www.auajournals.

org/journal/juro
JU Insight

Management and Oncologic Outcomes of Incidental Prostate

Cancer After Transurethral Resection of the Prostate in Denmark
Riccardo Leni , Andrew Julian Vickers , Klaus Brasso , et al.
Correspondence: Hein Vincent Stroomberg ([email protected]).
Full-length article available at https://doi.org/10.1097/JU.0000000000004159.

Study Need and Importance: Approximately 1 in 10

patients without prior prostate biopsy undergoing
surgery for lower urinary tract symptoms harbors
incidental prostate cancer; however, practice guide-
lines do not provide recommendations for its manage-
ment. We aimed at describing the oncologic outcomes
of patients with Grade Group (GG) 1 and GG2 prostate
cancer diagnosed at transurethral resection of the
prostate (TURP).
What We Found: Utilizing the Danish nationwide
population-based registries, we analyzed 24,494 Figure. Cumulative incidence of prostate cancerespecific (A) and
patients who underwent TURP from 2006 to 2022; other-cause death (B) stratified for initial transurethral
there were 1016 men with GG1 and 381 with GG2 resection of the prostate (TURP) diagnosis. Red indicates
disease. The 5-year cumulative incidence of further Grade Group (GG) 2; blue, GG1; black, nonmalignant. Shaded
MRIs or biopsies was 36% (95% CI 33%-39%) for areas are 95% CI.
GG1 and 30% (95% CI 25%-34%) for GG2. Fifteen-
year prostate cancer mortality was 8.4% (95% CI
5.3%-11%) for GG 1 and 14% (7.5%-21%) for GG2 was missing PSA values, which prevented subgroup
(Figure). A total of 270 men with GG1 underwent a analyses based on PSA levels.
biopsy after the TURP and 162 (60%) had no cancer; Interpretation for Patient Care: We observed high
in this group, prostate cancer mortality after 15 prostate cancer mortality after TURP with GG1/GG2,
years was 0.6% (95% CI 0%-1.8%). Men with post- likely due to unsampled high-grade cancer in the
TURP biopsy
GG2 had a prostate cancer mortal- peripheral zone. Patients with incidental prostate
ity of 30% (95% CI 9%-50%) 15 years post TURP. cancer should be further investigated to rule out high-
Limitations: The major limitation was the hetero- grade cancer. For patients with GG1 on TURP, once
geneous follow-up, which could lead to an over- a subsequent biopsy does not show cancer, a less
estimation of prostate cancer mortality compared to intense surveillance strategy should be discussed
a more standardized follow-up. Another limitation similarly to those with an initial nonmalignant biopsy.

THE JOURNAL OF UROLOGY® https://doi.org/10.1097/JU.0000000000004159

Ó 2024 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. Vol. 212, 692-700, November 2024
Printed in U.S.A.

692 j www.auajournals.org/jurology

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
 www.auajournals.org/journal/juro

Management and Oncologic Outcomes of Incidental

Prostate Cancer After Transurethral Resection of the
Prostate in Denmark
Riccardo Leni ,1,2,3 Andrew Julian Vickers ,1 Klaus Brasso ,4,5 Francesco Montorsi,2,3
Alberto Briganti ,2,3 Torben Kjær Nielsen ,4,5 Andreas Røder ,4,5
and Hein Vincent Stroomberg 4,6
1
Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, New York
2
Division of Experimental Oncology, Department of Urology, IRCCS San Raffaele Scientific Institute, Milan, Italy
3
Vita-Salute San Raffaele University, Milan, Italy
4
Copenhagen Prostate Cancer Center, Department of Urology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark
5
Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark
6
Section of Biostatistics, Department of Public Health, University of Copenhagen, Copenhagen, Denmark

Purpose: Approximately 1 in 10 patients without prior prostate biopsy under-

going surgery for lower urinary tract symptoms harbors incidental prostate
cancer; however, practice guidelines do not provide recommendations for its
management. We aimed at describing the oncologic outcomes of patients with
Grade Group (GG) 1 and GG2 prostate cancer diagnosed at transurethral
resection of the prostate (TURP).
Materials and Methods: This was a nationwide, population-based, observational
study of patients undergoing TURP in Denmark from 2006 to 2022 using the
Danish Prostate Registry. We estimated the cumulative incidence of further
biopsies and MRI, curative treatment, endocrine treatment, and cause-specific
mortality with competing risk analyses.

