The role of diagnostic ultrasound imaging for patients with known prostate cancer within an active surveillance pathway: A systematic review

Abstract

Background

The use of multiparametric magnetic resonance imaging (mpMRI) within active surveillance of prostate cancer programmes is identified by the UK National Institute for Health and Care Excellence (NICE guideline NG 131 2019) as having a role for monitoring disease. The widespread demands on mpMRI capacity may limit its use in surveillance. It is therefore timely to review the options that modern ultrasound imaging present to this cohort of patients in the monitoring of prostate cancer.

Methods

Between April and September 2020, 10 databases were searched to recruit studies for the review. Three reviewers evaluated the publications for inclusion. Characteristics including the inclusion criteria for the study cohort, how disease was determined, identification of disease progression, and the modality and mode of imaging used were reviewed. Given the paucity of full text articles, a meta-analysis was not possible. A narrative review was undertaken.

Results

In total, 12 studies, utilising the range of ultrasound parameters of B-mode, micro-ultrasound, colour Doppler, contrast ultrasound and elastography were included. The review demonstrated that micro-ultrasound offers promise as an imaging tool comparable with mpMRI. However, this is an emerging technology with limited availability. Analysis of the data further demonstrated that by combining the diagnostic features provided by multiple modes reviewed, ultrasound has a role in the diagnostic imaging of patients on active surveillance.

Conclusion

Providing a multiparametric approach is utilised, stable ultrasound findings may allow for increased intervals between biopsy for men on surveillance. The advent of micro-US offers promise as an imaging modality within an active surveillance pathway but requires further verification.

Keywords

Prostate cancer active surveillance watchful waiting ultrasound micro-ultrasound multi-parametric ultrasound Doppler

Introduction

Despite the fact that active surveillance (AS) has been advocated as a useful, safe and effective pathway in the management of prostate cancer (PCa) for over 10 years,^1,2 recent publications by Merriel et al.³ and Bruinsma et al.⁴ both highlight a lack of consistency in AS monitoring protocols. In particular, there is little consistency in the guidelines for the recommended use of multiparametric magnetic resonance imaging (mpMRI)⁴; however, the PROMIS study⁵ demonstrated that incorporating mpMRI into the diagnostic pathway as an initial test prior to prostate biopsy may improve the detection of clinically significant prostate cancer. The Prostate Cancer UK consensus statement³ highlighted that the application of mpMRI for monitoring disease is evolving but also acknowledged that there are significant gaps in the literature surrounding the best use of mpMRI in AS, despite mpMRI being included with the 2019 NICE guideline NG 131.⁶ For this to be a valuable imaging modality, regular access to high-quality mpMRI, in terms of both imaging protocols and reporting standards is required.⁷ There is known variability between the reporters of mpMRI despite the use of a standardised reporting tool.⁸ This variability, coupled with the challenges associated with mpMRI including poor visualisation of equivocal and/or mpMRI invisible lesions which may be present in patients on AS, has resulted in a reduced sensitivity of mpMRI being reported.⁹ Despite the limitations of mpMRI in AS, there are emerging data that support its role in selecting and monitoring men on such a pathway.^10,11 However, formalising this into routine practice, given the very real variation in AS programmes³ and in the variable access to imaging, is a significant challenge.¹² There are persistent and ongoing barriers to the widespread use of mpMRI for AS, including quality, cost and capacity.¹¹ It is therefore timely to review the option that the modern, alternative imaging modality of ultrasound presents to this cohort of patients.

Ultrasound technology

Publications from the early 2000s related to imaging of patients on monitoring programmes do mention ultrasound (US),¹³ but the technological capabilities of the early machines precluded this modality from having a useful role. Technological advances within the last decade have potentially changed that and there is now limited evidence that the use of US may, indeed, have a useful function in the identification of PCa.^7,14,15 A meta-analysis by Zhang et al. in 2019¹⁶ assessed the sensitivity and specificity of high frequency micro-ultrasound (micro-US) detection of PCa. This technology has been widely utilised in US imaging of superficial structures¹⁷ but is a relatively new technique in the field of transrectal imaging.¹⁸ Despite the limitations of US at frequencies in the range of 29 MHz, there are promising results from published studies^7,18 which supports its use in PCa. The 2019 meta-analysis¹⁶ suggests that micro-US is a more convenient and cost-effective method of imaging and detecting clinically significant PCa, although the study did not specifically evaluate its use in AS and concluded that a comprehensive evaluation remains a necessity.

There is significant heterogeneity across AS guidelines and protocols¹⁹ including regarding the use and timing of mpMRI. Recent publications suggest the role of micro-US¹⁶ adds further choice, but there is a lack of clarity around its use and may hinder or confuse clinician choice in pathway selection. Given the earlier technological limitations of standard frequency US it is, perhaps, understandable why clinicians may not have the confidence to use this modality in the monitoring of patients in such a sensitive and crucial pathway. If US cannot identify the critical window ²⁰ where intervention is required, consequently resulting in a negative impact on patient outcome, its use within a surveillance pathway is significantly restricted. However, the improvements in US described may lead to this modality being a useful adjunct to the monitoring of PCa and thus relieve demand on mpMRI.^21,22 Encouragingly, recent publications^7,23 have identified good correlation between mpMRI detected lesions of prostate imaging reporting and data system (PI‐RADS) 3, 4 and 5 and identifiable features on US. A range of US parameters to assess PCa, including standard frequency ultrasound imaging,²³ tissue perfusion assessment with Doppler, ²⁴ elastography²⁵ and the emerging high frequency micro-US,¹⁸ have also been investigated independently, leading to the suggestion that there may be a useful role for multiparametric ultrasound (mpUS) imaging in PCa,²⁶ other than primarily as a tool to guide tissue sampling biopsy procedures.

Aim

The aim of this study was to perform a systematic review to better understand the role of diagnostic US within the active surveillance pathway of PCa and to clearly identify where gaps in knowledge and data exist.

