Original Investigation | Obstetrics andGynecology
Relevance ofMolecular Profiling in PatientsWith Low-Grade Endometrial Cancer
StephanieW. Vrede,MD; Jenneke Kasius, MD, PhD; Johan Bulten,MD, PhD; Steven Teerenstra, PhD; Jutta Huvila, MD, PhD; Eva Colas, PhD; Antonio Gil-Moreno,MD, PhD;
Dorry Boll, MD, PhD; Maria Caroline Vos, MD, PhD; AnneM. van Altena, MD, PhD; Jasmin Asberger, MD, PhD; Sanne Sweegers, BSc; Willem Jan vanWeelden, MD;
Louis J. M. van der Putten, MD, PhD; Frédéric Amant, MD, PhD; Nicole C. M. Visser, MD, PhD; Marc P. L. M. Snijders, MD, PhD; Heidi V. N. Küsters-Vandevelde, MD, PhD;
Roy Kruitwagen, MD, PhD; Xavier Matias-Guiu, MD, PhD; Vit Weinberger, MD, PhD; Casper Reijnen, MD, PhD; JohannaM. A. Pijnenborg, MD, PhD
Abstract
IMPORTANCE Patients with low-grade (ie, grade 1-2) endometrial cancer (EC) are characterized by
their favorable prognosis compared with patients with high-grade (ie, grade 3) EC. With the
implementation of molecular profiling, the prognostic relevance of tumor gradingmight lose
attention. As most patients present with low-grade EC and have an excellent outcome, the value of
molecular profiling for these patients is unclear.
OBJECTIVE To determine the association of molecular profiling with outcomes among patients with
low-grade EC.
DESIGN, SETTING, AND PARTICIPANTS This retrospective cohort study included amulticenter
international European cohort of patients diagnosed with EC between 1994 and 2018, with a median
follow-up of 5.9 years. Molecular subgroups were determined by next-generation sequencing using
single-molecule molecular inversion probes and by immunohistochemistry. Subsequently, tumors
were classified as polymerase epsilon (POLE)-altered, microsatellite instable (MSI), tumor protein
p53 (TP53)-altered, or no specific molecular profile (NSMP). Patients diagnosed with any histological
subtypes and FIGO (International Federation of Gynecology and Obstetrics) stages of EC were
included, but patients with early-stage EC (FIGO I-II) were only included if they had known lymph
node status. Data were analyzed February 20 to June 16, 2022.
EXPOSURES Molecular testing of the 4molecular subgroups.
MAINOUTCOMES ANDMEASURES Themain outcomewas disease-specific survival (DSS) within
themolecular subgroups.
RESULTS A total of 393 patients with EC were included, with a median (range) age of 64.0
(31.0-86.0) years andmedian (range) bodymass index (BMI; calculated as weight in kilograms
divided by height in meters squared) of 29.1 (18.0-58.3). Most patients presented with early-stage
(290 patients [73.8%]) and low-grade (209 patients [53.2%]) disease. Of all patients, 33 (8.4%) had
POLE-altered EC, 78 (19.8%) hadMSI EC, 72 (18.3%) had TP53-altered EC, and 210 (53.4%) hadNSMP
EC. Across all molecular subgroups, patients with low-grade EC had superior 5-year DSS compared
with those with high-grade EC, varying between 90% to 100% vs 41% to 90% (P < .001).
Multivariable analysis in the entire cohort including age, tumor grade, FIGO stage, lymphovascular
space invasion, and themolecular subgroups as covariates found that only high-grade (hazard ratio
[HR], 4.29; 95%CI, 2.15-8.53; P < .001), TP53-altered (HR, 1.76; 95%CI, 1.04-2.95; P = .03), and FIGO
stage III or IV (HR, 4.26; 95% CI, 2.50-7.26; P < .001) disease were independently associated with
reduced DSS.
(continued)
Key Points
Question Is tumor molecular profile
associated with outcomes among
patients with low-grade
endometrial cancer?
Findings In this retrospective
multicenter cohort study of 393
patients, outcomes for patients with
low-grade endometrial cancer were not
associated with molecular subgroup.
Meaning These findings do not support
routine molecular profiling in patients
with low-grade endometrial cancer.
+ Supplemental content
Author affiliations and article information are
listed at the end of this article.
Open Access. This is an open access article distributed under the terms of the CC-BY License.
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 1/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
Abstract (continued)
CONCLUSIONS ANDRELEVANCE This cohort study found that patients with low-grade EC had an
excellent prognosis independent of molecular subgroup. These findings do not support routine
molecular profiling in patients with low-grade EC, and they demonstrate the importance of primary
diagnostic tumor grading and selective profiling in low-grade EC to increase cost-effectiveness.
