Powered by
05:57 CET
CT 2 - Clinical Trials in Radiology 1
Genitourinary Breast Interventional Radiology
Wednesday, February 27, 10:30 - 12:00
Type of session: Clinical Trials in Radiology
Topic: Genitourinary, Breast, Interventional Radiology
Moderators: V. P. Jackson (Tucson/US), M. Dewey (Berlin/DE)

Primary stenting of superficial femoral artery in patients with intermittent claudication has durable effects on Health-Related Quality of Life at 24 Months: results of a randomised controlled trial
H. Lindgren1, A. Gottsater2; 1Helsingborg/SE 2Malmö/SE

Purpose: This prospective randomised trial assess the 24-month impact of primary stenting with nitinol self-expanding stents compared to best medical treatment alone in patients with stable IC due to SFA disease on Health-Related Quality of Life.
Ethics committee approval: Written informed consent was obtained from all patients. The study was approved by the Medical Ethics Committee of Lund University (Dnr 2009/478).
Methods and materials: One hundred patients with stable IC due to SFA disease from seven Swedish hospitals treated with best medical treatment (BMT) were randomised to either the stent (n=48) or the control (n=52) group. Change in HRQoL 24 months after treatment was a primary outcome measure.
Results: Significantly better SF-36 Physical Component Summary (mean difference =5, 1; P=.024) and physical domain scores Physical Function (P=.012), Bodily Pain (P=.002), General Health (P=.037), and EQ5D (P=.010) were reported in intergroup and intragroup comparisons between the stent and the control group.
Limitations: The long inclusion period and large number of screening failures due to the strict inclusion and exclusion criteria might have created a bias in patient selection
Conclusion: In patients with IC caused by lesions in the SFA, primary stenting compared to BMT alone was associated with significant durable improvements in HRQoL, ABI, and walking distance.
Funding for this study: This study was supported by grants from the Gorthons Foundation, the Ernhold Lundström Foundation, Regional research funds and funds at Skåne University Hospital, the Albert Påhlsson Foundation, the Hulda Ahlmroth Foundation, Mediel AB, and from the Swedish state under the LUA/ALF agreement.

T. J. Kroencke; Augsburg/DE
Ultrasound-guided vs computed tomography-guided renal mass biopsy: a prospective randomised study
H. M. Farg, A. Abdelhamid, M. Zakaria, M. El-Ksas, A. El Mahdy, M. Abd El-Hameed, T. El-Diasty; Mansoura/EG

Purpose: The purpose of this study was to compare ultrasound (US) vs computed tomography (CT) guidance for renal mass biopsy.
Ethics committee approval: The study was approved by the local ethics committee. Written informed consent was obtained from all participants.
Methods and materials: This was a single-centre prospective randomised trial comparing US vs CT guidance for renal mass biopsy. Between June 2016 and May 2018, eligible patients with renal masses were randomly assigned in a 1:1 ratio to perform either US- or CT-guided biopsy using 18-gauge automated device. The primary outcome was to evaluate the diagnostic success rate in both groups. Secondary outcomes included total procedure time, total punctures, and complications.
Results: A total of 82 patients (men, 57%; mean age, 59.2 years) were included; 41 in each group. The diagnostic success rate was higher in the US group (90%) compared to the CT group (83%); (p <0.001). Four of the patients undergoing US-guided biopsy required re-biopsy compared to 7 patients in CT group; (p <0.001). US significantly shortened the total procedure time (150 seconds) compared to CT group (330 seconds). The complication rate was higher in the CT group (22%) than in the US group (12%). There were no procedure-related deaths in either group.
Limitations: Failed biopsy is not usually related to the image guidance procedure and it may be related to sampling error resulting from tumour haemorrhage or necrosis.
Conclusion: Ultrasound-guided renal mass biopsy has better diagnostic accuracy, less procedural time and lower complication rate than that of CT guidance.
Funding for this study: No funding from industry and/or a funding agency.

V. Bérczi; Budapest/HU
Self-compression mammography in clinical practice: a randomised clinical trial compared to standard compression mammography
V. Iotti, M. Guberti, L. Canovi, P. Giorgi Rossi, M. Bertolini, R. Vacondio, P. Pattacini; Reggio Emilia/IT