Methods

Search strategy

The search protocol for this study was submitted for registration to the PROSPERO database of the National Institute for Health Research²⁷ (registration number: CRD42020199936). Between April and September 2020, 10 databases were systematically searched for eligible articles: AMED, BNI, CINAHL, EMBASE, EMCARE, HMIC, Medline, PsycINFO, Web of Science and Google Scholar. Publications dated between January 2000 and September 2020 were included in the search. Boolean operators were used to construct and combine searches, and the search term ‘NOT’ was used to ensure articles which may use the search term ultrasound for guided biopsy rather than diagnostic procedures were excluded given the extensive use of this search term within literature. Table 1 summarises the search terms used. The results were filtered to include only English language abstracts. Little evidence was expected to be found during the search; therefore, no restrictions were placed upon type of articles included.

Table 1.

Search terms.

Population	Exposure		Outcomes
Patients with prostate cancer	Diagnostic ultrasound imaging		Active surveillance
1st term	2nd term	3rd term	4th term
prostate cancer OR cancer of the prostateOR prostate gland cancerOR prostate tumo* OR Prostat* neoplasm	diagnostic imaging ORUltrasou* OR UltrasonographyORsonograp* OR sonogram ORultrasound diagnosisOR ultrasound scanningOR trans-rectal ultrasou*OR TRUS	NOTBiops*	active surveillanceOR watchful waitingORexpectant management

Study selection

A preferred reporting item for systematic review and meta-analysis (PRISMA) protocol flow process was followed to identify relevant publications²⁸ (Figure 1). On identification of potentially relevant titles, the selected articles were independently screened by three reviewers with each full-text article or abstract assessed for relevance using the predetermined inclusion and exclusion criteria described in Table 2.

Figure 1.

Flow diagram of the process used to select studies for systematic review. Ten databases were systematically searched: AMED, BNI, CINAHL, EMBASE, EMCARE, HMIC, Medline, PsycINFO, Web of Science and Google Scholar.

Table 2.

Inclusion and exclusion criteria for study eligibility.

Inclusion criteria	Exclusion criteria
Any article related to the active surveillance of prostate cancerandAny article related to the use of ultrasound in the diagnosis or monitoring of prostate cancerAny article with evidence of clinical utility of ultrasound in AS	Conference abstracts which do not provide information suitable for extractionAbstracts with subsequent full articles and/or articles containing duplication of dataArticles describing the monitoring of PCa via other modalities; for example, MRIArticles using US in PCa patients but for reasons other than diagnosis or monitoring; for example, to guide surgical planning or to guide biopsiesArticles related to the use of diagnostic US for men high risk of PCaUntraceable or unrelated articlesCommentaries, editorials or opinion articlesArticles with no evidence of clinical utility (unless related to outcomes of interest within this review) Studies in languages other than English

Inclusion criteria

Exclusion criteria

Any article related to the active surveillance of prostate cancerandAny article related to the use of ultrasound in the diagnosis or monitoring of prostate cancerAny article with evidence of clinical utility of ultrasound in AS

Conference abstracts which do not provide information suitable for extractionAbstracts with subsequent full articles and/or articles containing duplication of dataArticles describing the monitoring of PCa via other modalities; for example, MRIArticles using US in PCa patients but for reasons other than diagnosis or monitoring; for example, to guide surgical planning or to guide biopsiesArticles related to the use of diagnostic US for men high risk of PCaUntraceable or unrelated articlesCommentaries, editorials or opinion articlesArticles with no evidence of clinical utility (unless related to outcomes of interest within this review) Studies in languages other than English

Whilst conference abstracts were included within the literature search, these were excluded from the review if the methodology was of a poor quality or if the data were of insufficient detail to confidently judge quality.

Quality assessment

To evaluate the robustness of the eligible papers, a risk of bias assessment was performed according to the Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2) tool.²⁹ A standardised form was utilised to extract the following data from each eligible article:

year of publication

type of study described

the cohort size

the criteria used to describe clinically significant PCa

the criteria used to describe disease progression

the modality of imaging used within the study

the mode of US imaging used within the study

Given the limited number of full-text original research articles available for review, and level of reporting of data, a meta-analysis of findings was not possible. A narrative review of the data was completed.

Results

The PRISMA diagram²⁸ was used to summarise the literature search (Figure 1). The narrow range of evidence sourced is an indication that diagnostic US is not in widespread use within AS protocols and few randomised control studies or peer reviewed multicentre trials exist. The articles included describe the clinical utility of the use of US in AS.

Evidence synthesis

The initial search identified 2094 titles with a further two papers found by forward backward reference searching. Duplicates were identified and excluded. This yielded 152 potentially eligible articles. Of these, 106 articles were excluded on primary screening because US was used solely to guide biopsy procedures. The remaining 46 published studies were reviewed for eligibility; 23 available articles were reviewed by reviewers 1 and 2 and 23 available articles reviewed by reviewer 1 and 3.

Inter-reviewer agreement

The inter-reviewer agreement was assessed using the Cohen k coefficient. Between reviewers 1 and 2, the Cohen k coefficient was 0.82 and between reviewers 1 and 3, 0.75. Despite the substantial agreement between reviewers 1 and 3, there were six publications in which these reviewers disagreed about their inclusion. A second review of these publications was made by reviewer 2 and a consensus agreement made.

Quality assessment

A total of 12 eligible articles were taken forward to the quality assessment stage. These were scored using the defined matrix and agreed parameters within the QUADAS-2 tool.²⁹ Three articles discuss the technology of US and its use in the diagnosis of PCa, rather than the role of US specifically in AS^23,30,31 but were included as the potential role of US could be inferred from the data extracted from these publications.

From the QA, seven publications had a low concern of applicability for inclusion,^{7,13,21,32–35} two had a moderate concern^30,36 and two had a high concern of applicability for inclusion^23,31 within the review. The characteristics of the studies were reviewed to assess the study inclusion criteria and study design, how disease was determined, including identification of disease progression if applicable and the modality and mode of imaging used. Study design, patient and inclusion characteristics for each published study are presented in Table 3. Disease and imaging characteristics are presented in Table 4.

Table 3.

