Conference PaperPDF Available

Proliferative inflammatory atrophy (PIA): New precusor lesion for prostate cancer.

Authors:

Abstract

In recent publications there is emerging evidence that inflamma¬tion may contribute to prostatic carcinogenesis. It is hypothesised that oxydative stress triggers the appearance of proliferative inflammatory atrophy (PIA) lesions which can degenerate into carcinoma. The results from recent studies connecting PIA with cancer have been contraversial. The aim of the study was to assess the risk of detecting prostate carcinoma on repeat biopsy based on the pre-biopsy PSA levels, presence and morphological characteristics of inflammation and PIA lesions on initial biopsy. This retrospective study included 208 patients with PSA level ≤ 10 ng/mL, submitted to repeat biopsy for detection of prostate carcinoma during the 2003-2008 period. The presence and number of PIA lesions were determined on initial biopsy, while a modified system of inflammation type grading, developed by Irani et al, was used for pathomorphological analysis. Factor analysis was used to assign specific inflammatory factor score to each subject. Patients were divided into two groups according to the type (more chronic, more acute) of inflammation, and also according to the presence of PIA. The presence and characteristics of carcinoma were determined on repeat biopsy. Inflammatory infiltration was present in most prostate biopsy specimens. Only 5.80% of subjects were free from inflammation, while inflammation of a varying type was present in the remaining 94.20% of study subjects. PIA was detected in 39.40% of patients. The majority of PIA lesions were grouped in biopsy cores with chronic inflammation (88.20%). Acute inflammation caused a significant decrease in FPSA and F/TPSA values (p < 0.001 both) thus biochemicaly mimicking prostate carcinoma. The presence of inflammation and PIA did not correlate with TPSA values. Patients with carcinoma on repeat biopsy showed more chronic (p < 0.001) inflammation on initial biopsy. Kaplan-Meier analysis indicated the risk of carcinoma detection on repeat biopsy to be greater in subjects with more chronic (p < 0.001) inflammation. The rate of PIA on initial biopsy was significantly higher in the group of subjects with carcinoma on repeat biopsy (p < 0.001). The subjects with carcinoma had a significantly greater number of PIA lesions per initial biopsy set (p < 0.001). Kaplan-Meier analysis yielded a higher risk of carcinoma detection in the group of subjects with PIA on initial biopsy (p < 0.001). ROC analysis yielded the optimal borderline PIA number on initial biopsy for the diagnosis of carcinoma on repeat biopsy of 1.01 (sensitivity 0.368, specificity 0.922); thus, subjects with two or more PIA lesions were at a higher risk of carcinoma detection on repeat biopsy (p < 0.001). Like carcinoma, the majority of PIA lesions (85.97%) were located in the peripheral zone of the prostate, while only 14.03% were found in the transition zone. On initial biopsy, a significantly greater proportion of PIA lesions were grouped in biopsy cores with carcinoma subsequently detected on repeat biopsy. The number of PIA lesions and age showed greatest independent predictive power as risk factors for the diagnosis of carcinoma on repeat biopsy (OR = 2.537; p < 0.001 and OR=1.060; p = 0.039, respectively), whereas F/TPSA and acute inflammation conferred a protective effect from prostate carcinoma (OR = 0.762; p < 0.001 and OR = 0.056; p < 0.001, respectively). Although study results can not be interpreted as definitive evidence for PIA to cause prostate carcinoma, they do indicate that the presence of chronic inflammation and PIA in initial biopsy is associated with a higher risk of subsequent carcinoma detection. Thus, PIA should be characterized as a facultative precancerous lesion. Acute inflammation has a protective role through its effect on FPSA and F/TPSA levels, decreasing the risk of carcinoma detection on repeat biopsy. Following negative biopsy findings in patients with borderline PSA levels, timely recognition of histologic inflammation and related PIA may point to the need of closer follow up and earlier repeat biopsy, thus resulting in earlier detection of carcinoma in high-risk patients.
PROLIFERATIVE INFLAMMATORY ATROPHY
(PIA): NEW PRECUSOR LESION FOR
PROSTATE CANCER
Clinical Department of Urology,
“Sestre milosrdnice”
University Hospital, Zagreb, Croatia
2nd croatian congress on urogenital and sexually transmitted infections
with international participation, Opatija, june 2010.
Goran Štimac
BPH
PROSTATITIS
PROSTATE
CANCER
COMMON SYMPTOMS, ETIOLOGY,
PATHOGENESIS?
