
==== Front
Curr Health Sci J
Curr Health Sci J
CHSJ
Current Health Sciences Journal
2067-0656
2069-4032
Medical University Publishing House Craiova

2023.3.10
10.12865/CHSJ.49.04.12
Original Paper
Should PI-RADS 3 be Subclassified According to ADC Values in the Transition Zone?
TEICA ROSSY VLADUT 1
CIOFIAC CRISTINA MIHAELA 1
FLORESCU LUCIAN MIHAI 2
GHEONEA IOANA-ANDREEA 2
1 School, University of Medicine and Pharmacy of Craiova, 200349 Craiova, Romania
2 Department of Radiology and Medical Imaging, University of Medicine and Pharmacy of Craiova, 200349 Craiova, Romania
Corresponding Author: : Cristina Mihaela Ciofiac Doctoral School University of Medicine and Pharmacy of Craiova 2 Petru Rareş Street, 200349 Craiova Romania ciofiacmcristina@gmail.com
Oct-Dec 2023
29 12 2023
49 4 564570
14 6 2023
12 10 2023
Copyright © 2023, Medical University Publishing House Craiova
2023
https://creativecommons.org/licenses/by-nc-sa/4.0/ This is an open-access article distributed under the terms of a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License, which permits unrestricted use, adaptation, distribution and reproduction in any medium, non-commercially, provided the new creations are licensed under identical terms as the original work and the original work is properly cited.
The premise of this paper starts from the fact that a more precise definition related to the intensity of the lesions on the ADC sequence can lead to a new subclassification of PI-RADS 3 lesions in the transitional zone and to an improvement of the specificity of the PI-RADS classification. The study was retrospective and included only patients who, based on prostate MRI examinations, contained exclusively PI-RADS 3 lesions, without other PI-RADS 4 or 5 lesions. The number of cases that meet all these conditions was reduced to 18, where a series of characteristics were noted for each one: PI-RADS 3 lesion area on the ADC sequence, the minimum and average ADC value of the lesion, the average ADC value of the transitional zone outside the lesion, PSA, prostatic volume, PSA density and biopsy result. The average ADC value of the negative lesions was 865(165) m2/sec, and of the positive ones was 869(118) m2/sec, which denies the hypothesis that there could be a value limit ADC to delimit PI-RADS 3 from PI-RADS 2 or 4 in this sample of patients. Furthermore, we reported the average ADC value of each such lesion to the adjacent unchanged transitional zone and obtained a greater difference of 432(163) m2/sec between the negative lesions and their adjacent transitional zone, compared to the difference of 399(127) m2/sec between the positive lesions and their adjacent ones.

Prostate
MRI
PIRADS 3
ADC
==== Body
pmcIntroduction

Prostate cancer remains the most frequent cancer in men worldwide [1, 2].

The Prostate Imaging Reporting and Data System (PI-RADS) is the structured reporting system for the assessment of magnetic resonance imaging (MRI) of the prostate.

It is divided into PI-RADS 1, 2 for low risk, 3 for medium risk, and 4.5 for high risk.

A lesion marked PI-RADS 1, 2, 4 or 5 is considered useful for the attending physician because the biopsy decision can be easily made in these cases, but PI-RADS 3 often brings ambiguity to this decision [3].

In the peripheral zone, a PI-RADS 3 lesion is characterized primarily on the diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) sequences, and secondarily on the DCE sequence where it has the possibility of being upgraded to PI-RADS 4, while in the transitional zone they are characterized primarily on the T2 weighted image (T2WI) sequence and then secondarily on the DWI and ADC sequences, without being influenced in any way by the dynamic contrast-enhanced (DCE) sequence, so the use of gadolinium does not bring benefits [4].

Although all sequences are exposed to signal noise in prostate MRI without an endorectal coil [5], it damages more the evaluation of the transition zone, where the reporting is done on the T2WI sequence and boils down to a greater number of characteristics (intensity, homogeneity, shape, edges, size) unlike the peripheral area where the ADC sequence complements DWI and the lesions are limited only to size and intensity.

In Figure 1 we can see how these artifacts look, but it is also a good example of the interpretation of the lesions because the peripheral lesion (blue arrow) looks similar to the transitional lesion (orange arrow) on the DWI sequence, but according to the PI-RADS 2.1 guide the peripheral lesion represents a very high risk (exceeds 15mm) and the transitional lesion represents an equivocal risk (if it does not exceed 15mm) because on the T2WI sequence that area shows no changes compared to the rest of the transitional area.