Study, patient, and inclusion characteristics

Author and Year	Study Design	Number of patients	Inclusion criteria	Study cohort inc. AS
Eure et al (2019)⁷	Prospective	9	Age ≥40 <80 years; Indication of biopsy required under AS protocolPSA ≤20 ng/mL Clinical stage <c T2c	Yes
Hruby et al (2001)¹³	Retrospective	180	PSA ≤15 ng/mLClinical stage T1b – T2N0M0Biopsy Gleason score ≤ 7	Yes
Staerman (2019)²¹	Prospective (Conference Abstract)	44	Biopsy proven PCa with Gleason score = 3 + 3	Yes
Press et al (2019)²³	Prospective	672	Patients who had an MRI-US fusion targeted biopsyTRUS grade possibleStandard MRI performedMRI suspicion score of > 1 (findings suspicious of PCa)	No
Murciano-Goroff et al (2014)³⁰	Prospective	70	Prostate Volume ≥ 20 ≤ 60 mLPSAD**** ≤ 0.15 ng/mL/cc	Yes
Perez (2019)³¹	Prospective (Conference Abstract)	55	PSA between 4.2 – 40 ng/mL	Yes
Eltemamy et al (2016)³²	Retrospective	875	PSA ≤20 ng/mLClinical stage** ≤ T2Biopsy Gleason score ≤ 3 + 4	Yes
Maffei et al (2011)³³	Prospective (Conference Abstract)	118	Indication of biopsy required under PRIAS*** protocol for AS of low risk cancer	Yes
Sauvain et al (2013)³⁴	Prospective	243	PSA ≤ 10 ng/MlNormal DRE	Yes
Shoji et al (2016)³⁵	Prospective	50	Age ≥ 40 < 80 yearsLife expectancy of >10 yearsPSA ≤20 ng/mL	Yes
Ko et al (2011)³⁶	Retrospective (Conference Abstract)	253	Patients in an active surveillance pathway	Yes
Weiss et al (2013)³⁷	Retrospective	756	Localised PCa treated with radical prostatectomyTRUS and endorectal MRI performed	Yes

*PSA – prostate specific antigen

** TNM staging system – T = tumour; N = node involvement and M = metastastic spread

*** PRIAS - Active Surveillance for Low-Risk Prostate Cancer Worldwide: The PRIAS Study (37)

**** PSAD - prostate specific antigen density = PSA / prostate volume

***** DRE – digital rectal examination

Table 4.

Definition of disease, imaging modality and mode characteristics

Author and Year	Definition of disease on imaging (or progression if applicable)	Imaging modality & mode utilised
Eure et al (2019)⁷	Lesions identified on imaging warranting biopsyGleason upgrade following biopsy	MpMRI TRUSStandard frequency B-ModeHigh frequency B-Mode (29MHz)
Hruby et al (2001)¹³	New or enlarging hypoechoic peripheral zone noduleIncrease in gland volume of ≥ 30%	TRUS - Standard frequency B-Mode
Staerman (2019)²¹	Suspicious areas	mpMRITRUS - High frequency B-Mode (29MHz)
Press et al (2019)²³	Presence of a demarcated or poorly demarcated hypoechoic ROI*** warranting biopsy	MpMRITRUS - Standard frequency B-Mode
Murciano-Goroff et al (2014)³⁰	n/a	MRI with endorectal coilTRUS - Standard frequency B-Mode
Perez (2019)³¹	Lesions identified on imaging warranting biopsy	MpMRITRUS - High frequency B-Mode (29MHz)
Eltemamy et al (2016)³²	Increase in: number of hypoechoic lesions demonstrated≥ 50% increase in lesion volume Gleason upgrade following biopsy	TRUS* - Standard frequency B-Mode
Maffei et al (2011)³³	Prostate volumeChanges in the PSAD	TRUS - Standard frequency B-Mode
Sauvain et al (2013)³⁴	TRUS - Suspicious lesions described as : hypoechogenic, weakly hypoechogenic or subtle, heterogenicFocal if under 5 mmNodular if over 5 mmPDUS**** - Suspicious lesions described as: hypoechogenic avascular lesion orweakly hypoechogenic or hypervascularised lesion orhypervascularised heterogenous or hypervascularised hypoechogenic lesion	TRUSPDUS
Shoji et al (2016)³⁵	TRUS: Diameter of suspicious lesionRate of change of the suspicious lesionDoppler grading (comparison with the normal anatomical blood supply (NABS) in the other unsuspicious parts of the prostate 0: no or 1: low - flow present but weaker or less than the NABS. Downgrade2: moderate: flow equal or similar to the NABS. Stable3: high: flow stronger or greater than the NABS. Upgrade	TRUSDoppler (not specified if colour or power)
Ko et al. (2011)³⁶	Prostate volumeChanges in the PSAD	TRUS - Standard frequency B-Mode
Weiss et al (2013)³⁷	n/a	MRI TRUS - Standard frequency B-Mode

*TRUS – Transrectal ultrasound

**PRIMUS - prostate risk identification using micro-ultrasound

***ROI – Region of interest

****PDUS – power Doppler ultrasound

The study characteristics are widely varied; in six studies, the prostate-specific antigen (PSA) was an indicator for inclusion;^{7,13,31,32,34,35} in four studies, the Gleason score provided an inclusion criterion;^13,21,30,32 as did the clinical stage and whether radical treatment was planned in three other studies.^7,13,32 Other inclusion criteria included patient’s age, the outcome of the digital rectal examination (DRE) and the prostate volume. Overall, however, it is difficult to draw direct comparisons between publications due to the heterogeneity of sample inclusion criteria applied. The target condition defined for this review is the histopathology-proven progression or upgrade of PCa. In all studies, the cohort of patients had histologically-proven PCa.

Imaging: Identifying disease and progression

Each of the publications was reviewed to determine how disease was defined and to ascertain the imaging modality and mode employed to identify disease, identify disease progression, or monitor disease. Table 4 describes the disease characteristics described within the publications.