Submitted March 18, 2024; accepted July 12, 2024; published July 31, 2024.
Recusal: Dr Matulewicz, reviewing board member of The Journal of Urology, was recused from the editorial and peer review processes due
to affiliation with Memorial Sloan Kettering Cancer Center. Dr Nocera, reviewing board member of The Journal of Urology, was recused from the
editorial and peer review processes due to affiliation with IRCCS Ospedale San Raffaele. Dr Capitanio, editorial board member of The Journal of
Urology, was recused from the editorial and peer review processes due to affiliation with Vita-Salute San Raffaele University.
Funding/Support: This work was supported in part by the National Institutes of Health/National Cancer Institute with a Cancer Center Support
Grant to Memorial Sloan Kettering Cancer Center (P30 CA008748), a SPORE Grant in Prostate Cancer from Dr H. Scher (P50-CA92629), and the
Sidney Kimmel Center for Prostate and Urologic Cancers.
Conflict of Interest Disclosures: Dr Vickers reported financial interest and/or other relationship with Opko, Arctic Partners, Steba, and
Insightec. Dr Stroomberg reported receiving travel expenses and speaking and lecture fees from MSD Denmark, and project support from Pfizer
outside the scope of this manuscript. Dr Nielsen reported prior financial interest and/or other relationship with Recordati, Amgen, and Astellas. Dr
Røder reported receiving payment for genetic analyses for Bayer Denmark and Pfizer Denmark, prior speaker's fee Astellas Nordic, and advisory
roles and consultancy for Pfizer, Amgen, MSD, Sanofi, Janssen, Astra-Zeneca, Recordati, Astellas, Bayer, Orion, Ferring, Medtronic, and Intuitive.
No other disclosures were reported.
Ethics Statement: The Danish Health Authorities developed an online platform where all Danish health registries can be accessed upon
regulatory approval. This platform is termed “The Research Machine” (Forskermaskinen) and is administered by the Danish Health Data Authority,
which is part of the Danish Ministry of Health. All data for this study were generated through this platform according to contemporary Danish data
protection policies. The study received Institutional Review Board approval (IRB No. P-2022-342).
Author Contributions:
Conception and design: Leni, Vickers, Nielsen, Røder, Stroomberg.
Data acquisition: Røder, Stroomberg.
Data analysis and interpretation: Leni, Vickers, Brasso, Montorsi, Briganti, Nielsen, Røder, Stroomberg.
Drafting the manuscript: Leni.
Critical revision of the manuscript for scientific and factual content: Vickers, Brasso, Montorsi, Briganti, Nielsen, Røder, Stroomberg.
Statistical analysis: Leni, Vickers, Stroomberg.
Supervision: Vickers, Brasso, Montorsi, Briganti, Røder, Stroomberg.
Corresponding Author: Hein Vincent Stroomberg, PhD, Copenhagen Prostate Cancer Center, Department of Urology, Copenhagen University
Hospital Rigshospitalet, Ole maaloes vej, opgang 75, 2. sal, afsnit 7521, Kobenhavn, Hovedstaden 2200, Denmark ([email protected]).

THE JOURNAL OF UROLOGY® https://doi.org/10.1097/JU.0000000000004159

Ó 2024 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. Vol. 212, 692-700, November 2024
Printed in U.S.A.

www.auajournals.org/jurology j 693
Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
694 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK

Results: Among 24,494 patients who underwent TURP, there were 1016 men with GG1 and 381 with GG2
prostate cancer. The 5-year cumulative incidence of further MRIs or biopsies was 36% (95% CI 33%-39%) for
GG1 and 30% (95% CI 25%-34%) for GG2 disease. Fifteen-year prostate cancer mortality was 8.4% (95% CI
5.3%-11%) for GG1 and 14% (7.5%-21%) for GG2. A total of 270 men with GG1 disease underwent a biopsy
after the TURP, and 162 (60%) had no cancer; in this group, prostate cancer mortality after 15 years was 0.6%
(95% CI 0%-1.8%). Men with post-TURP biopsy  GG2 had a prostate cancer mortality of 30% (95% CI 9%-
50%) 15 years post TURP. The major limitation was the heterogeneous follow-up, which could lead to an
overestimation of prostate cancer mortality compared to a more standardized follow-up.
Conclusions: We observed high prostate cancer mortality after TURP with GG1 or GG2, likely due to
unsampled high-grade cancer in the peripheral zone. Patients with incidental prostate cancer should be
further investigated to rule out high-grade cancer. For patients with GG1 on TURP, once a subsequent biopsy
does not show cancer, follow-up should be lessened similar to that of patients with an initial nonmalignant
biopsy.