Purpose: Self-compression for mammography may help reaching better compression, with a potential improvement of image quality and reduction of dose. To test the efficacy of a self-compression device, compared to radiographer-lead compression, to reduce the average-glandular-dose without affecting image quality. Secondary outcomes are pain, compression, thickness, time to perform mammography. Here we present preliminary results.
Ethics committee approval: Approved by the provincial Ethical Committee. All participants signed a written informed consent form.
Methods and materials: Women presenting for mammography for breast-cancer follow-up, symptoms, opportunistic screening or familial-risk were asked to participate in this prospective randomised trial and randomised to self-compression or to radiographer-compression. Anxiety was measured before mammography; pain (visual-assessment-scale) and discomfort were measured after. Image quality was assessed blindly by two independent radiologists and two radiographers. Room-in/room-out time has been measured.
Results: From January2018 to July2018, 496 women were enrolled, 245 and 251 in self- and radiographer-compression arm, respectively. Image quality (radiologists’ judgement p=0.62; radiographers’ judgement p=0.34, slightly better in self-compression for both), dose (0.94 vs. 0.96mGy, p=0.38) and pain (3.8 vs. 4.0, p=0.50) were similar in the two arms; compression was higher in self- than radiographer-arm (114 vs. 102daN, p<0.00005), with a small impact on thickness (46.4 vs. 48.1mm, p=0.13). Moderate/severe discomfort was reported by 7.9% vs 9.9% (p=0.46). Time was slightly higher 11.5 vs.10.2m (p=0.06). Results were similar in women with higher anxiety level.
Limitations: Screening population was not included.
Conclusion: Self-compression has no impact on dose, pain and discomfort, even if increases compression.
Funding for this study: Sustained by the institutional funds of the ReggioEmilia Local Health Authority–IRCCS.

A. Oktay; Izmir/TR
MRI vs mammography screening for high-risk women: an RCT - the influence of density and effectiveness
A. Geuzinge1, S. Saadatmand1, E. J. J. Rutgers2, R. M. Mann3, D. de Roy van Zuidewijn4, H. Zonderland2, R. Tollenaar5, M. Lobbes6, M. Ausems7, M. van 't Riet8, M. Hooning1, I. Mares-Engelberts1, E. Luiten9, E. Heijnsdijk1, C. Verhoef1, N. Karssemeijer3, J. Oosterwijk10, I.-M. Obdeijn1, H. de Koning1, M. Tilanus-Linthorst1; 1Rotterdam/NL 2Amsterdam/NL 3Nijmegen/NL 4Leeuwarden/NL 5Leiden/NL 6Maastricht/NL 7Utrecht/NL 8Delft/NL 9Breda/NL 10Groningen/NL

Purpose: American guidelines advise women with a familial breast cancer risk MRI-screening, but European mammography. FaMRIsc is the first randomised controlled trial comparing both strategies.
Ethics committee approval: The IRB of ErasmusUniversityMC approved FaMRIsc. All participants gave written informed consent.
Methods and materials: From January 2011 until December 2017, women with a cumulative lifetime risk (CLTR) of ≥20% without a BRCA1/2 mutation, aged 30–55 years, were in twelve Dutch centres randomised into two arms: A. yearly MRI-screening, clinical breast examination (CBE), and biennially mammography or B. yearly mammography and CBE. Primary outcomes were number, size and nodal stage of the detected breast cancers. Results were also stratified by mammographic density.
Results: More breast cancers were detected in the MRI-arm (N=675) compared to the Mx-arm (N=680) (41 vs 14, p<0.001), after on average 4.3 screening rounds per woman. In the MRI-arm invasive cancers were smaller (median size 8 vs 17 mm, p=0.006) and less often node positive (20% vs 71.4%, p=0.019). Sensitivity hardly differed (95.1% vs 92.9%, p=1), but false positive rate was higher in MRI-arm (160 vs 89.9/1000 woman-year at risk p<0.001). All tumours ≥T2 were in the two highest density categories. In ACR density A-C MRI detected more small tumours than mammography. Results of incident rounds and cost will be discussed.
Limitations: are discussed
Conclusion: More cancers were detected in the MRI-arm than in the mammography-arm and in a relevantly earlier stage also at low density. We will discuss at what cost.
Funding for this study: This study was supported by ZonMw, The Dutch Cancer Society, A Sisters Hope, Pink Ribbon, Stichting Coolsingel.

R. Schulz-Wendtland; Erlangen/DE
Variability in cancer-positive predictive value using PI-RADS v2: a multi-institutional study conducted by the SAR Prostate Cancer Disease Focused Panel (aim 2)
A. C. Westphalen1, A. B. Rosenkrantz2, C. McCulloch1, J. Anaokar3, S. arora4, N. Barashi5, J. Barentsz6, T. Bathala7, L. Bittencourt8, M. Booker9, V. Braxton4, D. Casalino5, S. Chang10, F. Coakley11, R. Dhatt12, S. C. Eberhardt13, B. Foster11, A. Froemming14, J. J. Futterer6, M. Gertner15, L. Gettle16, S. Ghai15, R. Gupta17, M. Hahn9, R. Houshyar18, C. Kim19, C. Kim20, C. Lall21, D. Margolis22, S. McRae7, A. Oto5, r. parsons3, N. Patel23, T. Polascik17, J. Starcevich13, B. Spilseth24, V. Tammisetti7, S. Taneja22, S. Verma25, C. Warlick24, A. Weinberger5, J. Yu19, R. J. Zagoria1; 1San Francisco/US 2New York City/US 3Philadelphia/US 4Nashville/US 5Chicago/US 6Nijmegen/NL 7Houston/US 8Rio de Janeiro/BR 9San Diego/US 10Vancouver/CA 11Portland/US 12Vancouver/US 13Albuquerque/US 14Rochester/US 15Toronto/CA 16Madison/US 17Durham/US 18Irvine/US 19Richmond/US 20Seoul/KR 21Orange/US 22New York/US 23Aurora/US 24Minneapolis/US 25Cincinnati/US