Whilst MRI was a ubiquitous imaging modality within the studies, the use and mode of US varied between publications. Most commonly, in eight studies, standard transrectal ultrasound (TRUS) was utilised;^{13,23,30,32,34–36,37} three studies investigated micro-US^21,31,33 and one study employed both standard and micro-US.⁷ Defining or identifying disease with the differing US modes, however, is less varied. Lesion identification, using either standard TRUS or micro-US, was the primary diagnostic feature in all studies. The authors used features inherent in US, such as changes to the echogenicity of the suspected lesions, changes to the size of lesions or whether lesions were well demarcated. Identifying changes on US is challenging in all studies, but data suggest that micro-US has improved sensitivity and specificity over standard TRUS.⁷

The next most frequent US mode employed within the studies reviewed was Doppler to assess the perfusion of the prostate and suspected lesions.^34,35 Doppler assessment is an operator-dependent technique and, as Sauvain et al.³⁴ identified, transducer pressure may compress microvasculature within small peripheral lesions, thereby compromising the accuracy of this mode of imaging. However, Shoji et al.³⁵ found that an upgrade in the Doppler signature of a lesion was a significant risk factor for biopsy-proven disease progression.

Sensitivity and specificity

To gain a better understanding of the sensitivity and specificity of each US mode in the detection of clinically significant PCa, data were extracted from each study where there was a confirmatory gold standard documented. The gold standard was either histopathological confirmation of PCa or, as in the two papers evaluating the reliability of prostate volume calculations with TRUS, MRI.^30,36 An overview of the study outcomes including the sensitivity and specificity of each imaging mode, where given, is documented in Table 5.

Table 5.

Sensitivity & specificity of US modes utilised within each study

Author and Year	Ultrasound mode utilised	Study outcomes &/or Sensitivity and Specificity of PCa detection
Eure et al (2019)⁷	Standard frequency B-Mode^AHigh frequency B-Mode ^B	^A Sensitivity 11%; specificity 93%^B Sensitivity 89%; specificity 45%
Hruby et al (2001)¹³	TRUS - Standard frequency B-Mode	25% of cohort with biopsy proven progression had changes on TRUS. 3.5% had an increase in volume size.
Staerman (2019)²¹	High frequency B-Mode	Overall cancer sensitivity 84%
Press et al (2019)²³	TRUS - Standard frequency B-Mode	The overall cancer detection rates were 46.2%, 58.6% and 76.0% for USG 0, 1 and 2, respectively (P < 0.001). A well-defined ROI had a significantly higher risk of a diagnosis of GS ≥7 PCa
Murciano-Goroff et al (2014)³⁰	TRUS - Standard frequency B-ModeMRI	Standard TRUS & MRI prostate volume calculations appear to overestimate prostate volume by 9.34% & 16.57% respectively
Perez (2019)³¹	TRUS - high frequency B-Mode^cmpMRI^d	^c Sensitivity 93.3%; specificity 27.5%^d Sensitivity 86.7%; specificity 40.0%
Eltemamy et al (2016)³²	Standard frequency B-Mode	Evidence of TRUS progression: 49% agreement at biopsy No evidence of TRUS progression: 66% agreement at biopsy; TRUS independently associated with biopsy upgrade of disease (OR 1.8, 95% CI 1.3 – 2.5, p <0.01)
Maffei et al (2011)³³	TRUS - Standard frequency B-Mode	Coefficient of Variation calculated with Pearson correlation coefficient of 0.04. Variability in volume measurements not adversely affect PSAD calculations.
Sauvain et al (2013)³⁴	PDUS	Sensitivity: 30% probability of having a significant cancer if PDS abnormal (However, 57% of all abnormal PDS had biopsy positive of all cancer grades inc insignificant PCa); Specificity 96%
Shoji et al (2016)³⁵	TRUSDoppler	Multivariate analyses demonstrate significant predictors of progression as follows: increase (≥25 %) in major axis diameter of lesion (hazard ratio, 6.672; 95 % CI, 1.097– 40.508; p = 0.022) upgrade of Doppler signature (hazard ratio, 4.091; 95 % CI, 1.673–24.878; p = 0.039)
Ko et al (2011)³⁶	TRUS - Standard frequency B-Mode	Coefficient of Variation calculated with Pearson correlation coefficient of 0.04. Variability in volume measurements not adversely affect PSAD calculations.
Weiss et al (2013)³⁷	Standard frequency B-Mode	Average prostate size measured with TRUS and eMRI correlated significantly with one another (R 0.801; P 0.0001), demonstrating a strong linear relationship (y0.891 x+2.622, R² 0.642).

Of the 12 articles, reliable sensitivity and specificity data could be extracted from only six.^{7,21,23,31–34} Sensitivity of over 80% was reported in four studies using micro-US (range 84%–93.3%)^7,21,31,33 and one study using standard TRUS. Press et al.²³ identified changes in US appearance of regions of interest (ROIs) as an indicator of disease progression with sensitivity increasing to a maximum of 76% in lesions that are well demarcated and of a widely different echogenicity to the background prostate. Equivocal sensitivity was reported by Eltemamy et al.³² with only 49% of biopsy-proven progression being identified on TRUS imaging. A far poorer sensitivity was inferred in the article by Hruby et al.¹³ with only 25% of the study cohort with biopsy-proven progression demonstrating any changes on TRUS, although it is acknowledged that this study was performed using now out-dated technology; a repeat of this study, with up to date imaging, may yield improved results.

Specificity is reported in only three studies^7,31,33 (range 27.5%–45%) and all used micro-US. Eltemamy et al.³² demonstrated a reasonable agreement between the TRUS findings and the histopathology results post-biopsy of 66% where there was no evidence of disease progression. No comparable sensitivity and sensitivity data could be extracted from the articles related to the assessment of prostate volume,^36,37 although both of these articles have argued that US measurements are comparable with MRI and can be used to assess disease progression. Hruby et al.¹³ however, had extremely poor results when using prostate volume to assess progression with only 3.5% of the cohort demonstrating any significant change. Whilst appearances on imaging may change with updated technology, volume measurements are an inherent calculation in any machine and, therefore, a degree of caution is required when using prostate gland volume measurement alone to assess progression. Shoji et al.³⁵ demonstrated that a change in ROI size, rather than overall gland volume, as well as changes to the Doppler signature, could both be used as a predictor of disease progression. This supports the findings by Sauvain et al.³⁴ that indicated a normal power Doppler signature was a strong indicator of the absence of clinically significant PCa. Combining changing US features of prostate appearance, ROI appearance, gland and ROI volumes and Doppler signature all suggest that mpUS has a role in the diagnostic imaging of patients on an AS pathway.