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372
Introduction
More than 85% of patients with endometrial cancer (EC) present with low-grade histology (ie, grade
1-2) and International Federation of Gynecology and Obstetrics (FIGO) early-stage (ie, I-II)
endometrioid EC and have a favorable prognosis, with a 5-year overall survival of 95%.1,2 Standard
treatment is hysterectomy with bilateral salpingo-oophorectomy, including lymph node staging for
patients with substantial risk of lymph nodemetastasis.2
The Cancer Genome Atlas defined 4 important prognostic molecular subgroups in EC based on
integrated genomic data: ultramutated tumors with polymerase epsilon (POLE; OMIM 174762)
alteration, microsatellite instability (MSI), copy-number-high with frequent tumor protein p53 (TP53;
OMIM 191170) alteration, and copy-number-low (also known as no specific molecular profile
[NSMP]). These subgroups increase insight in biological tumor behavior based onmolecular
signature beyond current morphological classification.3 Patients with TP53-altered tumors have the
worst outcome, representing 15% of all EC diagnoses and responsible for 50% to 70% of all
EC-relatedmortality.4,5
For decades, tumor grading and FIGO staging have been used to guide primary and adjuvant
treatment.6 Currently, with incorporation of themolecular classification to guide adjuvant treatment,
the prognostic relevance of tumor grading has gained less attention.7 Molecular profiling has been
shown to improve prognostication mainly in patients with high-grade EC, probably due to poor
interobserver reproducibility of morphological classification and the prognostic and intratumoral
heterogeneity of high-grade ECs.5,8 To our knowledge, no data have been reported about the
prognostic relevance of molecular profiling specifically in patients with low-grade EC. The aim of this
study is to determine the prognostic relevance of molecular profiling within low-grade EC. As most
patients present with low-grade EC and have an excellent outcome, we hypothesized that molecular
profiling might be less useful in these patients.
Methods
This cohort study was approved by the institutional review board of Radboud University Medical
Center and the institutional review boards of all participating centers. Data used in this study were
from previous published studies by our research group; therefore, informed consent was waived for
participants. This study followed the Strengthening the Reporting of Observational Studies in
Epidemiology (STROBE) reporting guideline.
Data Source
This retrospective Europeanmulticenter study used data from 4 previously published studies9-12 and
1 study that has not been published yet, all published by our research group. A baseline overview and
flowchart of the included studies are shown in eTable 1 and eFigure 1 in Supplement 1.
Patients
All patients were surgically treated between 1994 and 2018 (median, 2006). Inclusion criteria for this
study were: patients diagnosed with primary EC with all histological subtypes and FIGO stages, with
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 2/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
available EC tissue samples, fromwhich tumors were successfully classified accordingmolecular
profiling or the Proactive Molecular Risk Classifier for Endometrial Cancer13 classification. The
exclusion criterion was unknown lymph node status in FIGO early-stage disease.
Patients were classified into 1 of 4molecular subgroups according to the diagnostic algorithm
(Figure 1): POLE-altered, MSI, TP53-altered, and NSMP. Multiple-classifiers were classified as the
molecular subgroup with the best prognosis.14
DNAAnalysis
Representative areas of EC in the surgical specimen weremarked and selected for formalin-fixed
paraffin-embedded 20-μm thick sections. Slides were cut from these formalin-fixed paraffin-
embedded sections and stained with hematoxylin and eosin. Tumor areas were marked on these
slides, and the tumor cell percentage was estimated. These specimens were digested overnight at 56
°C in TET-lysis buffer (10 mmol/L Tris/hydrochloride, pH 8.5; 1 mmol/L ethylenediaminetetraacetic
acid, pH 8.0; and 0.01% polysorbate 20 [Tween-20; Thermo Fisher]) with 5% Chelex-100 (Bio-Rad)
and 0.2% proteinase K, with subsequent inactivation at 95 °C for 10 minutes. After this was
centrifugated, the supernatant was transferred into a clean tube. DNA concentration was
determined using the Qubit Broad Range Kit (Thermo Fisher Scientific).
Single-MoleculeMolecular Inversion ProbeDesign and Library Preparation
Samples were analyzed with single-molecule molecular inversion probes (smMIPs). The design
(Integrated DNA Technologies), as well as the library preparation, were previously published.15
Further detailed information on smMIP design, library preparation, and sequencing are provided in
the eMethods in Supplement 1.
Immunohistochemical Staining and Scoring
Detailed information about the immunohistochemical staining for p53 andmismatch repair
endonucleases PMS2 andMSH6 can be found in the eMethods in Supplement 1 and original
published studies.9,10 In brief, staining for p53 was considered outside reference range when more
than 80% of tumor cell nuclei showed strong expression (overexpression) or when there was
complete absence of nuclear staining (null expression). Mismatch repair deficiency was defined as
total loss of nuclear staining of PMS2 or MSH6 in the presence of a positive internal control.
Figure 1. Diagnostic Algorithm of Patients DiagnosedWithMolecular Profiling orWith Immunohistochemistry
Endometrial cancer
POLE-altered POLE wildtype
POLE-altered MSI TP53-altered NSMP
TP53-altered
or p53-abnormal
MSS
or MMR proficient
MSI
or MMR deficient
NSMP
or p53-wildtype
Histology
POLE status
MSI or MMR
status
TP53 or p53
status
Final classification
in presentation
MMR indicates mismatch repair protein; MSI,
microsatellite instability; and NSMP, no specific
molecular profile.