PROSTATE CANCER RISK FACTORS
AGING INHERITANCE
INFLAMMATION?
RACE
INFLAMMATION AND PROSTATE
CARCINOGESNESIS
EPIDEMIOLOGY GENETICS
INFLAMMATORY
TOXICOLOGY
MOLECULAR
PATHOLOGY
INFLAMMATION + UNIQUE PROSTATIC MILLEU + TIME.......
INCREASED OXYDATIVE STRESS (DECREASED DEFENCES)
+ TIME......................
PIA + GENOME INSTABILITY + OXYDATIVE STRESS
+ TIME......................
HGPIN + OXYDATIVE STRESS + TIME.....................
PROSTATE CANCER
Bostwick et al, 2004.
CELLULAR AND MOLECULAR MODEL OF EARLY
PROSTATE NEOPLASIA PROGRESSION
Specific morfphology of PIA:
- always tied to inflammation
- regenarative lesion:
hyperproliferaton, regeneration,
low apoptosis index, atrophy
- immature phenotype like
PIN and prostate cancer
- genetic mutations like prostate
cancer and PIN: GSTP1, p53
- zonal predisposition, adjecent to
PIN and prostate cancer
PIA – POSSIBLE NEW PRECURSOR
LESION FOR PROSTATE CANCER ?!
Zonal predisposition to prostate disease!
PROSTATE CANCER
DETECTION: 20 %
PROSTATE BIOPSY
TPSA<10 ng/ml, DRP (-)
Catalona, N Engl J Med, 1991.
PSA SENSITIVITY AND SPECIFICITY!?
ELIMINATES
50 %
UNNECESSARY
BIOPSIES
INDICATIONS: - PSA elevations, abnormal DRE
- premalignant lesions (ASAP, PIN)
- INFLAMMATION, PIA ?!
Djavan, Urol Clin N Am, 2003.
F/TPSA<20% PSAD >26
ng/ml/ccm
NEGATIVE BIOPSY - REPEAT BIOPSIES?
-Histological prostatitis has a high prevalence in cancer-free biopsies:
-Blumenfeld, Am J Surg Pathol, 1992. – 95% specimens after TURP
-Nadler, J Urol, 1995. – 97% prostate biopsy specimens
-Acute clinical prostatitis can contribute to the lack of PSA specificity!
-What about chronic or acute hystological (subclinical) inflammation?
-What about repeat prostate biopsies?
-Should we treat patients with asymptomatic prostatitis?
-Inflammation and PIA in biopsy – increased risk for prostate cancer?
INFLAMMATION AND DETECTION OF
PROSTATE CANCER - PITFALLS?
STUDY DESIGN AND AIMS
DESIGN: retrospective study included 208 patients with PSA level 10 ng/mL,
submitted to repeat biopsy for detection of prostate cancer during the 2003-2008
period.
INITIAL BIOPSY (8 CORES)
- PSA
- pathomorfological
inflammation grading
(localisation) (Irani et al.)
- presence and number
of PIA lesions (localisation)
TIME (months)
REPEAT BIOPSY (8 CORES)
- presence and characteristics
(localisation) of carcinoma
AIM: to assess the risk of detecting prostate cancer on repeat biopsy based on the:
- pre-biopsy PSA levels,
- presence and morphological characteristics of inflammation and
- presence and number of PIA lesions on initial biopsy.
INITIAL BIOPSY CORES:
PIA AND INFLAMMATION GRADING
Grade 0: no inflammation,
normalprostate epithelium
Grade I: MN infiltrates,
chronic inflammation
Grade II: MN and PMN infiltrates,
chronic active inflammation
Grade III: PMN infiltrates,
acute inflammation
PIA:
presence and number of
lesions in each biopsy core
110
88
78
69
80
91
94
88
78
89
89
99
95
89
84
95
14
18
31
34
29
20
27
21
6
13
10
6
4
8
3
4
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1. 2. 3. 4. 5. 6. 7. 8.
Initial biopsy cores
Acute inflammation
Chronic active inflammation
Chronic inflammation
No inflammation
RESULTS: DISTRIBUTION OF INFLAMMATION GRADES
IN INITIAL BIOPSY CORES
Inflammatory infiltration was present in most prostate biopsy specimens.
Only 5.80% of subjects were free from inflammation, while inflammation
of a varying type was present in the remaining 94.20% of study subjects.