Figure 1 Prostate MRI examination showing motion artifacts due to rectal contents. The anxious patient could no longer resist inside the MRI, so these are the only T2WI (left) and DWI (right) sequences obtained. The T2WI sequence shows a prostate with relatively homogeneous changes, but the DWI sequence (confirmed by the ADC sequence, not included in the figure) clarifies the report by the presence of two suspicious areas present both in the peripheral area on the right side (blue arrow) and in the area previous transitions (orange arrow), although both sequences were artifactual

Also, according to the PI-RADS 2.1 guide, a lesion with equivocal risk in the transitional zone can also be described as a high or very high-risk lesion according to the DWI and ADC sequences, while it is of low risk on the T2WI sequence (Figure 2).

Figure 2 Classification of PI-RADS lesions in the transitional area [4].

In the transitional area, the number of PI-RADS 3 lesions is more than that of PI-RADS 4 and 5 combined [6, 7] so the problem of risk stratification arises because less than 20% of them are proven in studies on a large group of patients to be clinically significant cancer [6, 8, 9, 10].

The DWI sequence analyzes the property of water molecules to be in Brownian motion in normal fluids, and to be restricted in areas where there are cell membranes, due to their hydrophobic nature.

In areas that contain more cells, such as cancer, the water diffusion restriction will be increased compared to that of adjacent normal tissues of the same type.

The ADC sequence calculates the net displacement of water molecules on a timescale that reflects diffusion sensitization, and the hyposignal of an area confirms the diffusion restriction visualized in the hypersignal on the DWI sequence (Figure 3).

Figure 3 Method of measurement of the suspicious lesion (red, left) and of the transitional zone without changes (green, left) in the 3DnetTM platform and the confirmation of diffusion restriction at the level of the suspicious zone on the DWI sequence (right). Despite the fact that there is an important diffusion restriction at the level of this PI-RADS 3 lesion, the histopathological result was negative

The premise of this paper starts from the fact that a more precise definition related to the intensity of the lesions on the ADC sequence can lead to a new subclassification of PI-RADS 3 lesions in the transitional zone and to an improvement of the specificity of the PI-RADS classification.

Materials and Methods

The study was retrospective and included MRIs performed between March 2020 and July 2023 in the Medical Imaging Department of the University of Medicine and Pharmacy of Craiova using a Philips Ingenia 3.0T MRI scanner, selecting only patients who, based on prostate MRI examinations, contained exclusively PI-RADS 3 lesions, visible on DWI and ADC sequences, without other lesions of the same prostatic half that are PI-RADS 4 or 5 at the level of the transitional zone or PI-RADS 3,4,5 lesions or with the appearance of prostatitis at the level of the peripheral zone.

These were filtered from a database, by three radiologists, who analyzed and re-evaluated the examinations with a prostate image quality score (PI-QUAL) of 5 or 4, which corresponded to the PI-RADS V2.1 criteria;

A final filtering was done at the end, where only the examinations where the histo-pathological examination was found were kept.

The number of cases that meet all these conditions was reduced to 18, where a series of characteristics were noted for each one: PI-RADS 3 lesion area on the ADC sequence, the minimum and average ADC value of the lesion, the average ADC value of the transitional zone outside the lesion, prostate-specific antigen (PSA), prostatic volume, PSA density and biopsy result.

The characteristics of the described lesion were calculated using the software integrated into the 3DnetTM platform provided by Biotronics3D (Figure 3), and the serological results were obtained from laboratories accredited by the Romanian accreditation association.

The cognitive-targeted biopsy was performed in the Craiova county emergency clinical hospital and we do not know the local protocol by which the biopsy of a patient with a PI-RADS 3 lesion is decided, which is why we left the search for the histopathological result at the very end of the data search.

Results

The final number of examinations that met all the conditions of the study was 18.

Of these, 4 (22%) were negative, 3 (16%) proved to be prostatitis, 7 (39%) were Gleason 6 (3+3), and the remaining 4 (22%) had clinically significant cancer (Figure 4).

Figure 4 Histopathological results of the 18 biopsies performed on PI-RADS 3 lesions

The average ADC value of the negative lesions was 865 (165) m2/sec, and of the positive ones (including Gleason 6) was 869 (118) m2/sec, which denies the hypothesis that there could be a value limit ADC to delimit PI-RADS 3 from PI-RADS 2 or 4 in this sample of patients.

Furthermore, we reported the average ADC value of each such lesion to the adjacent unchanged transitional zone and obtained a greater difference of 432 (163) m2/sec between the negative lesions and their adjacent transitional zone, compared to the difference of 399 (127) m2/sec between the positive lesions and their adjacent ones (Figure 5).

The fact that positive and negative lesions not only had similar values but even had a slight negative correlation denotes the lack of metric specificity of the ADC sequence in PI-RADS 3 lesions in the transitional zone.