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 3/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
Disease Classification
Early-stage disease was defined as FIGO stage I or II, and advanced-stage was defined as FIGO III or
IV. Low-grade EC was defined as grade 1 and grade 2 EC, and high-grade EC was defined as grade 3
endometrioid EC and nonendometrioid EC, according to the latest European Society of
Gynaecological Oncology, European Society for Radiotherapy and Oncology, and European Society
of Pathology andWorld Health Organization guidelines.2,16 The included patients in our retrospective
cohort received either full lymphadenectomy or no lymphadenectomy, as sentinel lymph node
procedure was not routinely incorporated yet.
Statistical Analysis
Statistical analyses were performed on SPSS version 25.0 (IBM) using χ2, Fisher exact test, Mann-
Whitney U test, Kaplan-Meier survival analysis, and univariable and multivariable Cox regression
analysis. For survival curves, including Hall-Wellner confidence bands, we used SAS version 9.4 (SAS
Institute). Two-tailed P < .05 was considered statistically significant. The assumption of
proportionality for the included variables was tested with log-minus-log curves and time-dependent
covariate (time × covariate). Disease-specific survival (DSS) was defined as time fromdate of surgery
to date of death from EC, all censored by date of last contact. We validated our data with the open
access database of Kandoth et al3 by performing Kaplan-Meier analysis. Method and baseline
characteristics can be found in the original article.3 Data were analyzed February 20 to June 16, 2022.
Results
Patients
In total, 689 patients were available with successful DNA analysis, of whom 296 (42.9%) were
excluded based on unknown lymph node status in FIGO early-stage disease (eFigure 1 in
Supplement 1). Baseline characteristics of the included vs excluded patients are shown in eTable 2 in
Supplement 1. Of 393 included patients, median (range) agewas 64.0 (31.0-86.0) years, andmedian
(range) bodymass index (calculated asweight in kilograms divided by height inmeters squared) was
29.1 (18.0-58.3) (Table 1). Baseline characteristics of the included patients according to the 4
molecular subgroups are shown in Table 1. Molecular subgroup distribution was 33 patients (8.4%)
with POLE-altered disease, 78 patients (19.8%) with MSI disease, 72 patients (18.3%) with TP53-
altered tumors, and 210 patients (53.4%) with NSMP. Low- and high-grade EC were equally
distributed in patients with POLE-altered andMSI tumors. Most patients with TP53-altered tumors
had high-grade EC, andmost NSMP tumorswere low-grade EC (Table 1). The EC-relatedmortality was
highest in the TP53-altered subgroup (33 patients [45.8%]) compared with the other molecular
subgroups (NSMP: 33 patients [15.7%], MSI: 6 patients [7.7%]; POLE-altered: 1 patients [3.0%]).
Excluded patients had similarly favorable DSS outcomes for all molecular subgroups within low-
grade EC.
Outcome
For the independent variables in Cox regressionmodels, the proportional hazard assumption was
checked. Results of testing the proportional hazard assumption show that all the variables were
satisfied.
The 5-year DSS of the included study cohort was worst for TP53-altered tumors and best for
POLE-altered tumors (Figure 2A). Across all molecular subgroups, patients with low-grade EC had an
outstanding 5-year DSS comparedwith patients with high-grade EC, varying between 90% to 100%
vs 41% to 90% (P < .001) (Figure 2B). For all the molecular subgroups in patients with grade 1 EC,
excellent 5-year DSSs were observed (Figure 2C). Patients with grade 2 EC and TP53-altered or NSMP
had 5-year DSSs of 85% to 95% (Figure 2D). Within the external validation cohort of 373 patients,
survival outcomes were similarly distributed across all the molecular subgroups, with 5-year DSSs
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 4/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
Table 1. Baseline Characteristics of the Included Study Cohort According to the 4Molecular Subgroups