Discussion

TRUS

Despite the poor results of standard TRUS reported by Hruby et al.¹³ the study by Eltemamy et al.³² reported TRUS progression criteria in terms of lesion size and site and concluded that stable TRUS findings may allow for increased intervals between biopsy for men on AS. Press et al.²³ demonstrated an increasing sensitivity for disease detection when there were greater changes in appearance of an ROI compared to the background gland echogenicity. Prostates with hypoechoic regions visible on TRUS were reported to experience worse oncological outcomes than men without, suggesting a correlation of ultrasonography ﬁndings with disease aggressiveness.²³ The findings of both Eltemamy et al.³² and Press et al.²³ demonstrate the potential use of TRUS to assist in the risk stratiﬁcation prior to biopsy among men with stable TRUS and/or low suspicion MRI ﬁndings. Men with low or equivocal MRI suspicion could potentially avoid biopsy if no discernible ROI is present on imaging using TRUS.²³ There are limitations to this study in that US ROIs not seen on MRI were not included or scrutinised. However, the purpose of this large cohort prospective study was to compare mpMRI lesions, and the findings do indicate a role for TRUS in this setting. In addition, only lesions with decreased echogenicity on US were considered for review despite previous papers^38,39 describing other US features indicating suspicious prostatic lesions in high-grade disease and which were subsequently correlated with radical prostatectomy. Not all lesions can be identified with ultrasound; some histologically proven lesions are known to be isoechoic to the surrounding prostate³⁸ and as such, operators would need to proceed with caution if B-mode US features alone were to be considered as diagnostic.

Micro-ultrasound

Despite the inherent limitations, US imaging, in particular micro-US, can detect clinically significant PCa.^15,22 The sensitivity for disease detection using micro-US ranged from 84% to 93.3%.^7,21,31,33 Indeed, Maffei et al.³³ upgraded the Gleason Score to ≥ 7 cancer in 31 patients using micro-US findings alone to indicate the site of confirmatory biopsy. The study by Staerman²¹ demonstrated a good concordance rate between mpMRI and micro-US in the identification of suspicious lesions and benign findings in biopsy-proven normal prostate. Micro-US has a reduced depth of field of imaging which hinders assessment of the whole gland, although technological improvements may reduce the negative impact of this fundamental ultrasound limitation. Despite this limitation, the findings of our review indicate that whilst TRUS can identify suspicious ROIs, particularly as disease grading increases, micro-US is an emerging technique that is comparable with MRI. However, this review has also identified that there is only a small number of researchers publishing in this field and few full-text papers are available for analysis, meaning potential difficulties with this new technology cannot be readily assessed. Further multicentre trials of this emerging technology are indicated.

Lesion detection

Being able to identify ROIs within a prostate is an important feature of any imaging mode, be that standard TRUS or micro-US. The key consideration for the use of imaging in AS is the ability to demonstrate longitudinal change of quantitative image-related variables that can be used confidently as a clinical predictor for disease progression, such as diameter or volume of an ROI or the Doppler signature.³⁵ Shoji et al.³⁵ describe a high specificity associated with Doppler signature when analysed over the duration of the patient’s follow-up, although they found two major limitations. The first was the limitation of visibility of very low-volume cancer, rendering Doppler imaging difficult to assess. The second was that the threshold volume of clinically significant cancer that could be identified was 0.5 mL; therefore, small but clinically significant cancer may be missed or underestimated by imaging. Volume may be calculated using either the three dimensions with the standard semi-ellipsoid formula inherent within ultrasound machines or using planimetry as used in MRI. Calliper placement for any calculation requires clear demarcation of the tissue-tumour boundary which ultimately depends on clear grey-scale visibility on ultrasound. This inherent limitation of ultrasound may render volume calculations unreliable. However, volume calculations have been investigated and good correlation with gold standard MRI demonstrated. Murciano-Goroff et al.³⁰ identified the optimum parameters for prostate volume calculations as different methods produced disparate volumes. Ko et al.³⁶ found that in 95% of their cohort, variability in TRUS-guided prostate volume measurement did not affect PSAD calculations sufﬁciently to affect management. The study by Weiss et al.³⁷ demonstrated a high degree of correlation between TRUS- and MRI-based prostate volumes. The authors concluded that given the high degree of accuracy and reproducibility, in the hands of an experienced sonographer, a TRUS-based examination is a reasonable modality for estimating prostate size in all patients with disease of the prostate. In patients with biopsy-proven disease progression, 15% had changes to the ROI volume as the sole indicator of progression on TRUS.³² Whilst the volume of the ROI is limited by visibility and a finite size, the evidence suggests that the use of this, as an US parameter to monitor disease progression, is clearly beneficial on an AS pathway. However, it is noted that the lesion volume size in these studies were compared to lesion volume on mpMRI and evidence to support direct validation with histologically proven volumes was not found during this literature search.

Perfusion characteristics

Perfusion characteristics of the ROI are a good indicator of normality.³⁴ Sauvain et al.³⁴ identified that a normal power Doppler signature (PDS) was associated with a 96% probability of not having a high-risk cancer. They concluded that a normal PDS may be used to delay biopsy in patients with low risk disease. Doppler grading, assessed as a longitudinal variable, also had a positive bearing on assessment of disease progression in the study by Shoji et al.³⁵ A limitation of any Doppler technique is its reliance on operator performance. Unfortunately, transducer pressure may compress microvasculature within small peripheral lesions, which may lead to misinterpretation of the Doppler signature.³⁵ A further consideration is the qualitative assessment of the Doppler signature. Shoji et al.³⁵ used a subjective Doppler grade of blood flow signal within the lesion and classified from grade 0 to 3. Quantitative assessment of perfusion using solely Doppler has demonstrated that the mean speed of coloured pixels and speed-weighted pixel density are good discriminators for PCa in periurethral and the peripheral regions,⁴⁰ although this technique has not been widely tested nor utilised in an AS pathway. Technological advances in the identification of microvasculature are now available on modern ultrasound machines but remain under-investigated in the field of prostate cancer. This technique is worthy of consideration in future research given that no published studies have been identified in this systematic review and may contribute to the understanding of perfusion in prostate cancer progression.

