biomedicines

Article
Performance of Radiological and Biochemical Biomarkers in
Predicting Radio-Symptomatic Knee Osteoarthritis Progression
Ahmad Almhdie-Imjabbar 1, *, Hechmi Toumi 1,2 and Eric Lespessailles 1,2

1 Translational Medicine Research Platform, PRIMMO, University Hospital Center of Orleans,

45100 Orleans, France; hechmi.toumi@chu-orleans.fr (H.T.); eric.lespessailles@chu-orleans.fr (E.L.)
2 Department of Rheumatology, University Hospital Center of Orleans, 45100 Orleans, France
* Correspondence: ahmad.almhdie@chu-orleans.fr

Abstract: Imaging biomarkers permit improved approaches to identify the most at-risk patients
encountering knee osteoarthritis (KOA) progression. This study aimed to investigate the utility of
trabecular bone texture (TBT) extracted from plain radiographs, associated with a set of clinical,
biochemical, and radiographic data, as a predictor of long-term radiographic KOA progression. We
used data from the Foundation for the National Institutes of Health (FNIH) Biomarkers Consortium
dataset. The reference model made use of baseline TBT parameters adjusted for clinical covariates and
radiological scores. Several models based on a combination of baseline and 24-month TBT variations
(TBT∆TBT) were developed using logistic regression and compared to those based on baseline-only
TBT parameters. All models were adjusted for baseline clinical covariates, radiological scores, and
biochemical descriptors. The best overall performances for the prediction of radio-symptomatic,
radiographic, and symptomatic progression were achieved using TBT∆TBT parameters solely, with
area under the ROC curve values of 0.658 (95% CI: 0.612–0.705), 0.752 (95% CI: 0.700–0.804), and
0.698 (95% CI: 0.641–0.756), respectively. Adding biochemical markers did not significantly improve
the performance of the TBT∆TBT-based model. Additionally, when TBT values were taken from the
entire subchondral bone rather than just the medial, lateral, or central compartments, better results
were obtained.

Keywords: trabecular bone texture; knee osteoarthritis; conventional radiograph; X-rays; prediction
Citation: Almhdie-Imjabbar, A.;
of knee osteoarthritis progression; longitudinal variations; biochemical biomarkers
Toumi, H.; Lespessailles, E.
Performance of Radiological and
Biochemical Biomarkers in Predicting
Radio-Symptomatic Knee
Osteoarthritis Progression. 1. Introduction
Biomedicines 2024, 12, 666. https:// Bone is among the key factors in the pathological process of OA, as illustrated by
doi.org/10.3390/biomedicines12030666 the numerous features depicted by several imaging modalities, such as Magnetic Reso-
Academic Editor: Gurjit Singh
nance Imaging (MRI)-based bone marrow lesions [1], dual-energy X-ray absorptiometry
(DXA)-based tibial subchondral bone mineral density [2], X-ray-based subchondral bone
Received: 14 February 2024 thickening, osteophytes [3], and the trabecular bone texture (TBT) of subchondral tibial
Revised: 8 March 2024 bone [4].
Accepted: 14 March 2024 TBT is a promising imaging biomarker for the prediction of radiographic knee os-
Published: 16 March 2024
teoarthritis (KOA) outcomes (incidence, progression, and total knee arthroplasty) [5–10].
TBT-based prediction models have, furthermore, demonstrated to be robust and flexible, as
they have been trained and validated on different cohorts [11].
Copyright: © 2024 by the authors.
Searching for tools for predicting KOA progression is not limited to imaging biomark-
Licensee MDPI, Basel, Switzerland. ers, as KOA is also characterized by an imbalance between pro-inflammatory (procatabolic)
This article is an open access article and anti-inflammatory cytokines and growth factors. Thus, the corresponding imbalance
distributed under the terms and between tissue degradation and formation can be assessed and monitored by both syn-
conditions of the Creative Commons ovial and blood-based biomarkers. Biochemical trial enrichment biomarkers have been
Attribution (CC BY) license (https:// suggested to improve successful therapy development for KOA [8,12].
creativecommons.org/licenses/by/ In the present study, we included biochemical data obtained from the Foundation for
4.0/). the National Institutes of Health (FNIH) OA Biomarkers Consortium, Bethesda, MD, USA,