Characteristic
Patients by molecular subtype, No. (%)
P valueTotal (N = 393) POLE-alt (n = 33) MSI (n = 78) TP53-alt (n = 72) NSMP (n = 210)
Patient characteristics
Age, median (range), y 64.0 (31.0-86.0) 58.0 (31.0-78.0) 65.0 (43.0-83.0) 64.5 (35.0-82.0) 63.5 (37.0-86.0) .001
BMI, median (range) 29.1 (18.0-58.3) 31.3 (18.4-58.3) 29.5 (21.90-46.9) 31.2 (21.2-41.1) 27.0 (18.0-38.9) .004
Primary treatment
Lymph node dissection
No 12 (3.1) 0 3 (3.8) 2 (2.8) 7 (3.3)
.26
Yes 376 (95.7) 33 (100) 75 (96.2) 67 (93.1) 201 (95.7)
Pelvic 214 (56.9) 21 (63.6) 47 (62.7) 25 (37.3) 121 (60.2)
Para-aortic 13 (3.4) 0 1 (1.3) 4 (6.0) 8 (4.0)
Pelvic and para-aortic 54 (14.4) 4 (12.1) 8 (10.7) 12 (17.9) 30 (14.9)
Unknown which nodes 95 (25.3) 8 (24.2) 19 (25.3) 26 (38.8) 42 (20.9)
Unknown 5 (1.3) 0 0 3 (4.2) 2 (1.0)
Final pathologic characteristics
Histology
EEC 318 (80.9) 28 (84.8) 69 (88.5) 41 (56.9) 180 (85.7)
<.001
Non-EEC 75 (19.1) 5 (15.2) 9 (11.5) 31 (43.1) 30 (14.3)
Grade
1-2 209 (53.2) 17 (51.5) 41 (52.6) 13 (18.1) 138 (65.7)
<.001
3 184 (46.8) 16 (48.5) 37 (47.4) 59 (81.9) 72 (34.3)
Myometrial invasion
<50% 197 (50.1) 13 (39.4) 42 (53.8) 32 (44.4) 110 (52.4)
.14>50% 194 (49.4) 19 (57.6) 35 (44.9) 40 (55.6) 100 (47.6)
Unknown 2 (0.5) 1 (3.0) 1 (1.3) 0 0
LVSI
No 304 (77.4) 27 (81.8) 64 (82.1) 41 (56.9) 172 (81.9)
<.001
Yes 89 (22.6) 6 (18.2) 14 (17.9) 31 (43.1) 38 (18.1)
Lymph nodes
N0 305 (77.6) 29 (87.9) 68 (87.2) 46 (63.9) 162 (77.1)
.02
N1 43 (10.9) 1 (3.0) 5 (6.4) 13 (18.1) 24 (11.4)
Pelvic 18 (41.9) 1 (100) 2 (40.0) 6 (46.2) 9 (37.5)
Para aortic 7 (16.3) 0 0 4 (30.8) 3 (12.5)
Pelvic and para-aortic 6 (13.9) 0 2 (40.0) 0 4 (16.7)
Unknown which nodes 12 (27.9) 0 1 (20.0) 3 (23.0) 8 (33.3)
No information 40 (10.2) 3 (9.1) 5 (6.4) 13 (18.1) 24 (11.4)
FIGO stage
Early (I-II) 290 (73.8) 27 (81.8) 68 (87.2) 37 (51.4) 158 (75.2)
<.001
Advanced (III-IV) 103 (26.2) 6 (18.2) 10 (12.8) 35 (48.6) 52 (24.8)
Adjuvant treatment
None 97 (24.7) 6 (18.2) 15 (19.2) 17 (23.6) 59 (28.1)
.02
Radiotherapy 225 (57.3) 20 (60.6) 56 (71.8) 34 (47.2) 115 (54.8)
EBRT 67 (29.8) 8 (40.0) 15 (26.8) 16 (47.1) 28 (24.3)
VBT 89 (39.6) 6 (30.0) 25 (44.6) 7 (20.6) 7 (20.6)
ERBT+VBT 47 (20.9) 5 (25.0) 10 (17.9) 5 (14.7) 5 (14.7)
Unknown 22 (9.8) 1 (5.0) 6 (10.7) 6 (17.6) 6 (17.6)
Chemotherapy 33 (8.4) 2 (6.1) 2 (2.6) 13 (18.1) 16 (7.6)
Chemoradiation 34 (8.7) 5 (15.2) 4 (5.1) 6 (8.3) 19 (9.0)
Unknown 4 (1.0) 0 1 (1.3) 2 (2.8) 1 (0.5)
Mortality
Recurrence 74 (18.8) 1 (3.1) 12 (15.8) 30 (50.8) 31 (15.4) <.001
Mortality 90 (22.9) 2 (6.1) 8 (10.3) 38 (52.8) 42 (20.0) <.001
EC-related mortality 73 (18.6) 1 (3.0) 6 (7.7) 33 (45.8) 33 (15.7) <.001
Abbreviations: alt, altered; EBRT, external beam radiation therapy; EC, endometrial
cancer; EEC, endometrioid endometrial cancer; FIGO, Federation International of
Gynecology and Obstetrics; LVSI, lymphovascular space invasion; MSI, microsatellite
instability; NSMP, No specific molecular profile; VBT, vaginal brachytherapy.
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 5/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
varying between 98% to 100% in low-grade EC and 62% to 100% in high-grade EC (P = .02)
(eFigure 2 in Supplement 1).
In multivariable analysis of the entire cohort, high-grade (hazard ratio [HR], 4.29; 95% CI, 2.15-
8.53; P < .001), TP53-altered (HR, 1.76; 95% CI, 1.04-2.95; P = .03), and FIGO advanced-stage (HR,
4.26; 95% CI, 2.50-7.26; P < .001) disease were independently associated with reduced DSS. Among
patients with low-grade EC, FIGO advanced stagewas independently associatedwith a reducedDSS,
but none of the of molecular subgroups were. However, the number of events was low and the
estimatedHR’s were of similarmagnitude as in the entire cohort (Table 2). Among patients with high-
grade EC, only FIGO advanced-stage remained associated as an independent prognostic factor for a
reduced DSS (eTable 3 in Supplement 1). Including the diagnostic year in the multivariable Cox
regression analyses did not change the results of the Cox regression analyses.