Perfusion can also be evaluated using contrast-enhanced ultrasound (CEUS). Whilst good results have been demonstrated for CEUS of lesions 0.5 mL or larger, (sensitivity of 58–69% and specificity of 93–95%),²⁶ a study by Qi et al.⁴¹ found that three section CEUS was able to detect 92.3% of patients with cancer, whereas standard TRUS identified only 70.7%. The cohort for both studies, however, all had clinically significant cancer (Gleeson ≥ 7) and had subsequent prostatectomy. Techniques quantifying perfusion using contrast are available but have not been investigated in the assessment of prostate cancer under active surveillance. The role of this has not been evaluated within this review, given the absence of published date. There is, indeed, an absence of evidence for the use of CEUS in an AS pathway and, given its limitations of scan duration and contrast agent cost (£92.00 per scan 2019/2020 NHS tariff),⁴² it is unlikely to be considered as a viable imaging alternative to mpMRI for patients on AS. However, the strong evidence presented by Sauvain et al.³⁴ and Shoji et al.³⁵ suggests that perfusion of the prostate and/or ROI should be considered as part of the mpUS assessment.

Elastography

An US imaging mode that has been trialled in the diagnosis of PCa is elastography, utilising both strain and shear wave techniques. Zhai et al.⁴³ performed a rigorous study of consecutive patients scanned prior to prostatectomy specifically looking at acoustic radiation force impulse (ARFI). The authors noted that the bilateral stiffness asymmetry created by PCa in ARFI images may provide a convenient means to identify suspicious malignancy in the prostate by being able to differentiate between stiff abnormal tissue and stiffness caused by chronic benign prostatic calcifications that in strain elastography (SE) create false-positive findings. A further study by Pelzer et al.⁴⁴ comparing SE with endorectal MRI (eMRI) demonstrated that SE detected PCa in 46 of the 50 positive cancer cohort (92%), whilst eMRI detected PCa in 42 (84%). SE was found to be more sensitive in the apical and mid prostate; eMRI more sensitive in the base (peripheral zones) and transitional zone. In both studies, the patients all had known clinically significant PCa. The limitations of both studies, and with the technique generally, are the uncertainty as to how ARFI or SE can assist in AS when the elastography findings in a normal prostate or in the presence of benign disease are unknown.

Strengths and limitations

This systematic review is the first specifically evaluating the role of all modes of US within an AS pathway. A key strength in of this study is the quality assessment, systematic approach to searching and the robust review of the literature available by a team of reviewers. The eligibility of articles was rigorously reviewed by three independent reviewers which resulted in 12 publications eligible for inclusion. Of these, only eight full-text articles were available despite the authors of the published abstracts being contacted to determine if their data were available as a journal publication. The paucity of full-text articles is an indication of the limited current use of US within an AS pathway. Indeed, the variability of inclusion criteria between studies reflects the lack of consistency of practice within this clinical field. This variation is a challenge for clinicians and patients when planning care pathways and when radical intervention is required.⁴ The review has demonstrated that micro-US offers promise as an imaging tool comparable with mpMRI in AS with five^{7,21,23,31,33} of the 12 papers included related to studies assessing the utility of this modality. However, this review has relied on data related to this emerging technique extracted from conference abstracts with only one full-text paper being eligible for inclusion. Bias could potentially be introduced into the review findings, although the authors acknowledge the lack of published trials. The small cohort of researchers publishing in this field limits the scope of papers available. Literature searching for new and updated relevant publications will continue.

A second limitation of this review is the variability of the modes of US imaging used within the included studies. A meta-analysis was not possible because of the wide heterogeneity across the studies which reduced the comparative data eligible for review. It has been difficult to combine the results of the varied studies to produce a generalisation for clinical practice. Fusion-guided ultrasound imaging has a significant role to play in prostate biopsy following mpMRI and may aid targeted follow-up.⁴⁵ Fusion-guided imaging has not been included in this systematic review, as there have been no studies evaluating its use in an active surveillance programme. This remains an area for future research and evaluation. However, this review has indicated that US does have a role in monitoring disease progression provided a multiparametric approach, which includes prostate volume, lesion volume, lesion demarcation, lesion echogenicity and an assessment of the Doppler signature as baseline imaging modes, is utilised. Micro-US shows promise despite the limitations of the publications included. Its use in the routine clinical setting remains uncertain and the technique, including the confidence of image interpretation, will require careful evaluation to fully understand its usefulness and acceptability.

Context of clinical practice

There are significant capacity issues within the National Health Service (NHS) in the UK⁴⁶ that limits access to MRI. Waiting times published in the NHS England January 2020 diagnostic dataset⁴⁶ document a duration from MRI referral received to examination performed of between 19 and 25 days. Priority is given to urgent cancer diagnosis over follow-up in many institutions, and therefore the ability to fulfil the NICE criteria, ⁶ where the evidence for such practice continues to be updated,³ is likely to be limited. There is, however, less of a capacity issue within alternative imaging modalities, such as in diagnostic US, and any evidence that would support the safe transfer of diagnostic imaging from MRI would be welcome at a performance level. Indeed, waiting times for US in England are, on average, 10 days.⁴⁶ Whilst capacity for diagnostic imaging is already at a premium, the full impact of the 2020 coronavirus pandemic has yet to be fully realised, but it is certain to have resulted in significant delays for diagnosis and hindered any imaging required on a surveillance pathway.⁴⁷ There is mounting evidence of the impact of COVID-19 on the PCa pathway with delays to diagnosis being reported in many of the UK cancer alliances,⁴⁸ and this is an issue that is likely to be repeated worldwide. The challenges of providing healthcare to the level required, and comparable with pre-COVID-19 delivery, will be faced for many years to come, although significant interruption to normal practice does provide an opportunity for services to be redesigned and a restart button be pushed.⁴⁹ Now is the time to review existing pathways and seek alternatives that are safe for patients, sustainable for future delivery and release capacity in the high demand modalities, such as MRI. Randomised control trials are essential if this technique is to be embedded into everyday clinical practice and US offered as a much needed, viable alternative to mpMRI for patients on AS.