Biomedicines 2024, 12, 666. https://doi.org/10.3390/biomedicines12030666 https://www.mdpi.com/journal/biomedicines

Biomedicines 2024, 12, 666 2 of 11

namely urinary C-terminal crosslinked telopeptide type II collagen (uCTX-II), crosslinked

N-telo peptide of type I collagen (sNTXI) serum, and hyaluronic acid (sHA) serum, which
have previously been evaluated for their role in the progression of the disease [8,13,14].
The restriction to utilizing these three biochemical biomarkers stemmed from their demon-
strated superior predictive capacity within the FNIH dataset [8]. Combining imaging
biomarkers and molecular biomarkers has received limited attention [15] despite its po-
tential to improve the prediction of KOA progression, and to stratify therapeutic interven-
tions [16,17]. Thus, in addition to the baseline parameters employed by the reference model
(age, sex, Body Mass Index (BMI), Kellgren–Lawrence (KL), joint space narrowing in the
medial tibial plateau (JSNM), and TBT at baseline), we examined the use of longitudinal
TBT variations associated with the aforementioned biochemical markers (uCTX-II, sNTXI,
and sHA) for the prediction of the radiographic, symptomatic, and radio-symptomatic
progression of KOA in the FNIH dataset.

2. Materials and Methods

2.1. Patients
This study included data from 600 patients (one knee per patient) obtained from the
nested case–control study of the Osteoarthritis Initiative (OAI) cohort, previously identified
by the FNIH Biomarkers Consortium [7]. The OAI permission group of the National
Institute of Mental Health Data Archive, Bethesda, MD, USA, gave us access to the raw
data used in our study. Only one knee per patient was included to avoid inter-organ
correlation [18]. We recall that the inclusion criteria imposed by the FNIH dataset included
clinical, biological, and radiological descriptors of included patients whose knees had
KL scores of 1–3 at baseline and whose relevant information concerning medial knee OA
progression was available at 24-month and 48-month control points.

2.2. Definition of Radiographic and Symptomatic Progression

Radiographic progression was defined as a loss in medical joint space width of less
than 0.7 mm during the first 24 months of the study, but a loss of at least 0.7 mm during the
period from 24 months to 48 months, while symptomatic progression was defined by an
insignificant worsening of pain during the first 24 months but a significant worsening of
pain during the period from 24 months to 48 months of the study, expressed by an increase
in the Western Ontario McMaster Universities Osteoarthritis (WOMAC) pain score of at
least 9 points (on a 0–100 normalized score) [19]. Radio-symptomatic progressors were
defined as patients with both radiographic and symptomatic progression. Non-progressors
were defined as patients with no radiographic progression in the medial or lateral tibial
plateaus, and no symptomatic progression in both the index and contralateral knees.
One participant with missing BMI data at baseline and two participants with missing
biochemical data were excluded from our study. Of the remaining 597 knees, 192 were
both radiographic and symptomatic (radio-symptomatic) progressors (Group 1), 102 were
radiographic-only progressors (Group 2), 103 were symptomatic-only progressors (Group
3), and 200 were non-progressors (Group 4).

2.3. Trabecular Bone Texture Analysis

Several well-known approaches have been used to describe the fractal dimension of
XR image texture, including fractal signature analysis [20], the Whittle estimator [9], and
the quadratic variation method (VAR) [9,10]. These three various fractal analysis techniques
all produced reliable outcomes in terms of their ability to predict KOA progression [10].
The VAR approach, used in [6,11], was retained for the studies presented in this paper. As
previously reported [11], the cut-off scale was observed around 500 mm on the empirical
variograms. From each knee X-ray, the TBT parameters were calculated from a patchwork
of 16 ROIs covering the entire tibial subchondral bone structure (Figure 1). Two fractal
parameters were extracted, corresponding to the texture complexity computed for the
two micro (µ: below 400 mm) and milli (m: above 600 mm) scales of observation, and
 Two fractal parameters were extracted, correspon
Biomedicines 2024, 12, 666
puted for the two micro (µ: below 400 mm) and m
3 of 11
vation, and filtered in both horizontal (HF) and v
scriptors
filtered for each
in both horizontal patchwork.
(HF) and vertical (VF) directions, yielding 64 descriptors for
each patchwork.