Key Words: transurethral resection of the prostate, incidental prostate cancer,

prostate cancer–specific mortality

TRANSURETHRAL resection of the prostate (TURP) is a METHODS

standard approach for management of lower uri-
Study Population
nary tract symptoms unresponsive to medical Data were extracted from the Danish Prostate Registry
treatment.1,2 Prostate cancer is found in the surgi- (DanProst) using a previously described methodology.12,13
cal specimen in approximately 10% to 20% of cases, Our analysis focused on men who underwent TURP after
most frequently Grade Group (GG) 1.2,3 This is 2005, with either GG1, GG2, or nonmalignant findings,
typically referred to as incidental prostate cancer.3 without prior histopathological assessment of the pros-
International guidelines do not provide recom- tate. We based our analyses only on men with low- or
mendations on prostate cancer diagnostic workup intermediate-risk prostate cancer, as higher-risk prostate
after prostate cancer diagnosed incidentally.2,4 One cancers would normally not be considered candidates for
view is that incidental cancers (stage T1a and T1b) observation only.4 In contrast with the prior analysis for
patients with a nonmalignant TURP, we did not consider
do not differ from T1c cases and usual predictors
patients diagnosed before 2006 both because of the change
such as PSA and grade should guide decision- in pathology reporting of the 2005 International Society of
making the same way.5 Others have argued that Urological Pathology classification and because in the
the natural history of incidental prostate cancer early era of PSA testing in Denmark TURP was used as a
differs from that of T1c tumors and therefore should palliative procedure for patients with advanced prostate
be included in the guidelines as a unique entity.6 cancer.11
These differing views may be the consequence of a
paucity of data on the long-term outcomes of inci- Outcome Measures
dental prostate cancer. While many studies have The data from DanProst are combined with other national
reported on prevalence, few have examined subse- registries through a person-specific registration number.
Follow-up and treatment data were extracted from the
quent morbidity and mortality.3,7-10
Danish National Patient Registry and causes of death
We previously reported that the risk of dying of
were extracted from the Danish Causes of Death Registry,
prostate cancer after a TURP in which no cancer respectively, using the December 31, 2022 update.14,15 For
was found is 1.4% at 15 years. We speculated that if the initial treatment strategy, we considered active sur-
prostate cancer is not found with limited sampling, veillance (AS) or watchful waiting based on either the
then the likelihood of harboring aggressive disease relevant code in the National Patient Registry or, in
in the peripheral zone is low.11 However, it is not case no entry was found in the registry, based on any MRI
known whether, if prostate cancer is found on or biopsy within 2 years from diagnosis without any
limited sampling, this is a proxy of unsampled form of definitive treatment within 6 months of the first
higher-grade cancer. assessment. Curative treatment was defined as either
In Denmark, since 1995, reporting of histopath- radical prostatectomy (RP) or external beam radiation
therapy (EBRT). Hormonal treatment was defined as any
ological findings of tissue retrieved from TURP has
hormone-related drug for the treatment of prostate cancer
been mandatory, and all assessments were stored in
(gonadotropin-releasing hormone analogues, surgical castra-
a nationwide registry.11 We used this dataset to tion, androgen receptor inhibitors, or CYP17A1 inhibitors) as
investigate the clinical follow-up and long-term previously described, which all represent treatment modal-
oncologic outcomes of patients who had an inci- ities for advanced disease in Denmark.4,13 Due to the absence
dental diagnosis of GG1 or GG2 prostate cancer on of guideline recommendations, outcomes after incidental
TURP. prostate cancer can be influenced by the choice to perform

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK 695

further assessments.4,16 Hence, we initially measured the

incidence of MRIs and biopsies after TURP, and we compared
postoperative PSA levels between those who underwent vs
did not undergo a subsequent biopsy.7 Finally, we estimated
the incidence of curative treatment, hormonal treatment,
prostate cancer mortality, and overall mortality with
competing risk analyses.

Statistical Analyses
Median follow-up for patients without an event was
defined from the date of diagnosis with the reverse
Kaplan-Meier method. All cumulative incidences were
calculated with the Aalen-Johansen estimator from the
date of diagnosis until the outcome of interest, with
censoring at last follow-up, at the date of a patient
emigrating to another country, or December 31, 2022,
whichever occurred first. Thirty-three men emigrated
within 6 months from the TURP, of whom 2 had GG1 in
the specimen. These patients were censored at the date of
TURP. Any treatment related to prostate cancer and all-
Figure 1. Schematic flowchart of the study population. DanProst
cause mortality was considered a competing event when
indicates Danish Prostate Registry; GG, Grade Group; TURP,
estimating the incidence of further MRIs and biopsies transurethral resection of the prostate.
after TURP; all-cause mortality was a competing event for
curative treatment and hormonal treatment; other-cause
mortality was a competing event for prostate cancer– TURP also had higher PSA values compared to pa-
specific death. Analyses are presented stratified for TURP tients with nonmalignant histology or GG1. We
histology: GG1, GG2, and nonmalignant as control. For found a slightly higher comorbidity burden for pa-
patients with GG1 we undertook several additional ana-
tients diagnosed with GG2 (any comorbidity in 16%
lyses. First, we estimated prostate cancer mortality ac-
cording to whether patients underwent subsequent
of patients with GG2 vs 13% for GG1 vs 11% for
biopsies, with subsequent biopsy as a time-dependent co- nonmalignant). There were 20% to 25% with
variate to address immortal-time bias from time of missing data on preoperative PSA, more frequently
TURP.16,17 We then analyzed prostate cancer mortality in for patients diagnosed before 2015 (Supplementary
a subgroup of men who underwent a biopsy, according to Table 1, https://www.jurology.com).
histopathological findings on reevaluation: nonmalig-
nant, GG1, or  GG2, respectively, from the date of the
biopsy. The incidence of further MRIs or biopsies and Table 1. Baseline Patient Characteristics
curative treatment (RP or EBRT) was estimated ac-
Characteristic Nonmalignant GG1 GG2
cording to age and comorbidities, reported according to
(n [ 22,140) (n [ 1016) (n [ 381)
the 2010 update of the Charlson Comorbidity Index,
with differences assessed by Gray’s test.18 Moreover, 2 Age, median (IQR), y 70 (64-76) 71 (65-77) 74 (68-79)
multivariable Cox proportional hazard models for risk of Charlson Comorbidity Index, No. (%)
0 19,374 (89) 868 (87) 309 (83)
MRI or biopsy and prostate cancer–specific mortality 1-2 2206 (10) 116 (12) 53 (14)
from time of TURP were fitted to account for the effect >2 307 (1.4) 14 (1.4) 9 (2.4)
of baseline characteristics and initial management on Missing 253 (1.1) 18 (1.8) 10 (2.6)
longer-term outcomes. All analyses were conducted in R PSA
Median (IQR), ng/mL 2.6 (1.3-4.5) 2.5 (1.4-4.7) 3.8 (1.9-8.4)
4.3.2 with the packages prodlim (v.2023.08.28) and sur-
Missing, No. 5596 220 80
vival (v.3.5.7). Year of diagnosis, No. (%)
2006-2014 12,273 (55) 538 (53) 189 (50)
2015-2021 9867 (45) 478 (47) 192 (50)
RESULTS First assessment within 2 y, No. (%)
Biopsy 147 (0.6) 173 (17) 52 (14)
Baseline Characteristics MRI 342 (1.5) 151 (15) 47 (12)
Our primary study population consisted of 24,494 None 21,651 (98) 692 (68) 282 (74)
Primary treatment within 2 y, No. (%)
patients who underwent TURP from 2006 to 2022, RP 19 (0.1) 35 (3.4) 19 (5.0)
with either GG1 (n [ 1016), GG2 (n [ 381), or no EBRT 68 (0.3) 8 (0.8) 10 (2.6)
cancer in the specimen (n [ 22,140) as control. Hormones 17 (0.1) 13 (1.3) 22 (5.8)
AS/WWa 1419 (6.4) 429 (42) 147 (39)
Figure 1 displays the participant flowchart. Base-
line characteristics and first follow-up assessments Abbreviations: AS, active surveillance; EBRT, external beam radiation therapy; GG,
are reported in Table 1. The median age was 71 Grade Group; IQR, interquartile range; RP, radical prostatectomy; WW, watchful
waiting.
years (interquartile range [IQR] 65-77) for GG1 and a
AS/WW is defined as either MRI or biopsy within 2 years and no treatment within 6
74 years (IQR 68-79) for GG2. Patients with GG2 on months from that assessment.