Conclusion

A range of publications has been evaluated and has had to include conference abstracts due to the paucity of data available leading to a limitation of this systematic review. However, it has demonstrated that there may be a role for mpUS for patients with known PCa within an active surveillance pathway despite the evidence being less robust than the evidence for the use of mpMRI. Although not all lesions can be identified on ultrasound, an imaging protocol that combines prostate volume, lesion volume, lesion demarcation, lesion echogenicity and an assessment of the Doppler signature will provide a reasonable sensitivity and specificity but any such model will need testing prior to implementation. Whilst the data demonstrate that standard TRUS is not comparable with mpMRI, it does indicate that, providing an mpUS approach is utilised, stable TRUS findings may allow for increased intervals between biopsy for men on AS. The advent of micro-US, with its reported sensitivity in the range of 84%–93.3%, offers more promise for a truly comparable imaging modality to relieve capacity issues within MRI. Further research is needed to optimise and evaluate mpUS and micro-US for the monitoring of patients with low risk PCa.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics Approval

Ethical approval is not required for this systematic review.

Guarantor

PP.

Contributorship

PP, MT and PW researched literature and conceived the study. PP and TP performed the initial review of literature. PP, MT and PW reviewed the literature for eligibility for inclusion in the systematic review. PP wrote the first draft of the article. All authors reviewed and edited the article and approved the final version of the article.

Acknowledgments

We would like to thank Trevor Parker for his assistance and guidance in this systematic review.

ORCID iD

Pamela Parker

References

Klotz

Vesprini

Sethukavalan

, et al. Long-term follow-up of a large active surveillance cohort of patients with prostate cancer. J Clin Oncol 2015; 33: 272–U75.

Da Silva

Cagiannos

Lavallée

, et al. An assessment of Prostate Cancer Research International: active surveillance (PRIAS) criteria for active surveillance of clinically low-risk prostate cancer patients. Can Urol Assoc J 2017; 11: 238.

Merriel

SWD

Hetherington

Seggie

, et al. Best practice in active surveillance for men with prostate cancer: a Prostate Cancer UK consensus statement. Bju Int 2019; 124: 47–54.

Bruinsma

Zhang

Roobol

, et al. The Movember Foundation's GAP3 cohort: a profile of the largest global prostate cancer active surveillance database to date. Bju Int 2018; 121: 737–744.

Brown

Ahmed

Faria

, et al. Multiparametric MRI to improve detection of prostate cancer compared with transrectal ultrasound-guided prostate biopsy alone: the PROMIS study. Health Technol Assess 2018; 22: 1–176.

NICE. Prostate cancer: diagnosis and management NICE guideline 131. UK: National Institute for Health and Care Excellence, 2019.

Eure

Fanney

Lin

, et al. Comparison of conventional transrectal ultrasound, magnetic resonance imaging, and micro-ultrasound for visualizing prostate cancer in an active surveillance population: a feasibility study. Can Urol Assoc J 2019; 13: E70--E77.

Vargas

Hötker

Goldman

, et al. Updated prostate imaging reporting and data system (PIRADS v2) recommendations for the detection of clinically significant prostate cancer using multiparametric MRI: critical evaluation using whole-mount pathology as standard of reference. Eur Radiol 2016; 26: 1606–1612.

Branger

Maubon

Traumann

, et al. Is negative multiparametric magnetic resonance imaging really able to exclude significant prostate cancer? The real-life experience. 2016; 119: 449--445.

10.

Klotz

Pond

Loblaw

, et al. Randomized study of systematic biopsy versus magnetic resonance imaging and targeted and systematic biopsy in men on active surveillance (ASIST): 2-year postbiopsy follow-up. Eur Urol 2020; 77: 311–317.

11.

Glass

Dall'Era

MA.

Use of multiparametric magnetic resonance imaging in prostate cancer active surveillance.

Bju Int 2019; 124: 730–737.

12.

RCR. Magnetic Resonance Imaging (MRI) equipment, operations and planning in the NHS. London: The Royal College of Radiologists, www.rcr.ac.uk/sites/default/files/cib_mri_equipment_report.pdf (2017, accessed 7 February 2021).

13.

Hruby

Choo

Klotz

, et al. The role of serial transrectal ultrasonography in a ‘watchful waiting' protocol for men with localized prostate cancer. Bju Int 2001; 87: 643–647.

14.

Maffei

Paciotti

Domanico

, et al . Diagnostic performance of micro-ultrasound in patients under active surveillance for low-risk prostate cancer. Eur Urol Suppl 2019; 18: e3419.

15.

Bevilacqua

Maffei

Paciotti

, et al . Diagnostic performance of micro-ultrasound in a contemporary cohort of patients in active surveillance for localized prostate cancer: a single-institutional experience. Eur Urol Suppl 2019; 18: e3155--e3376.

16.

Zhang

Wang

, et al. Micro-ultrasound imaging for accuracy of diagnosis in clinically significant prostate cancer: a meta-analysis. Front Oncol 2019; 9: 1368.

17.

Lockwood

Turnball

Christopher

, et al. Beyond 30 MHz [applications of high-frequency ultrasound imaging. IEEE Eng Med Biol Magz 1996; 15: 60–71.

18.

Ghai

Eure

Fradet

, et al. Assessing cancer risk on novel 29 MHz micro-ultrasound images of the prostate: creation of the micro-ultrasound protocol for prostate risk identification. J Urol 2016; 196: 562–569.

19.

Bates

Kostakopoulos

Ayers

, et al. A narrative overview of active surveillance for clinically localised prostate cancer. Semin Oncol Nurs 2020; 36: 151045.

20.

Iremashvili

Manoharan

Kava

, et al. Predictive models and risk of biopsy progression in active surveillance patients. Urol Oncol 2017; 35: 37.e1–e8.