Figure 2. Five-Year Disease-Specific Survival in Patients with Endometrial Cancer in the Entire Cohort and byMolecular Subtype and Grade
0
No. at risk
0
33
78
20
32
75
30
32
75
40
27
66
60
24
47
50
27
57
0.8
DS
S
Follow-up, mo
0.6
0.4
0.2
10
33
77
POLE-alt
MSI
210 187 177 166 117148197NSMP
72 46 37 32 162559TP53-alt
Entire cohortA
0
No. at risk
0
6
14
20
6
14
30
6
14
40
6
14
60
5
11
50
6
14
1.0
0.8
DS
S
Follow-up, mo
0.6
0.4
0.2
10
6
14
POLE-alt
MSI
62 59 59 58 515760NSMP
4 4 4 3 334TP53-alt
Among patients with grade 1 diseaseC
0
No. at risk
0
11
27
20
11
26
30
11
26
40
9
25
60
9
21
50
9
22
1.0
0.8
DS
S
Follow-up, mo
0.6
0.4
0.2
10
11
27
POLE-alt
MSI
76 74 71 68 455876NSMP
9 6 5 5 446TP53-alt
Among patients with grade 2 diseaseD
0
No. at risk, (grade)
0
17
16
20
17
15
30
17
15
40
15
12
60
14
10
50
15
12
0.8
DS
S
Follow-up, mo
0.6
0.4
0.2
10
17
16
POLE-alt, (1-2)
POLE-alt, (3)
41 40 40 39 323641MSI, (1-2)
37 35 35 27 152136MSI, (3)
138
72
133
54
130
47
126
40
96
21
115
33
136
61
NSMP, (1-2)
NSMP, (3)
13 10 9 8 7710TP53-alt, (1-2)
59 36 28 24 91849TP53-alt, (3)
Stratified by gradeB
Log-rank P =.55Log-rank P =.94
Log-rank P <.0001 Log-rank P <.001
Grade 3
POLE-alt
MSI
NSMP
TP53-alt
POLE-alt
Grade 1-2
MSI
NSMP
TP53-alt
POLE-alt
MSI
NSMP
TP53-alt
POLE-alt
MSI
NSMP
TP53-alt
POLE-alt
MSI
NSMP
TP53-alt
1.0 1.0
Alt indicates altered; MSI, microsatellite instability; NSMP, no specific molecular profile.
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 6/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
Discussion
This cohort study assessed whether molecular profiling is associated with outcomes in patients with
low-grade EC. Interestingly, patients with low-grade EC had very favorable 5-year DSSs independent
of themolecular subgroups comparedwith patients with high-grade EC. Furthermore, high-grade EC,
as well as TP53-altered tumors and FIGO advanced-stage disease, were independently associated
with decreased DSS. Among patients with low-grade EC, none of themolecular subgroups were
independently associated with reduced DSS.
Our study supported previous findings3,13 regarding the excellent prognosis for POLE-altered
EC, good or intermediate prognosis for MSI and NSMP EC, and poor prognosis for TP53-altered
tumors when analyzing all histological subtypes. Moreover, this study illustrated that themolecular
subgroups were mainly discriminative among high-grade EC.3,8 To our knowledge, no previous
studies have evaluated outcomes for themolecular subgroups within patients with low-grade EC.We
analyzed the open access data of Kandoth et al3 to validate our results.
Molecular profiling has been proposed to be performed routinely in all patients with EC.2,17
However, as most patients with EC are diagnosed with low-grade disease, it is questioned whether
this strategy is beneficial and cost-effective. Our data on low-grade EC demonstrate that full
molecular profiling may not be necessary (except for screening for Lynch syndrome).18 Multivariate
analyses did not show any statistically significant association of the molecular subgroups among
patients with low-grade EC. However, the number of events was low in this subgroup analysis.
Analyzing the HRs, the high HR of TP53-altered tumors could still be associated with a reduced DSS
in patients with low-grade EC. We question whether this is mainly attributable to grade 2 EC, as
shown in the DSS curve of TP53-altered tumors within grade 2 EC. Poor interobserver reproducibility
is mainly observed within grade 2 and 3 EC; in these patients, the use of immunohistochemical or
molecular markers could be recommended, eg, TP53 genomic or expression analysis in patients with
doubtful low-grade (grade 2) EC.4,8,19,20 In this way, binary grading (low vs high) with molecular
profiling or immunohistochemistry could be optimized with respect to reproducibility.2
Molecular profiling is demanding for health care facilities and comes with high costs, which can
be especially challenging in low-income countries. Therefore, primary clinical management of EC
should be guided based onmorphological tumor characteristics, consideration of
Table 2. Cox Regression Univariable andMultivariable Analysis of Disease-Specific Survival in the Entire Cohort andWithin Low-Grade EC
Variable
Entire cohort Low-grade EC
Univariable DSS Multivariable DSS, 73 events Univariable DSS Multivariable DSS, 12 events
HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value
Patient age (continuous) 1.04 (1.02-1.07) .001 1.02 (0.99-1.05) .08 NAa NAa NAa NAa
Grade
1-2 1 [Reference]
<.001
1 [Reference]
<.001
NAa
NAa
NAa
NAa
3 7.70 (4.13-14.35) 4.29 (2.15-8.53) NAa NAa
Molecular subgroup
POLE-alt 0.17 (0.02-1.27) .09 0.16 (0.02-1.16) .07 0.00 (0.00-0.00) .99 0.00 (0.00-0.00) .98
MSI 0.45 (0.19-1.11) .08 0.51 (0.21-1.22) .13 0.73 (0.15-3.40) .69 0.65 (0.13-3.02) .58
TP53-alt 4.14 (2.53-6.75) <.001 1.76 (1.04-2.95) .03 1.58 (0.19-12.63) .66 2.94 (0.33-25.83) .63
NSMP 1 [Reference] NA 1 [Reference] NA 1 [Reference] NA 1 [Reference] NA
LVSI
No 1 [Reference]]
<.001
1 [Reference]
.64
1 [Reference]
.30
1 [Reference]
.78
Yes 3.78 (2.37-6.00) 1.13 (0.67-1.88) 2.27 (0.48-10.57) 1.28 (0.24-6.88)
FIGO
Stage I-II 1 [Reference]
<.001
1 [Reference]
<.001
1 [Reference]
.01
1 [Reference]
.008
Stage III-IV 7.02 (4.35-11.33) 4.26 (2.50-7.26) 4.57 (1.43-14.56) 5.38 (1.55-18.62)
Abbreviations: alt, altered; EC, endometrial cancer; FIGO, Federation International of
Gynecology and Obstetrics; HR, hazard ratio; LVSI, lymphovascular space invasion; MSI,
microsatellite instability; NA, not applicable; NSMP, no specific molecular profile.