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
696 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK

Initial Management After TURP 10-year incidence of hormonal treatment for these
Most men, irrespective of having incidental prostate groups was 5.1% (95% CI 3.6%-6.6%), 15% (95% CI
cancer, had no assessment (either MRI or biopsy) 11%-19%), and 1.6% (95% CI 1.4%-1.8%; Supple-
after TURP. The 5-year overall incidence of further mentary Table 3, https://www.jurology.com). At 15-
assessments was 27% (95% CI 24%-29%) for GG1 and year follow-up, prostate cancer mortality was 8.4%
22% (95% CI 18%-26%) for GG2, compared to 5.2% (95% CI 5.3%-11%) for GG1, 14% (95% CI 7.5%-21%)
(95% CI 4.8%-5.4%) for those with a nonmalignant for GG2, and 0.9 (95% CI 0.7%-1.2%) for nonmalig-
TURP (Supplementary Table 2, https://www.jurology. nant cases, and did not differ between those un-
com). Post-TURP biopsy was performed in 1114 pa- dergoing vs not undergoing further biopsies
tients (270 with GG1 on TURP, 85 with GG2, and 759 (Table 3, Figure 2, and Supplementary Figure 2,
without cancer in the specimen). A total of 162 (60%) https://www.jurology.com). In men with GG1 inci-
patients with GG1 on TURP who underwent a sub- dental prostate cancer, per-year increase in age was
sequent biopsy had no evidence of cancer. The fre- associated with increased prostate cancer mortality
quency of  GG2 on biopsy was 61 (22%) and 31 (37%) (HR 1.11, 95% CI 1.04-1.17, P < .001; Supplemen-
for men with GG1 or GG2 incidental prostate cancer, tary Table 5, https://www.jurology.com). For pa-
respectively (Table 2). The median PSA post TURP tients with GG1 on TURP, the 5-year incidence of
was 1.2 ng/mL (IQR 0.6-3 ng/mL) in patients without curative treatment depended on age: 17% for pa-
a biopsy and 1.3 ng/mL (IQR 0.6-2.9) for patients who tients younger than 70 years of age, 11% for pa-
underwent a biopsy (P [ .4; Supplementary Figure 1, tients aged 70 to 75 years, and 1.5% for those older
https://www.jurology.com), indicating that the choice than 75 years (Supplementary Table 4, https://www.
to order a biopsy depended on parameters other than jurology.com). In 270 men with GG1 on TURP un-
immediate postoperative PSA. For patients with GG1 dergoing subsequent biopsies we found an associa-
on TURP, the 5-year incidence of further MRIs or tion between histopathologic assessment and
biopsies depended on age: 55% for patients younger 15-year risk of prostate cancer mortality, with a 30%
than 70 years of age, 31% for patients aged 70 to 75 (95% CI 9%-50%) risk in men with GG2 on follow-up
years, and 7.2% for those older than 75 years (Sup- biopsies, compared to 0.6% (95% CI 0%-1.8%) and
plementary Table 4, https://www.jurology.com). Per- 2.1% (95% CI 0%-6.3%) risk in men with nonmalig-
year increase of age at TURP (hazard ratio [HR] nant or GG1 on biopsy, respectively (Table 3 and
0.92, 95% CI 0.91-0.93, P < .001) and year of TURP Supplementary Figure 2, https://www.jurology.com).
 2015 (HR 1.33, 95% CI 1.04-1.68, P [ .021) were
associated with increased risk of further assessment
by biopsy or MRI in men with GG1 on initial TURP DISCUSSION
(Supplementary Table 5, https://www.jurology.com). We found that in Danish patients diagnosed with
prostate cancer on TURP, nearly 1 in 4 did not un-
Long-Term Follow-Up
dergo any further evaluation (MRI or biopsies)
Median follow-up for those who did not die was 8.6
within 5 years and only 1 in 10 underwent any form
years (IQR 4.9-13 years); 6397 men were followed
of treatment thereafter. Elderly patients had a very
for 10 years or more, and of those there were 241
low incidence of further MRI, biopsies, and defini-
men with GG1, 47 with GG2, and 6109 with
tive treatment. Overall, prostate cancer mortality
nonmalignant TURP. The 10-year cumulative inci-
was high even in men diagnosed with GG1 on
dence of curative treatment (RP or EBRT) was 12%
(95% CI 9.6%-14%) for GG1, 19% (95% CI 15%-23%)
for GG2, and 2.3% (95% CI 2.1%-2.5%) for patients Table 3. Prostate Cancer Mortality in the Study Population
with initial nonmalignant findings. The respective Risk, % (95% CI)