21.

Staerman

The utility of 29 MHz high resolution micro-ultrasound and mpMRI in the management of Gleason six prostate cancer with active surveillance. Eur Urol, Suppl 2019; 18: e3502.

22.

22. Lughezzani

Pereira

Sanchez

, et al. Comparison of micro-ultrasound and multiparametric MRI imaging for prostate cancer: a multicentre prospective analysis. 2020; 203: e806.

23.

Press

Rosenkrantz

Huang

, et al. The ultrasound characteristics of regions identified as suspicious by magnetic resonance imaging (MRI) predict the likelihood of clinically significant cancer on MRI-ultrasound fusion-targeted biopsy. BJU Int 2019; 123: 439–446.

24.

Mischi

Kuenen

MPJ

Beerlage

, et al. Prostate cancer localization by contrast-ultrasound-dispersion imaging: results from a pilot study. Eur Urol, Suppl 2014; 13: e1054--e1054a.

25.

Boehm

Salomon

Beyer

, et al. Shear wave elastography for localization of prostate cancer lesions and assessment of elasticity thresholds: implications for targeted biopsies and active surveillance protocols. J Urol 2015; 193: 794–800.

26.

Postema

Mischi

de la Rosette

, et al. Multiparametric ultrasound in the detection of prostate cancer: a systematic review. World J Urol 2015; 33: 1651–1659.

27.

NIHR. PROSPERO. International prospective register of systematic reviews. National Institute of Health Research, www.crd.york.ac.uk/PROSPERO/ (2020, accessed 7 February 2021).

28.

Moher

Liberati

Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097.

29.

Whiting

Rutjes

AWS

Westwood

, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011; 155: 529–536.

30.

Murciano-Goroff

Wolfsberger

Parekh

, et al. Variability in MRI vs. ultrasound measures of prostate volume and its impact on treatment recommendations for favorable-risk prostate cancer patients: a case series. Radiat Oncol 2014; 9: 200.

31.

Perez

TVR.

Initial results comparing high resolution micro-ultrasound with multiparametric magnetic resonance imaging for prostate cancer detection. Eur Urol, Suppl 2019; 18: e3499.

32.

Eltemamy

Leapman

Cowan

, et al. Serial anatomic prostate ultrasound imaging during prostate cancer active surveillance. J Urol 2016; 195.

33.

Maffei

Staerman

Paciotti

, et al. The utility of micro-ultrasound in patients under active surveillance for low-risk prostate cancer: a PIRAS study extension. J Urol 2020; 203: E339–E.

34.

Sauvain

Rohmer

Papavero

, et al. Value of transrectal power Doppler sonography in the detection of low-risk prostate cancers. Diagn Intervent Imaging 2013; 94: 60–67.

35.

Shoji

Ukimura

De Castro Abreu

, et al. Image-based monitoring of targeted biopsy-proven prostate cancer on active surveillance: 11-year experience. World J Urol 2016; 34: 221–217.

36.

Landis

Carter

, et al. 1632 effect of intra-observer variation in prostate volume calculations in prostate cancer active surveillance group. J Urol 2011; 185: e654–e5.

37.

Weiss

Wein

Malkowicz

, et al. Comparison of prostate volume measured by transrectal ultrasound and magnetic resonance imaging: is transrectal ultrasound suitable to determine which patients should undergo active surveillance?. Urol Oncol – Semin Original Investig 2013; 31: 1436–1440.

38.

Harvey

Pilcher

Richenberg

, et al. Applications of transrectal ultrasound in prostate cancer. Br J Radiol 2012; 85 Spec No 1: S3–17.

39.

Franiel

Teichgraber

Asbach

, et al. Prostate imaging – an update. RoFo Fortschritte auf dem Gebiet der Rontgenstrahlen und der Bildgebenden Verfahren 2015; 187: 751–759.

40.

Potdevin

Moskalik

Fowlkes

, et al. Doppler quantitative measures by region to discriminate prostate cancer. Ultrasound Med Biol 2001; 27: 1305–1310.

41.

Sun

, et al. Value of three-section contrast-enhanced transrectal ultrasonography in the detection of prostate cancer. J Clin Ultrasound 2017; 45: 304–309.

42.

NHSi. 2019/20 National Tariff Payment System: National Health Service Improvement (NHSi), https://improvement.nhs.uk/resources/national-tariff/ (2019, accessed 7 February 2021).

43.

Zhai

Rosenzweig

Palmeri

, et al. Acoustic radiation force impulse imaging of human prostates: initial in vivo demonstration. Ultrasound Med Biol 2012; 38: 50–61.

44.

Pelzer

Heinzelbecker

Weiß

, et al. Real-time sonoelastography compared to magnetic resonance imaging using four different modalities at 3.0T in the detection of prostate cancer: Strength and weaknesses. Eur J Radiol 2013; 82: 814–821.

45.

Kasivisvanathan

Rannikko

Borghi

, et al. MRI-targeted or standard biopsy for prostate-cancer diagnosis. N Engl J Med 2018; 378: 1767–1777

46.

NHSE. Diagnostic imaging dataset statistical release Jan 2019–Jan 2020. National Health Service England, www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2020/05/Provisional-Monthly-Diagnostic-Imaging-Dataset-Statistics-2020-05-21-1.pdf (2020, accessed 7 February 2021).

47.

Sikora

The potential impact of COVID-19 on cancer mortality in the UK: Oncology Central, www.oncology-central.com/in-focus/in-focus-coronavirus/the-potential-impact-of-covid-19-on-cancer-mortality-in-the-uk/ (2020, accessed 7 February 2021).

48.

Popert

Kum

Macaskill

, et al. Our first month of delivering the prostate cancer diagnostic pathway within the limitations of COVID-19 using local anaesthetic transperineal biopsy. BJU Int 2020; 126: 329--332.

49.

BIR. The “Radiology Reset Button” has been pressed!: British Institute of Radiology, www.mybir.org.uk/CPBase__event_detail?id=a173Y00000Eff4YQAR&site=a0N2000000COvFsEAL (2020, accessed 7 February 2021).