a Cox regression analysis within patients with low-grade EC did not include age and
grade as variables.
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 7/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
immunohistochemistry in doubtful cases, and selective molecular profiling in patients with high-
grade or advanced-stage disease to guide adjuvant treatment decisions.21
To our knowledge, this is the first study to address the prognostic relevance of molecular
profiling in low-grade EC. Our study consisted of a large study population, with known lymph node
status in FIGO early-stage disease to prevent bias by undiagnosed stage III. Furthermore, our results
are comparable with the data of the Cancer Genome Atlas research network.3
Limitations
This study has a few limitations, including those owing its retrospective design. First, differences in
the methods between the included studies exist. More than 80% of the cohort was assessed with
complete molecular profiling and less than 20%with the immunohistochemistry surrogates of
molecular profiling according to the Proactive Molecular Risk Classifier for Endometrial Cancer
criteria. However, immunohistochemistry surrogate analysis has been established as a reliable
alternative for molecular profiling.13 Second, the original diagnosis was used without centralized
pathology review; however, slides were from large referral hospitals, and diagnoses were made by
expert gynecological pathologists. This makes our study applicable to daily practice. Third, race and
ethnicity have not been reported in our study. Although we fully agree that these patients’ race and
ethnicity might impact outcomes in several diseases, within Europe they are not routinely
documented in patient files.22 To evaluate whether race and ethnicity might have impacted our
results, we performed additional analyses within the Kandoth et al open access database.3 Race was
not statistically different between patients with low- vs high-grade EC or between EC-related
mortality.3 However, in patients with Black race, TP53-altered tumors was more frequently present,
supporting previous findings of a study by Lu et al23 that these women more often were diagnosed
with nonendometrioid EC. Therefore, it seems probable that molecular subgroups override the
prognostic relevance of race. Fourth, patients were diagnosed between 1994 and 2018, a time
spanning more than 24 years, and this could have biased the survival findings because of different
treatment strategies over time. Including the diagnostic year in the multivariable Cox regression
analyses did not change the results of the Cox regression analyses. Furthermore, although therewere
significantly more patients with low-grade EC among the excluded patients, the DSS for excluded
patients showed similar favorable outcomes for all molecular subgroups within low-grade EC.
Conclusions
The findings of this cohort study suggest that routine molecular profiling would not be beneficial in
patients with low-grade EC due to their excellent prognosis independent of molecular subgroup. Our
data demonstrate the importance of primary diagnostic tumor grading and do not support routine
molecular profiling in low-grade EC as a cost-effective approach.
ARTICLE INFORMATION
Accepted for Publication:October 21, 2022.
Published:December 16, 2022. doi:10.1001/jamanetworkopen.2022.47372
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Vrede SW
et al. JAMA Network Open.
Corresponding Author: JohannaM. A. Pijnenborg, MD, PhD, Department of Obstetrics and Gynaecology,
RadboudUniversityMedical Center, Geert Grooteplein Zuid 10, PO Box 9101, 6500HB, Nijmegen, the Netherlands
(hanny.ma.pijnenborg@radboudumc.nl).