5y 10 y 15 y
Table 2. Results on the First Follow-Up Biopsy, Irrespective of a
According to initial TURP histology
Its Timing, According to the Initial Transurethral Resection of
Nonmalignant 0.14 (0.09-0.19) 0.56 (0.43-0.69) 0.95 (0.74-1.2)
the Prostate Histology GG1 2.9 (1.8-4.0) 4.9 (3.3-6.5) 8.4 (5.3-11)
Biopsy conclusion, No. of men (% with biopsy) GG2 5.2 (2.9-7.6) 11 (7.3-16) 14 (7.5-21)
According to further biopsyb
Nonmalignant GG1 TURP GG2 TURP Without further biopsy 3.2 (1.8-4.6) 5.3 (3.4-7.2) 8.4 (5.0-12)
Characteristic TURP (n [ 759) (n [ 270) (n [ 85) With further biopsy 2.1 (2.7-3.9) 3.8 (1.2-6.5) 8.5 (1.7-15)
Patients with GG1 on TURP, according to further biopsy reportb
No cancer, No. (%) 285 (38) 162 (60) 41 (48) Nonmalignant 0.62 (0-1.82) 0.62 (0-1.82) 0.62 (0-1.82)
Cancer GG, No. (%) GG1 2.1 (0-6.3) 2.1 (0-6.3) 2.1 (0-6.3)
1 57 (7.5) 47 (17) 13 (15) GG 2 11 (2.8-20) 15 (4.1-26) 30 (9.0-50)
2 99 (13) 28 (10) 16 (19)
3 318 (42) 33 (12) 15 (18) Abbreviations: GG, Grade Group; TURP, transurethral resection of the prostate.
a
Time calculated from the date of TURP.
b
Abbreviations: GG, Grade Group; TURP, transurethral resection of the prostate. Time calculated from the date of post-TURP biopsy.

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK 697

Figure 2. Cumulative incidence of prostate cancer–specific (A) and other-cause death (B) stratified for initial transurethral resection of the
prostate (TURP) diagnosis. Red indicates Grade Group (GG) 2; blue, GG1; black, nonmalignant. Shaded areas are 95% CI.

TURP. This finding supports that incidental pros- The 15-year incidence of death from prostate
tate cancer requires further assessments due to cancer was 8.4% in patients with GG1 on TURP but
initial inadequate sampling.5 was 0.6% after a nonmalignant post-TURP biopsy
A prior study from the same registry evaluated and 2.1% if the subsequent biopsy showed GG1,
long-term outcomes after a TURP with nonmalig- indicating that it is high-grade cancer in the pe-
nant histology, demonstrating that the 15-year ripheral zone that causes death from the disease
prostate cancer mortality was 1.4%, similar to the and must be therefore ruled out. Although the con-
findings of our updated cohort.11 Clearly, even with fidence intervals were wide, our results show that
a limited sampling of the prostate, patients who do post-TURP biopsies hold prognostic information as
not have any cancer on TURP can be reassured of men with nonmalignant or GG1 on biopsy have a
having a low risk of aggressive disease on follow- low risk of subsequent prostate cancer mortality,
up.11 Here we show that if low-grade prostate can- whereas men with GG2 have a high risk of prostate
cer is found by the limited sampling associated with cancer mortality. Moreover, the 0.6% disease-
TURP, further assessment of the peripheral zone specific mortality for patients with a nonmalignant
should be performed, since finding GG1 on TURP is post-TURP biopsy was lower than the 1.9%
a proxy for unsampled higher-grade cancer. observed for patients from the same registry who
One of the few studies on the oncologic outcomes had an initial nonmalignant systematic biopsy.21
of incidental prostate cancer is an analysis of the Critically, baseline PSA in the initial nonmalignant
SEER (Surveillance, Epidemiology, and End Re- biopsy population was higher than baseline pre-
sults) database.9 Scheipner et al found that the 6- TURP PSA, and hence the 2 populations are not
year prostate cancer mortality for patients with directly comparable. However, the observed lower
incidental GG1 varied according to preoperative disease-specific mortality supports that patients can
PSA and type of subsequent treatment, with a 2% be reassured that if cancer is excluded on a biopsy
mortality for those not treated, comparable to the after incidental prostate cancer, they need similar
2.9% mortality observed at 5 years in the current monitoring to that of men with an initial nonma-
study.9 Results from the SEER database, even if lignant biopsy according to their individual risk of
limited to a shorter follow-up, are in line with our prostate cancer mortality. If, on the other hand,
findings, indicating that prostate cancer mortality GG1 is found on a post-TURP biopsy, then AS
for incidental GG1 is higher compared to patients should be the management of choice.
with GG1 monitored with contemporary AS pro- A high proportion of patients with GG2 on TURP
tocols, who typically have disease-specific mortality who underwent a subsequent biopsy had high-grade
around 1% or less at very long-term follow-up.19,20 cancer. Considering that the upper 95% CI bound for