Purpose: To determine the variability of positive predictive value (PPV) across centres according to PI-RADS v2 scores.
Ethics committee approval: IRB-approved, HIPPA-compliant study, with waiver of signed informed consent.
Methods and materials: This retrospective included 3,479 men who underwent prostate mpMRI at 26 DFP-member expert centres from several countries and had a lesion classified as PI-RADS ≥ 2. Exams scored as PI-RADS 1 (i.e. no visible lesion) were not included. MR-targeted biopsy results were the standard of reference. PPV were derived from generalised estimating equations regression to estimate variability across centres.
Results: Across all centres, the PPV for PI-RADS score ≥ 2 was 33.8% (range: 12.9%, 68.6%), for PI-RADS ≥ 3 was 37.5% (range, 14.7%, 68.6%), and for PI-RADS ≥ 4, 48.4% (26.4%, 75.9%). The PPV stratified by PI-RADS scores were as follow: score 2=6.9% (range, 4.2%, 50.0%); score 3=15.3% (3.2%, 54.0%), score 4=38.4% (17.7%, 70.0%), and score 5=71.9% (40.0%, 97.1%).
Limitations: Outcomes based on biopsy modalities that are different and not perfect.
Conclusion: Enormous variation in the PPV exists across expert centres, and the overall PPV among these centres was alarmingly low. Based on the results, a PPV ≥ 50% for PI-RADS v2 score ≥ 4 is proposed as a performance standard, with centres below this benchmark warranting quality improvement efforts. Many of the participating centres did not achieve this threshold.
Funding for this study: University of California San Francisco, Department of Radiology and Biomedical Imaging Seed Grant # 16-43.

H. Kressel; Boston/US
Role of PI-RADS v2 in clinical models: a multi-institutional study conducted by the SAR Prostate Cancer Disease Focused Panel (aim 3)
A. C. Westphalen1, A. B. Rosenkrantz2, C. McCulloch1, J. Anaokar3, S. arora4, N. Barashi5, J. Barentsz6, T. Bathala7, L. Bittencourt8, M. Booker9, V. Braxton4, D. Casalino5, S. Chang10, F. Coakley11, R. Dhatt10, S. C. Eberhardt12, B. Foster11, A. Froemming13, J. J. Futterer6, M. Gertner14, L. Gettle15, S. Ghai16, R. Gupta17, M. Hahn9, R. houshyar18, C. Kim19, C. Kim20, C. Lall18, D. Margolis21, S. McRae7, A. Oto5, r. parsons3, N. Patel22, T. Polascik17, J. Starcevich12, B. Spilseth23, V. Tammisetti7, S. Taneja21, S. Verma24, C. Warlick23, A. Weinberger5, J. Yu19, R. J. Zagoria1; 1San Francisco/US 2New York City/US 3Philadelphia/US 4Nashville/US 5Chicago/US 6Nijmegen/NL 7Houston/US 8Rio de Janeiro/BR 9San Diego/US 10Vancouver/CA 11Portland/US 12Albuquerque/US 13Rochester/US 14Toronto/CA 15Madison/US 162n2/CA 17Durham/US 18Orange/US 19Richmond/US 20Seoul/KR 21New York/US 22Aurora/US 23Minneapolis/US 24Cincinnati/US

Purpose: To determine if PI-RADS v2 improves the prediction of high Gleason score prostate cancer beyond PSA, PSA density (PSAD), and clinical stage (T1C versus T2/T3).
Ethics committee approval: IRB-approved with waiver for signed consent.
Methods and materials: This retrospective study included 3,479 men who underwent prostate MRI at one of 26 institutions in the USA, Canada, Brazil, South Korea, and Netherlands. Up to 4 lesions/case were classified using PI-RADS. Cognitive, fusion, and in-bore MR-targeted biopsies were performed. Data was collected using REDCap. The impact of PI-RADS on clinical variables was determined using logistic regressions derived area under the ROC curves. Confidence intervals were calculated using cluster-corrected (to account for multiple lesions/person) bootstrapping.
Results: PI-RADS score, logPSA, logPSAD, and palpable nodule were predictors of GS≥3+4 on uni- (all P<0.001) and multivariate analyses (palpable nodule=0.03, others P<0.001). The PI-RADS AU-ROC was 73.9% (72.6%, 75.3%). Adding PI-RADS to the clinical model raised the AU-ROC from 60.7% (59.1%, 62.2%) to 80.8% (79.5%, 82.2%). The differences between the AU-ROC of the PI-RADS and combined model (6.8%; 5.8%, 7.9%), and of the clinical and combined model (20.1%; 19.0%, 21.4%) were statistically significant (P < 0.001).
Limitations: All centres are expert sites, so results may not apply to all locales.
Conclusion: PI-RADS predicts GS≥3+4 prostate cancer better than clinical variables, but its performance may be enhanced by knowledge of PSA, PSAD, and/or presence of palpable nodule.
Funding for this study: UCSF Department of Radiology Seed Grant # 16-43.

E. Rud; Oslo/NO
This website uses cookies. Learn more