In almost each set of biopsies we found different patterns of inflammation grades
Factor analysis: specific inflammatory score for each patient (prostate)
RESULTS: REDUCTION OF INFLAMMATION GRADING
0
NO INFLAMMATION
3
ACUTE
INFLAMMATION
(PMN infiltrate)
2
CHRONIC ACTIVE
INFLAMMATION
(MN+PMN infiltrate)
1
CHRONIC
INFLAMMATION
(MN infiltrate)
REGROUPING
(REDUCTION OF CLASSSIFICATION BY FACTOR ANALYSIS)
MORE CHRONIC
INFLAMMATION
GROUP
MORE ACUTE
INFLAMMATION
GROUP
RESULTS: SUBCLINICAL INFLAMMATION RELATED TO
PREBIOPSY TPSA, FPSA AND F/TPSA VALUES
Chronic Acute
Inflammation type in initial biopsy (groups)
TPSA
( p = 0,431; Mann-Whitney)
Chronic Acute
Inflammation type in initial biopsy (groups)
FPSA
( p < 0,001; Mann-Whitney)
Chronic Acute
Inflammation type in initial biopsy (groups)
F/TPSA
( p < 0,001; Mann-Whitney)
Acute subclinical inflammation caused a significant decrease in FPSA and F/TPSA
values biochemicaly mimicking prostate carcinoma.
f/tPSA=18%
ACUTE
CHRONIC
RESULTS: SCATTERGRAM OF TPSA VS F/TPSA IN PATIENTS
WITH ACUTE AND CHRONIC INFLAMMATION
TPSA
F/TPSA
Inflammation:
18
25
6
21
22
20
12
18
0
1
3
1
4
1
2
3
1
1
0
0
0
0
1
1
0%
20%
40%
60%
80%
100%
1. 2. 3. 4. 5. 6. 7. 8.
Initial biopsy cores
PIA + acute
inflammation
PIA + chronic active
inflammation
PIA + chronic
inflammation
RESULTS: DISTRIBUTION OF PIA RELATED TO
INFLAMMATION GRADES IN INITIAL BIOPSY CORES
PIA was detected on initial biopsy in 39.40% of patients.
The majority of PIA lesions grouped in cores with chronic inflammation (88.20%).
N
(%)
PROSTATE CARCINOMA
IN REPEAT BIOPSY TOTAL
NO YES
INFLAMMATION
TYPE IN
INITIAL BIOPSY
MORE
CHRONIC
36
(17,30)
68
(32,70)
104
(50,00)
MORE
ACUTE
66
(31,70)
38
(18,30)
104
(50,00)
TOTAL 102
(49,00)
106
(51,00)
208
(100,00)
χ2 = 17,314, df = 1, p < 0,001
RESULTS: DETECTION OF CANCER IN REPEAT BIOPSY
RELATED TO INFLAMMATION TYPE ON INITIAL BIOPSY
Patients with carcinoma on repeat biopsy showed more chronic inflammation
on initial biopsy.
Prostate cancer in repeat biopsy
Inflammation score in initial biopsy
(p < 0,001; Mann-Whitney test)
No Yes
More acute
More chronic
RESULTS: DETECTION OF CANCER IN REPEAT BIOPSY
RELATED TO INFLAMMATION SCORE ON INITIAL BIOPSY
Time (months) after initial biopsy
Proportion of cancer free (cumulative)
Chronic inflammation group
Acute inflammation group
Censored
follow up
(p < 0,001; Log-Rank test)
RESULTS: KAPLAN-MEIER SURVIVAL CURVES FOR
CHRONIC AND ACUTE INFLAMMATION GROUPS
Kaplan-Meier analysis indicated the risk of cancer detection on repeat biopsy
to be greater in subjects with more chronic inflammation in initial biopsy.
END POINT: time
to positive biopsy
RESULTS: DETECTION OF CANCER IN REPEAT BIOPSY
RELATED TO PRESENCE OF PIA IN INITIAL BIOPSY
N
(%)
PROSTATE CANCER
IN REPEAT BIOPSY TOTAL
NO YES
PRESENCE
OF PIA IN
INITIAL
BNIOPSY
NO 76
(36,50)
50
(24,10)
126
(60.60)
YES 26
(12,50)
56
(26,90)
82
(39,40)
TOTAL 102
(49,00)
106
(51,00)
208
(100,00)
χ2 = 16,27, df = 1, p < 0,001
The presence of PIA on initial biopsy was significantly higher in the group of
subjects with carcinoma on repeat biopsy.