Author Affiliations:Department of Obstetrics and Gynecology, Radboud University Medical Center, Nijmegen,
the Netherlands (Vrede, van Altena, vanWeelden, van der Putten, Kruitwagen, Pijnenborg); Department of
Obstetrics and Gynecology, Canisius-Wilhelmina Hospital, Nijmegen, the Netherlands (Vrede, vanWeelden,
Snijders); Department of Gynecologic Oncology, AmsterdamMedical Centers and Center of Gynecologic Oncology
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 8/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
Amsterdam, Amsterdam, the Netherlands (Kasius); Department of Pathology, Radboud University Medical Center,
Nijmegen, the Netherlands (Bulten, Sweegers); Department of Health Evidence, Radboud University Medial
Center, Nijmegen, the Netherlands (Teerenstra); Department of Pathology, University of Turku, Turku, Finland
(Huvila); Biomedical Research Group in Gynecology, Vall Hebron Institute of Research, Universitat Autònoma de
Barcelona, Centro de Investigación Biomédica en Red Cáncer, Barcelona, Spain (Colas); Gynecological Department,
Vall HebronUniversity Hospital, Centro de Investigación Biomédica en Red Cáncer, Barcelona, Spain (Gil-Moreno);
Pathology Department, Vall Hebron University Hospital, Centro de Investigación Biomédica en Red Cáncer,
Barcelona, Spain (Gil-Moreno); Department of Obstetrics and Gynecology, Catharina Hospital Eindhoven, the
Netherlands (Boll); Departement of Obstetrics and Gynecology, Elisabeth-Tweesteden Hospital, the Netherlands
(Vos); Department of Obstetrics and Gynecology, Medical Center–University of Freiburg, Freiburg, Germany
(Asberger); Department of Oncology, KU Leuven, Leuven, Belgium (Amant); Department of Gynaecologic
Oncology, Netherlands Cancer Institute and AmsterdamMedical Centers, Amsterdam, the Netherlands (Amant);
Department of Pathology, Stichting Laboratory for Pathology and Medical Microbiology, Eindhoven, the
Netherlands (Visser); Department of Pathology, Canisius Wilhelmina Hospital, Nijmegen, the Netherlands
(Küsters-Vandevelde); Department of Obstetrics and Gynecology, School for Oncology and Reproduction,
Maastricht University Medical Center, Maastricht, the Netherlands (Kruitwagen); Department of Pathology and
Molecular Genetics and Research Laboratory, Hospital Universitari Arnau de Vilanova, University of Lleida, Institut
de Recerca Biomèdica de Lleida, Centro de Investigación Biomédica en Red Cáncer, Lleida, Spain (Matias-Guiu);
Department of Obstetrics and Gynecology, University Hospital in Brno andMasaryk University, Brno, Czechia
(Weinberger); Department of Radiation Oncology, Radboud University Medical Center, Nijmegen, the
Netherlands (Reijnen).
Author Contributions:Drs Vrede and Pijnenborg had full access to all of the data in the study and take
responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Vrede, Kasius, Boll, vanWeelden, van der Putten, Reijnen, Pijnenborg.
Acquisition, analysis, or interpretation of data: Vrede, Kasius, Bulten, Huvila, Colas, Gil-Moreno, Vos, van Altena,
Asberger, Sweegers, van Weelden, van der Putten, Amant, Visser, Snijders, Küsters-Vandevelde, Kruitwagen,
Matias-Guiu, Weinberger, Reijnen, Pijnenborg.
Drafting of the manuscript: Vrede, Sweegers, Kruitwagen, Reijnen, Pijnenborg.
Critical revision of the manuscript for important intellectual content: Vrede, Kasius, Bulten, Huvila, Colas,
Gil-Moreno, Boll, Vos, van Altena, Asberger, van Weelden, van der Putten, Amant, Visser, Snijders, Küsters-
Vandevelde, Matias-Guiu, Weinberger, Reijnen, Pijnenborg.
Statistical analysis: Vrede, Reijnen.
Administrative, technical, or material support: Vrede, Huvila, Colas, van Altena, Asberger, Sweegers, vanWeelden,
van der Putten, Snijders, Küsters-Vandevelde, Matias-Guiu, Reijnen, Pijnenborg.
Supervision: Kasius, Bulten, Gil-Moreno, Boll, van der Putten, Amant, Kruitwagen, Matias-Guiu, Weinberger,
Pijnenborg.
Conflict of Interest Disclosures:None reported.
Data Sharing Statement: See Supplement 2.
Additional Contributions: Astrid Eijkelenboom, PhD (Tumor Genetics, Radboud University Medical Center)
helped with the interpretation of complex cases of molecular profiling and was not compensated for this work.
REFERENCES
1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2021. CA Cancer J Clin. 2021;71(1):7-33. doi:10.3322/
caac.21654
2. Concin N, Matias-Guiu X, Vergote I, et al. ESGO/ESTRO/ESP guidelines for themanagement of patients with
endometrial carcinoma. Int J Gynecol Cancer. 2021;31(1):12-39. doi:10.1136/ijgc-2020-002230
3. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic
characterization of endometrial carcinoma. Nature. 2013;497(7447):67-73. doi:10.1038/nature12113
4. Jamieson A, Thompson EF, Huvila J, Gilks CB, McAlpine JN. p53abn endometrial cancer: understanding the
most aggressive endometrial cancers in the era of molecular classification. Int J Gynecol Cancer. 2021;31(6):
907-913. doi:10.1136/ijgc-2020-002256
5. Leon-Castillo A, HorewegN, Peters EEM, et al. Prognostic relevance of themolecular classification in high-grade
endometrial cancer for patients staged by lymphadenectomy and without adjuvant treatment. Gynecol Oncol.
2022;164(3):577-586. doi:10.1016/j.ygyno.2022.01.007
6. Morice P, Leary A, Creutzberg C, Abu-RustumN, Darai E. Endometrial cancer. Lancet. 2016;387(10023):
1094-1108. doi:10.1016/S0140-6736(15)00130-0
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 9/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023
7. Bosse T, Powell M, Crosbie E, et al. 595 Implementation of collaborative translational research (TransPORTEC)
findings in an international endometrial cancer clinical trials program (RAINBO). Int J Gynecol Cancer. 2021;
31(suppl 3):A108-A109. doi:10.1136/ijgc-2021-ESGO.171
8. Bosse T, Nout RA, McAlpine JN, et al. Molecular classification of grade 3 endometrioid endometrial cancers
identifies distinct prognostic subgroups. Am J Surg Pathol. 2018;42(5):561-568. doi:10.1097/PAS.