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
698 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK

15-year prostate cancer mortality for patients who for future studies.10 Moreover, we did not observe
had GG1 on TURP and  GG2 at biopsy is 50%, this significant differences in PSA levels after TURP for
demonstrates the need for further assessments after those who did vs did not undergo further biopsies,
incidental findings of prostate cancer. Naturally, indicating that factors other than PSA, such as age,
many patients undergoing biopsy after TURP do not influenced the decision to biopsy. Missingness of
have high-grade cancer, suggesting that other pa- other important variables such as post-TURP biopsy
rameters such as prostate MRI findings might be and MRI is expected to be limited due to the oblig-
used to stratify patients and to identify who can atory nature of registering these procedures
safely avoid a biopsy. We also speculate that patients within Denmark. We postulate that MRI findings
who were previously investigated (ie, those with also influenced the decision to biopsy. However, in
prior nonmalignant biopsies) have better oncologic this population registry, we could only determine
outcomes. Nonetheless, until it is established that a whether patients underwent an MRI, but we had no
biopsy can be safely avoided if an MRI is nonsuspi- access to the reports, nor could we determine if the
cious, we do not recommend evaluating patients with biopsy was MRI targeted, and we acknowledge this
incidental prostate cancer only with MRIs. as a further limitation. That said, guidelines do not
It must be noted that in 2022 Denmark had a provide recommendations on the use of MRI in the
prostate cancer mortality of 47.6 per 100,000, which is follow-up of patients with incidental prostate can-
higher with respect to Europe (31.9 per 100,000) and cer, rendering this issue unavoidable in population-
the United States (20.4 per 100,000).22 Life expec- based observational research.4-6 Another possible
tancy explains part of this finding: for patients diag- limitation is that cause-specific mortality is subject
nosed with localized prostate cancer from 2007 to to misclassification in the Danish Death Registry.25
2016, age at diagnosis decreased, and because of It was previously demonstrated that the Registry
improved life expectancy in the Danish male popula- overestimated the proportion of death attributable
tion, age at death increased.23 The increased life ex- to prostate cancer in up to 20% of cases in men
pectancy allows men with localized prostate cancer diagnosed with prostate cancer.25 That said, a 20%
enough time to progress to advanced stage and die of relative difference in the risk of death would not
the disease. Notably, the median age in our cohort change our conclusions that the risk of mortality is
was 71 years for GG1 and 74 for GG2; this, together high even for GG1 cancer but low for patients not
with year of diagnosis, explains the low incidence of found to have higher-grade disease on biopsy.
further assessments and treatments, which for pa- Finally, we acknowledge that the current clinical
tients with GG1 was almost tenfold lower in men practice for the management of incidental prostate
older than 75 years compared to those younger than cancer in Denmark was not as standardized as for
70 years. Moreover, patients with GG2 on TURP un- GG1 on biopsy, and thus long-term outcomes are
derwent fewer subsequent assessments, plausibly due influenced by the initial management after the
to age difference. The high disease-specific mortality procedure, including frequency of PSA evaluations,
is explained both by higher-grade cancer on follow-up repeated MRIs, and biopsies. Such limitation high-
biopsies and by different surveillance intensity in lights the need for standardized follow-up in these
elderly patients. Several studies are underway to patients.
determine the optimal treatment intensity for elderly
patients, who are typically underrepresented in trial
populations and are mostly managed with conserva- CONCLUSIONS
tive strategies. For instance, the ongoing SPCG-19 We found high disease-specific mortality in patients
randomized trial will compare immediate treatment with GG1 or GG2 on TURP, likely due to unsampled
vs observation or hormonal treatment alone in pa- high-grade cancer in the peripheral zone. Therefore,
tients with high-risk nonmetastatic prostate cancer of patients with incidental prostate cancer who are
75 years of age or more.24 otherwise candidates for further treatment should be
This registry-based study has some limitations offered reevaluation with MRI or biopsies. Patients
previously noted for Danish data.12 Many men with GG1 on TURP can be reassured that they are at
diagnosed prior to 2014 had missing PSA values, low risk if they undergo a biopsy with no high-grade
meaning that subgroup analyses by PSA level were cancer being found. When no cancer is found upon
not possible. However, median PSA prior to the follow-up biopsy, a less intense surveillance strategy
TURP was almost threefold lower compared to PSA should be discussed similarly to those with an initial
prior to biopsy in the DanProst registry, and PSA nonmalignant biopsy. Further research is warranted
typically has an 80% reduction after the procedure; to determine whether prostate MRI has a role in the
therefore, adopting the usual stratifications (4 ng/mL diagnostic workup after a TURP with incidental
and 10 ng/mL) would be of unclear significance in the prostate cancer, ideally in cohorts where further as-
post-TURP setting, but this can be of specific interest sessments are performed systematically.