RESULTS: DETECTION OF CARCINOMA IN REPEAT BIOPSY
RELATED TO NUMBER OF PIA IN INITIAL BIOPSY
Number of PIA in initial biopsy
Prostate cancer in repeat biopsy (groups)
No Yes
(p < 0,001; Mann-Whitney test)
The subjects with carcinoma had a significantly greater number of PIA lesions
per initial biopsy set.
NO
YES
Censored
follow-up
Time (months) after initial biopsy
Proportion of cancer free (cumulative)
Presence of PIA in
Initial biopsy:
RESULTS: KAPLAN-MEIER SURVIVAL CURVES FOR
GROUPS WITH PIA AND WITHOUT PIA IN INITIAL BIOPSY
Kaplan-Meier analysis yielded a higher risk of carcinoma detection in the group
of subjects with PIA on initial biopsy.
(p < 0,001; Log-Rank test)
END POINT: time
to positive biopsy
RESULTS: ZONAL PREDISPOSITION
RIGHT LOBE LEFT LOBE
Mid
Base Base
Mid
Apex
TRANSITION
ZONE
PERIPHERAL
ZONE
12,30 %
17,50 %
17,00 %
13,50 %
12,90 % 14,90 %
5,26 % 8,77 %
Apex
PIA
in initial biopsy
PROSTATE CANCER
in repeat biopsy
RIGHT LOBE LEFT LOBE
Apex
Mid
Base Base
Mid
Apex
TRANSITION
ZONE
PERIPHERAL
ZONE
15,74 %
17,02 %
17,02 %
14,89 %
12,34 % 14,04 %
3,83 % 5,11 %
Like carcinoma, the majority of PIA (85.97%) were located in the peripheral zone of
the prostate, while only 14.03% were found in the transition zone.
A significantly greater proportion of PIA lesions were grouped in biopsy cores with
carcinoma subsequently detected on repeat biopsy (p < 0,05, Mann-Whitney test).
ROC analysis yielded the borderline PIA number on initial biopsy for the
diagnosis of cancer on repeat biopsy of 1.01 (sensitivity 0.368, specificity 0.922).
Subjects with two or more PIA lesions were at a higher risk of carcinoma detection
on repeat biopsy.
RESULTS: PREDICTION OF CANCER DETECTION
ACCORDING TO NUMBER OF PIA LESIONS
1 PIA
1-specificity
Sensitivity
(p < 0,001; Wilcoxon test)
Optimal cut-off
value =1.01
AGE
FPSA
TPSA
Number
of
PIA
F/TPSA
Inflammation
factor
score
*
*
*
*
*
* *
*
*
*
*
*
*
RESULTS: SPEARMAN¨S
CORRELATION ANALYSIS MATRIX
- statitically significant correlation
RESULTS: MULTIVARIATE ANALYSES OF INITIAL BIOPSY
PARAMETERES TO PREDICT PROSTATE CANCER
INITIAL BIOPSY
PARAMETERS
WALD
TEST df p
VALUE
ODDS
RATIO CONFIDENCE INTERVALS
AGE 4,276 1 0,039 1,060 1,003 1,119
TPSA 0,347 1 0,556 1,197 0,658 2,175
FPSA 0,179 1 0,673 0,404 0,006 27,137
F/TPSA 26,271 1 < 0,001 0,762 0,687 0,846
NUMBER OF PIA 18,728 1 < 0,001 2,537 1,664 3,869
INFLAMATION
SCORE 15,359 1 < 0,001 0,056 0,013 0,237
The number of PIA lesions and age showed greatest independent predictive
power as risk factors for the diagnosis of carcinoma on repeat biopsy.
F/TPSA and inflammation score (more acute inflammation) were found to be
important protective factors.
METHOD: BACKWARD STEWISE LOGISTIC REGRESSION
Histological prostatitis has a high prevalence in cancer-free
prostate biopsy specimens
Acute inflammation has a protective role through its effect on
FPSA and F/TPSA levels, decreasing the risk of carcinoma on
repeat biopsy:
treatment could increase PSA sensitivity and specificity
treatment could decrease the rate of unnecessary biopsies
tretament of prostatitis – long-term prostate cancer prevention?
PIA with chronic inflammation is associated with a higher risk of
subsequent prostate cancer detection.
PIA should be characterized as a facultative precancerous lesion
timely recognition could result in earlier detection of carcinoma in
high-risk patients
CONCLUSIONS!
QUESTIONS?
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.