0000000000001020
9. Reijnen C, Küsters-Vandevelde HVN, Prinsen CF, et al. Mismatch repair deficiency as a predictive marker for
response to adjuvant radiotherapy in endometrial cancer. Gynecol Oncol. 2019;154(1):124-130. doi:10.1016/j.ygyno.
2019.03.097
10. Reijnen C, Küsters-Vandevelde HVN, Ligtenberg MJL, et al. Molecular profiling identifies synchronous
endometrial and ovarian cancers as metastatic endometrial cancer with favorable clinical outcome. Int J Cancer.
2020;147(2):478-489. doi:10.1002/ijc.32907
11. vanWeeldenWJ, van der Putten LJM, Inda MA, et al. Oestrogen receptor pathway activity is associated with
outcome in endometrial cancer. Br J Cancer. 2020;123(5):785-792. doi:10.1038/s41416-020-0925-4
12. van der Putten LJ, Visser NC, van de Vijver K, et al. L1CAM expression in endometrial carcinomas: an ENITEC
collaboration study. Br J Cancer. 2016;115(6):716-724. doi:10.1038/bjc.2016.235
13. Kommoss S, McConechyMK, Kommoss F, et al. Final validation of the PROMISEmolecular classifier for
endometrial carcinoma in a large population-based case series. Ann Oncol. 2018;29(5):1180-1188. doi:10.1093/
annonc/mdy058
14. León-Castillo A, Gilvazquez E, Nout R, et al. Clinicopathological and molecular characterisation of ‘multiple-
classifier’ endometrial carcinomas. J Pathol. 2020;250(3):312-322. doi:10.1002/path.5373
15. Eijkelenboom A, Kamping EJ, Kastner-van Raaij AW, et al. Reliable next-generation sequencing of formalin-
fixed, paraffin-embedded tissue using single molecule tags. J Mol Diagn. 2016;18(6):851-863. doi:10.1016/j.
jmoldx.2016.06.010
16. WHO Classification of Tumours Editorial Board.WHO Classification of Tumours. 5th ed. International Agency
for Research on Cancer; 2020.
17. McAlpine J, Leon-Castillo A, Bosse T. The rise of a novel classification system for endometrial carcinoma;
integration of molecular subclasses. J Pathol. 2018;244(5):538-549. doi:10.1002/path.5034
18. KöbelM, Nelson GS. Letter in response to:McAlpine J, Leon-Castillo A, Bosse T. The rise of a novel classification
system for endometrial carcinoma; integration of molecular subclasses. J Pathol 2018; 244: 538-549. J Pathol.
2018;245(2):249-250. doi:10.1002/path.5068
19. Vrede SW, vanWeeldenWJ, Visser NCM, et al; ENITEC Consortium. Immunohistochemical biomarkers are
prognostic relevant in addition to the ESMO-ESGO-ESTRO risk classification in endometrial cancer. Gynecol Oncol.
2021;161(3):787-794. doi:10.1016/j.ygyno.2021.03.031
20. Visser NCM, van derWurff AAM, IntHout J, et al. Improving preoperative diagnosis in endometrial cancer
using systematic morphological assessment and a small immunohistochemical panel.Hum Pathol. 2021;117:68-78.
doi:10.1016/j.humpath.2021.08.006
21. Hamilton CA, Pothuri B, Arend RC, et al. Endometrial cancer: a Society of Gynecologic Oncology evidence-
based review and recommendations. Gynecol Oncol. 2021;160(3):817-826. doi:10.1016/j.ygyno.2020.12.021
22. Sheikh A, Netuveli G, Kai J, Panesar SS. Comparison of reporting of ethnicity in US and European randomised
controlled trials. BMJ. 2004;329(7457):87-88. doi:10.1136/bmj.38061.593935.F7
23. LuKH,BroaddusRR. Endometrial cancer.NEngl JMed. 2020;383(21):2053-2064. doi:10.1056/NEJMra1514010
SUPPLEMENT 1.
eTable 1. Baseline Characteristics of the 4 Previous Published Studies and 1 Unpublished Study
eFigure 1. Study Flowchart
eMethods.Detailed Information on DNA Analysis, smMIP Design and Library Preparation, Sequencing, and
Immunochemistry Analysis
eTable 2. Baseline Characteristics of the Included vs Excluded Patients
eFigure 2.Disease-Specific Survival Curves of the Validation Cohort
eTable 3. Cox Regression Univariable andMultivariable Analysis of Disease-Specific Survival in Patients With High-
Grade Disease
eReferences
SUPPLEMENT 2.
Data Sharing Statement
JAMANetworkOpen | Obstetrics andGynecology Relevance of Molecular Profiling in Patients With Low-Grade Endometrial Cancer
JAMA Network Open. 2022;5(12):e2247372. doi:10.1001/jamanetworkopen.2022.47372 (Reprinted) December 16, 2022 10/10
Downloaded From: https://jamanetwork.com/ by a University of Turku User  on 01/17/2023