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK 699

REFERENCES
1. Coyne KS, Sexton CC, Thompson CL, et al. The 9. Scheipner L, Incesu R-B, Morra S, et al. Charac- 18. Quan H, Li B, Couris CM, et al. Updating and
prevalence of lower urinary tract symptoms teristics of incidental prostate cancer in the validating the Charlson Comorbidity Index and
(LUTS) in the USA, the UK and Sweden: results United States. Prostate Cancer Prostatic Dis. score for risk adjustment in hospital discharge
from the Epidemiology of LUTS (EpiLUTS) study. Published online October 23, 2023. doi:10.1038/ abstracts using data from 6 countries. Am J
BJU Int. 2009;104(3):352-360. doi:10.1111/j.1464- s41391-023-00742-7 Epidemiol. 2011;173:676-682. doi:10.1093/aje/
410X.2009.08427.x kwq433
10. Cheng BK-C, Castellani D, Chan IS-H, et al.
2. Gravas S, Gacci M, Gratzke C, et al. Summary Incidence, predictive factors and oncological 19. Carlsson S, Benfante N, Alvim R, et al. Long-term
paper on the 2023 European Association of outcomes of incidental prostate cancer after outcomes of active surveillance for prostate
Urology guidelines on the management of non- endoscopic enucleation of the prostate: a cancer: the Memorial Sloan Kettering Cancer
neurogenic male lower urinary tract symptoms. systematic review and meta-analysis. World J Center experience. J Urol. 2020;203(6):1122-1127.
Eur Urol. 2023;84(2):207-222. doi:10.1016/j. Urol. 2022;40(1):87-101. doi:10.1007/s00345- doi:10.1097/JU.0000000000000713
eururo.2023.04.008 021-03756-9
20. Maggi M, Cowan JE, Fasulo V, et al. The long-
3. Capogrosso P, Capitanio U, Vertosick EA, et al. 11. Hilscher M, Røder A, Helgstrand JT, et al. Risk of term risks of metastases in men on active
Temporal trend in incidental prostate cancer prostate cancer and death after benign transure- surveillance for early stage prostate cancer.
detection at surgery for benign prostatic hyper- thral resection of the prostateda 20-year J Urol. 2020;204(6):1222-1228. doi:10.1097/JU.
plasia. Urology. 2018;122:152-157. doi:10.1016/j. population-based analysis. Cancer. 2022;128(20): 0000000000001313
urology.2018.07.028 3674-3680. doi:10.1002/cncr.34407
21. Kawa SM, Stroomberg HV, Larsen SB, et al. A
4. Mottet N, van den Bergh RCN, Briers E, et al. 12. Stroomberg HV, Larsen SB, Lanth
en GS, et al. nationwide analysis of risk of prostate cancer
EAU-EANM-ESTRO-ESUR-SIOG guidelines on Danish Prostate Registry (DanProst)dan updated diagnosis and mortality following an initial
prostate cancerd2020 update. Part 1: screening, version of the Danish Prostate Cancer Registry, negative transrectal ultrasound biopsy with
diagnosis, and local treatment with curative methodology, and early results. J Med Syst. long-term followup. J Urol. 2022;208(1):100-108.
intent. Eur Urol. 2021;79(2):243-262. doi:10.1016/ 2023;47(1):98. doi:10.1007/s10916-023-01991-8 doi:10.1097/JU.0000000000002491
j.eururo.2020.09.042
13. Stroomberg HV, Larsen SB, Kjñr Nielsen T, 22. Bergengren O, Pekala KR, Matsoukas K, et al.
5. Mottet N, Rouviere O, van der Kwast TH. Inci- Helgstrand JT, Brasso K, Røder A. Outcomes of 2022 Update on prostate cancer epidemiology
dental prostate cancer: a real need for expansion biopsy grade group 1 prostate cancer diagnosis in and risk factorsda systematic review. Eur Urol.
in guidelines?. Eur Urol Oncol. 2022;5(2):259-260. the Danish population. Eur Urol Oncol. 2023;84(2):191-206. doi:10.1016/j.eururo.2023.04.
doi:10.1016/j.euo.2021.04.006 2024;7(4):770-777. doi:10.1016/j.euo.2023.10.005 021