Figure 5 Average ADC values of the biopsied lesions sorted according to the histopathological result

The average PSA value of negative lesions was 5.23 (1.31) ng/ml, of positive lesions 8.72 (3.58) ng/ml and of cases with prostatitis 13.57 (3, 53) ng/ml, which says more about PI-RADS 3 lesions than the ADC value (Figure 6).

Two of the three total positive lesions that had a PSA greater than 10 were the only ones from the entire group of patients that had positive lesions diagnosed at the same histopathological examination in the contralateral half.

Figure 6 PSA values of the biopsied lesions sorted according to the histopathological result

PSA density, calculated by the ratio between PSA and prostate volume can be seen in Figure 7, where all 4 negative, non-prostatitis cases had PSA density below 0.1 and the 4 positive, clinically significant cases had PSA density above 0.1.

It should be noted that the two cases with the highest PSA densities are at the same time the only ones that showed positive lesions in the other half of the prostate as well.

Figure 7 The PSA density values of the prostates from which the lesions were biopsied, sorted according to the histopathological result

The age of the patients and the calculation of the minimum ADC value from each lesion did not correlate significantly with the histopathological results, but were added to the final data table (Table 1).

Table 1 All data obtained, sorted according to the histopathological result

AGE

	PSA (ng/ml)

	LESION DIAGNOSTIC

	CONTRALATERAL DIAGNOSIS

	LESION AREA

	LESION MIN ADC

	LESION AVG ADC

	NON-SUSPECT AREA AVG ADC

	PROSTATE VOLUME

	
67

	5,73

	Negative

	-

	0,39 cm2

	351

	676

	1222

	78,83 cm3

	
51

	6,23

	Negative

	-

	2,11 cm2

	897

	1181

	1312

	74,99 cm3

	
74

	3,30

	Atyp. hyperplasia

	-

	0,19 cm2

	577

	747

	1297

	56,33 cm3

	
70

	5,65

	Atyp. hyperplasia

	-

	0,73 cm2

	450

	661

	1284

	82,01 cm3

	
67

	12,44

	Prostatitis

	-

	0,44 cm2

	771

	969

	1343

	71,67 cm3

	
69

	10,74

	Prostatitis

	-

	0,46 cm2

	680

	899

	1330

	171,34 cm3

	
79

	17,52

	Prostatitis

	-

	1,16 cm2

	784

	927

	1302

	184,52 cm3

	
67

	7,38

	Gleason 6(3+3)

	-

	0,4 cm2

	707

	818

	1063

	49,94 cm3

	
63

	4,02

	Gleason 6(3+3)

	-

	0,67 cm2

	566

	711

	1050

	42,10 cm3

	
71

	9,21

	Gleason 6(3+3)

	-

	0,57 cm2

	488

	714

	1175

	38,36 cm3

	
70

	3,65

	Gleason 6(3+3)

	-

	0,75 cm2

	745

	915

	1252

	48,32 cm3

	
73

	11,60

	Gleason 6(3+3)

	-

	0,83 cm2

	570

	857

	1303

	63,23 cm3

	
76

	6,02

	Gleason 6(3+3)

	-

	0,93 cm2

	783

	942

	1233

	91,06 cm3

	
68

	16,40

	Gleason 6(3+3)

	7(3+4)

	1,69 cm2

	893

	1096

	1350

	63,10 cm3

	
64

	8,97

	Gleason 7(3+4)

	-

	0,8 cm2

	771

	905

	1385

	69,04 cm3

	
77

	10,32

	Gleason 7(3+4)

	7(4+3)

	0,9 cm2

	625

	957

	1513

	35,20 cm3

	
75

	8,57

	Gleason 7(4+3)

	-

	0,81 cm2

	691

	919

	1261

	56,59 cm3

	
75

	9,83

	Gleason 8(4+4)

	-

	0,43 cm2

	204

	732

	1374

	85,53 cm3

	

Discussions

We chose to look for a method of subclassifying PI-RADS 3 lesions according to the ADC value, and for this we looked for a very specific patient profile: with exclusively PI-RADS 3 lesions in the transitional zone and without other lesions at least the same half of the prostate, this is mainly because the biopsies to which we had access were cognitive-guided and sorted in the histopathology report as lesions in the left half, respectively in the right half.

This specific profile led to a small number of subjects but increased confidence in the results obtained.

PSA value and PSA density proved specific, reconfirming the conclusions of Schoots et al. [11] according to which unnecessary biopsies can be avoided by applying a PSA density threshold of 0.1 in PI-RADS 3 lesions.

This threshold is also motivated by the report of patient #13 from Table 1, with PSA values of 6.02 and PSA density of 0.07, initially his PSA was monitored over a 6-month interval and it increased slightly during this period, which is why it was subsequently biopsied and confirmed with Gleason 6 (3+3).

