–34.

[35] Lee S-Y, Hsu H-H, Chen Y-C, Huang C-C, Wong Y-C, Wang L-J, et al.Evaluation of renal function of angiomyolipoma patients afterselective transcatheter arterial embolization. Am J Med Sci2009;337(2):103–8.

[36] Seyam RM, Bissada NK, Kattan SA, Mokhtar AA, Aslam M, Fahmy WE,et al. Changing trends in presentation, diagnosis and management ofrenal angiomyolipoma: comparison of sporadic and tuberous sclero-sis complex-associated forms. Urology 2008;72(5):1077–82.

[37] Han Z, Liang P, Yu X, Cheng Z, Liu F, Yu J. Ultrasound-guidedpercutaneous microwave ablation of sporadic renal angiomyoli-poma: preliminary results. Acta Radiol 2015;56(1):56–62.

[38] Guo H, Gan W, Yang R, Yan X, Lian H, Ji C, et al. Laparoscopic cool-tipradiofrequency ablation in renal angiomyolipoma. Urology 2009;74(Sup 4):S188.

[39] Makki A, Ostraat A, Hoyer S, Graumann O. T Kjaergaard Arecryoablation a feasible and safe treatment modality for renalangiomyolipoma? J Endourol 2016;30:A245.


http://refhub.elsevier.com/S2588-9311(19)30054-9/sbref0145














































http://handbook.cochrane.org/














http://www.effectivehealthcare.ahrq.gov/ehc/products/322/998/MethodsGuideforCERs_Viswanathan_IndividualStudies.pdf

http://www.effectivehealthcare.ahrq.gov/ehc/products/322/998/MethodsGuideforCERs_Viswanathan_IndividualStudies.pdf






























































[40] Swanson J, Atwell TD, Schmit GD, Kurup AN, Weisbrod AJ, Chow Gk,et al. Percutaneous thermal ablation of renal angiomyolipomas:feasibility and safety. J Vasc Interv Radiol 2012;23(3):Sup1.

[41] Boorjian SA, Frank I, Inman B, Lohse CM, Cheville JC, Leibovich BC,et al. The role of partial nephrectomy for the management ofsporadic renal angiomyolipoma. Urology 2007;70(6):1064–8.

[42] Heidenreich A, Hegele A, Varga Z, von Knobloch R, Hofmann R.Nephron-sparing surgery for renal angiomyolipoma. Eur Urol2002;41(3):267–73.

[43] Minervini A, Giubilei G, Masieri L, Lanzi F, Serni S, Carini M. Simpleenucleation for the treatment of renal angiomyolipoma. BJU Int2007;99(4):887–91.

[44] Golan S, Johnson SC, Maurice MJ, Kaouk JH, Lai WR, Lee BR, et al.Safety and early effectiveness of robot-assisted partial nephrectomyfor large angiomyolipomas. BJU Int 2017;119(5):755–60.

[45] Liu W, Qi L, Chen M, Wang L, Liu L, Zhu B, et al. LaparoscopicRetroperitoneal Enucleation-Separation Surgery for Renal Angio-myolipoma: Perioperative and Oncologic Outcomes Based on aRandomized Controlled Trial. J Endourol 2016;30(8):901–5.

[46] Lin C-Y, Yang C-K, Ou Y-C, Chiu K-Y, Cheng C-L, Ho H-C, et al. Long-term outcome of robotic partial nephrectomy for renal angiomyo-lipoma. Asian J Surg 2018;41(2):187–91.

[47] Ng WM, Lim CY, Chiu A, Huang KS. Laparoscopic retroperitoneo-scopic partial nephrectomy for renal angiomyolipoma. Int J Urol2016;23:Sup1.

[48] Msezane L, Chang A, Shikanov S, Deklaj T, Katz MH, Shalhav AL, et al.Laparoscopic nephron-sparing surgery in the management of angio-myolipoma: a single center experience. J Endourol 2010;24(4):583–7.

[49] Bardin F, Chevallier O, Bertaut A, Delorme E, Moulin M, Pottecher P,et al. Selective arterial embolization of symptomatic and


asymptomatic renal angiomyolipomas: a retrospective study ofsafety, outcomes and tumor size reduction. Quant Imaging MedSurg 2017;7(1):8–23.

[50] Chan CK, Yu S, Yip S, Lee P. The efficacy, safety and durability ofselective renal arterial embolization in treating symptomatic andasymptomatic renal angiomyolipoma. Urology 2011;77(3):642–8.

[51] Chick CM, Tan B-S, Cheng C, Taneja M, Lo R, Tan Y-H, et al. Long-termfollow-up of the treatment of renal angiomyolipomas after selectivearterial embolization with alcohol. BJU Int 2010;105(3):390–4.

[52] Duan X-H, Zhang M-F, Ren J-Z, Han X-W, Chen P-F, Zhang K, et al.Urgent transcatheter arterial embolization for the treatment ofruptured renal angiomyolipoma with spontaneous hemorrhage.Acta Radiol 2016;57(11):1360–5.

[53] Sheth RA, Feldman AS, Paul E, Thiele EA, Walker TG. Sporadic versusTuberous Sclerosis Complex-Associated Angiomyolipomas: Predic-tors for Long-Term Outcomes following Transcatheter Emboliza-tion. J Vasc Interv Radiol 2016;27(10):1542–9.

[54] Takebayashi S, Horikawa A, Arai M, Iso S, Noguchi K. Transarterialethanol ablation for sporadic and non-hemorrhaging angiomyoli-poma in the kidney. Eur J Radiol 2009;72(1):139–45.

[55] Urbano J, Paul L, Cabrera M, Alonso-Burgos A, Gómez D. Elective andEmergency Renal Angiomyolipoma Embolization with EthyleneVinyl Alcohol Copolymer: Feasibility and Initial Experience. J VascInterv Radiol 2017;28(6):832–9.

[56] Tso WK, Wong S, Tsang J, Tung H, Tai KS, Tam PC, et al. Embo-lotherapy with a mixture of lipiodol and ethanol for renal angio-myolipoma J HK Coll Radiol (8):2005;72–82.

[57] Tillou X, Boutemy F, Remond A, Petit J. [Contribution of curative andpreventive embolization for renal angiomyolipomas treatment].Prog Urol 2010;20(9):627–32.































































management of sporadic renal angiomyolipomas: a …...1. introduction renal angiomyolipoma (aml) is...

Documents