6. Capitanio U, Autorino R, Bandini M, et al. Inci- 14. Lynge E, Sandegaard JL, Rebolj M. The Danish 23. Andersen MCM, Stroomberg HV, Brasso K,
dental prostate cancer (cT1a–cT1b) is a relevant National Patient Register. Scand J Public Helgstrand JT, Røder A. Diagnostic age, age
clinical and research entity and should be fully Health. 2011;39(7 suppl l):30-33. doi:10.1177/ at death and stage migration in men dying
discussed in the International Prostate Cancer 1403494811401482 with or from prostate cancer in Denmark.
guidelines. Eur Urol Oncol. 2022;5(2):256-258. Diagnostics (Basel). 2022;12(5):1271. doi:10.
doi:10.1016/j.euo.2021.03.005 15. Helweg-Larsen K. The Danish Register of Causes 3390/diagnostics12051271
of Death. Scand J Public Health. 2011;39(7 suppl
7. Yilmaz M, Toprak T, Suarez-Ibarrola R, Sigle A, l):26-29. doi:10.1177/1403494811399958 24. Loffeler S. Immediate Curative vs Conservative
Gratzke C, Miernik A. Incidental prostate can- Treatment in Older Men With M0, High-Risk
cer after holmium laser enucleation of the 16. Assel M, Sjoberg D, Elders A, et al. Guidelines for Prostate Cancer (GrandP/SPCG19). 2023.
prostateda narrative review. Andrologia. reporting of statistics for clinical research in Accessed February 15, 2024. https://clinicaltrials.
2022;54(3):e14332. doi:10.1111/and.14332 urology. Eur Urol. 2019;75(3):358-367. doi:10. gov/study/NCT05448547
1016/j.eururo.2018.12.014
8. Elkoushy MA, Elshal AM, Elhilali MM. Incidental 25. Nguyen-Nielsen M, Møller H, Tjønneland A, Borre
prostate cancer diagnosis during holmium laser 17. Zabor EC, Assel M. On the need for landmark M. Causes of death in men with prostate cancer:
enucleation: assessment of predictors, survival, analysis or time-dependent covariates. results from the Danish Prostate Cancer Registry
and disease progression. Urology. 2015;86(3):552- J Urol. 2023;209(6):1060-1062. doi:10.1097/ (DAPROCAdata). Cancer Epidemiol. 2019;59:249-
557. doi:10.1016/j.urology.2015.06.002 JU.0000000000003459 257. doi:10.1016/j.canep.2019.02.017

EDITORIAL COMMENT

We commend Leni et al for their Danish population- Interestingly, a relatively small proportion of
based retrospective cohort study of biopsy-na€ıve pa- men underwent further testing after incidental
tients diagnosed with grade group (GG) 1 and GG2 diagnosis following TURP. The 2-year cumulative
prostate cancer following transurethral resection of incidence of undergoing either an MRI or biopsy
the prostate (TURP) between 2006 and 2022.1 While was only 30% for GG1 and 24% for GG2dyounger
the incidental diagnosis of prostate cancer is not un- men were more likely to receive testing. Yet 1 in 5
common (almost 10% in this study), there are limited patients with incidental GG1 disease and 1 in 3 with
studies assessing long-term outcomes of incidental incidental GG2 disease had GG  2 prostate cancer
prostate cancer and, in turn, a lack of clear guidelines. on biopsy. This suggests that patients with transi-
This study provides clinically relevant insights that tion zone prostate cancer may harbor higher-grade
can inform the management of such patients. disease in the peripheral zone that potentially did

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.
700 MANAGEMENT AND OUTCOMES OF INCIDENTAL PROSTATE CANCER AFTER TURP IN DENMARK

not cause a notable elevation in PSA levels in cancer (2.9%-5.2% vs 0%-1.9% in active surveil-
screened men. lance).2,4 Thus, these data suggest that incidental
Prior studies have demonstrated that MRI alone is prostate cancer should be risk-stratified with a
not an effective replacement for subsequent biopsies timely “confirmatory” biopsy after considering life
and that GG1 cancers may exist in the context of expectancy.
high-risk diseases with high rates of upgrading/
adverse pathology.2,3 Importantly, compared to low- Mary O. Strasser,1 Neal A. Patel,1
grade prostate cancer managed with active surveil- and Bashir Al Hussein Al Awamlh 1
lance, the 10-year disease-specific mortality rate is 1
Department of Urology, Weill Cornell Medicine
significantly higher in TURP-diagnosed prostate New York, New York

REFERENCES
1. Leni R, Vickers A, Brasso K, et al. Management Androl Urol. 2021;10(6):2809-2819. doi:10.21037/ considering active surveillance. BJU Int.
and oncologic outcomes of incidental prostate tau-20-1370 2019;123(5):846-853. doi:10.1111/bju.14554
cancer after transurethral resection of the prostate
in Denmark. J Urol. 2024;212(5):692-700. doi:10. 3. Kaye DR, Qi J, Morgan T, et al; Michigan Uro- 4. Newcomb L, Schenk J, Zheng Y, et al. Long-term
1097/JU.0000000000004159 logical Surgery Improvement Collaborative. Patho- outcomes in patients using protocol-directed
logical upgrading at radical prostatectomy for active surveillance for prostate cancer. JAMA.
2. Shill D, Roobol M, Ehdaie B, Vickers AJ, Carlsson patients with Grade Group 1 prostate cancer: im- 2024;331(24):2084-2093. doi:10.1001/jama.2024.
SV. Active surveillance for prostate cancer. Transl plications of confirmatory testing for patients 6695

Copyright © 2024 American Urological Association Education and Research, Inc. Unauthorized reproduction of this article is prohibited.