The main purpose of this paper still remains the correlation between the ADC value of PI-RADS 3 lesions in the transitional zone and the risk of cancer, including the clinically insignificant one, and in our small group of patients we are dealing with an even slightly negative correlation, so that we expect that even in larger groups of patients there will not be a threshold between biopsiable and non-biopsiable PI-RADS 3 lesions.

Abreu-Gomez et al. [12] concluded in another context that the size of visible lesions on the ADC sequence was associated with the pathological stage, while the ADC metrics were not, while Zhang et al. [13] argue that applying an ADC threshold of 900m2/sec to all PI-RADS 3 lesions proved to be useful only when correlated with PSA density.

Since the threshold we are looking for in the ADC value seems to be found in the PSA density rather than in other imaging aspects, future work will include studies on several biopsied PI-RADS 3 lesions, preferably under ultrasound guidance, in which we will look for ways to associate serological results with imaging ones in search of more effective risk stratifications between cancer and inflammatory changes.

Conclusions

PI-RADS 3 lesions did not prove to be subclassifiable according to ADC values, because in our sample they alone or compared to other measurements did not prove relevant for risk stratification, while PSA value and PSA density proved useful.

Conflict of interests

None to declare.
==== Refs
References

1 Bergengren O Pekala KR Matsoukas K Fainberg J Mungovan SF Bratt O Bray F Brawley O Luckenbaugh AN Mucci L Morgan TM Carlsson SV Update on Prostate Cancer Epidemiology and Risk Factors-A Systematic Review Eur Urol 2023 84 2 191 206 37202314
2 Culp MB Soerjomataram I Efstathiou JA Bray F Jemal A Recent Global Patterns in Prostate Cancer Incidence and Mortality Rates Eur Urol 2020 77 1 38 52 31493960
3 Natale C Koller CR Greenberg JW Pincus J Krane LS Considering Predictive Factors in the Diagnosis of Clinically Significant Prostate Cancer in Patients with PI-RADS 3 Lesions Life (Basel) 2021 11 12 1432 1432 34947963
4 Mythreyi C Lauren H Cassandra V ASSESSMENT AND REPORTING Mythreyi C PI-RADS: Prostate Imaging-Reporting and Data System 2019 Chicago ACR - ESUR-AdMeTech 10 32
5 Lee G Oto A Giurcanu M Prostate MRI: Is Endorectal Coil Necessary? A Review Life (Basel) 2022 12 4 569 569 35455060
6 Felker ER Lee-Felker SA Feller J Margolis DJ Lu DS Princenthal R May S Cohen M Huang J Yoshida J Greenwood B Kim HJ Raman SS In-bore magnetic resonance-guided transrectal biopsy for the detection of clinically significant prostate cancer Abdom Radiol (NY) 2016 41 5 954 962 27118268
7 Thai JN Narayanan HA George AK Siddiqui MM Shah P Mertan FV Merino MJ Pinto PA Choyke PL Wood BJ Turkbey B Validation of PI-RADS Version 2 in Transition Zone Lesions for the Detection of Prostate Cancer Radiology 2018 288 2 485 491 29786491
8 Bastian-Jordan M Magnetic resonance imaging of the prostate and targeted biopsy, Comparison of PIRADS and Gleason grading J Med Imaging Radiat Oncol 2018 62 2 183 187 28990727
9 Osses DF van Asten Kieft GJ Tijsterman JD Prostate cancer detection rates of magnetic resonance imaging-guided prostate biopsy related to Prostate Imaging Reporting and Data System score World J Urol 2017 35 2 207 212 27287889
10 Venderink W van Luijtelaar Bomers JGR van der Hulsbergen-van de Barentsz JO Sedelaar JPM Futterer JJ Results of Targeted Biopsy in Men with Magnetic Resonance Imaging Lesions Classified Equivocal, Likely or Highly Likely to Be Clinically Significant Prostate Cancer Eur Urol 2018 73 3 353 360 28258784
11 Schoots IG Padhani AR Risk-adapted biopsy decision based on prostate magnetic resonance imaging and prostate-specific antigen density for enhanced biopsy avoidance in first prostate cancer diagnostic evaluation BJU Int 2021 127 2 175 178 33089586
12 Abreu-Gomez J Walker D Alotaibi T McInnes MDF Flood TA Schieda N Effect of observation size and apparent diffusion coefficient (ADC) value in PI-RADS v2.1 assessment category 4 and 5 observations compared to adverse pathological outcomes Eur Radiol 2020 30 8 4251 4261 32211965
13 Zhang Y Zeng N Zhang FB Rui Huang Tian Y Performing Precise Biopsy in Naive Patients with Equivocal PI-RADS, Version 2, Score 3, Lesions: An MRI-based Nomogram to Avoid Unnecessary Surgical Intervention Clin Genitourin Cancer 2020 18 5 367 377 32771334