Figure 1. Illustration of the 16 regions of interest (ROIs) automatically selected, covering the entire
Figure 1. Illustration of the 16 regions of interest (ROIs)
tibial subchondral bone structure. Medial, central, and lateral ROIs are highlighted in green, blue,
and red, respectively.
tibial subchondral bone structure. Medial, central, and l
2.4. Biochemical Parameters
and red, respectively.
We included in our prediction models the biochemical data (BIO: uCTX-II, sNTXI, and
sHA) obtained from the FNIH OA Biomarkers Consortium.

2.4. Biochemical Parameters

2.5. Prediction Models
As described earlier, the reference model involved the use of baseline clinical covariates
We included in our prediction models the bi
(CLIN: age, sex, and BMI), radiological readings (KL and JSNM), and TBT descriptors,
whereas the proposed model involved, in addition, the use of longitudinal variations in TBT
and sHA) obtained from the FNIH OA Biomarkers
(∆TBT) parameters. All models were developed using nested logistic regression [5,7,10,21]
to evaluate their prediction performance based on the TBT parameters of the complete
tibial subchondral bone structure (Models 1–5), or those of the medial tibial plateau (TBTM)
(Model 6), lateral tibial plateau (TBTL) (Model 7), or central tibial plateau (TBTC) (Model 8).
2.5. Prediction Models
Figure 1 shows the location of these regions. All prediction models were evaluated using a
10-fold cross-validation, repeated 300 times, to avoid overfitting.
• AsTBTdescribed
Model 1: ← CLIN + KL + JSNMearlier, the
(Reference reference model invo
model)
• Model 2: ∆TBT ← CLIN + KL + JSNM
• ates (CLIN:
Model 3: age,
TBT + ∆TBT ← CLINsex,+ KLand
+ JSNMBMI), radiological reading
• Model 4: TBT + ∆TBT
• whereas
Model 5: TBT the
+ ∆TBTproposed
← BIO + CLIN + KL model
+ JSNM involved, in addition
• Model 6: TBTM + ∆TBTM ← CLIN + KL + JSNM
• TBT (∆TBT)
Model 7: TBTL + ∆TBTL parameters.
← CLIN + KL + JSNM All models were develo
• Model 8: TBTC + ∆TBTC ← CLIN + KL + JSNM
[5,7,10,21] to evaluate their prediction performanc
The model with the descriptor on the left of (←) is adjusted for the descriptor(s) on
complete tibial subchondral bone structure (Models
the right of (←).

2.6. Statistical Analysis

eau (TBTM) (Model 6), lateral tibial plateau (TBT
AUC was used as the preferred metric to determine the most predictive model [22,23].
(TBTC)
Other statistical (Model 8).predictive
metrics (positive Figure value 1 shows
(PPV), negative the location
predictive of the
value (NPV),

evaluated using a 10-fold cross-validation, repeated

• Model 1: TBT ← CLIN + KL + JSNM (Referenc
• Model 2: ∆TBT ← CLIN + KL + JSNM
Biomedicines 2024, 12, 666 4 of 11

and balanced accuracy (BACC)) were also used to further evaluate the performance of the
different models studied [6,24].
The Akaike Information Criterion (AIC) method was employed to select the most ap-
propriate set of TBT parameters, ensuring the maintenance of good predictive performance
in the proposed prediction models while reducing the risk of overfitting. Furthermore, the
DeLong method was used to assess whether the difference in AUC between the proposed
models and the reference model was statistically significant. Specifically, the test statistic
derived from DeLong’s method facilitated the computation of the p-value [25]. In this study,
a given model is deemed statistically significant compared to the reference model if the
p-value is less than 0.05.
Inspired by [7], the primary analysis evaluated the ability of the 64 TBT parameters
and their variations over 24 months to predict KOA radio-symptomatic progression (knees
with both radiographic and symptomatic progression (Group 1; 193 progressors) compared
to knees without both radiographic and symptomatic progression (Groups 2, 3 and 4;
406 controls). The secondary analyses included 4 different scenarios:
• Scenario 1 evaluated the proposed models to predict any progression (knees with either
radiographic or symptomatic progression, or both (Groups 1, 2, and 3; 397 progressors)
compared to knees without any progression (Group 4; 200 controls).
• Scenario 2 evaluated the proposed models to predict all progression (knees with either
radiographic or symptomatic progression (Groups 2 and 3; 205 progressors) compared
to knees without any progression (Group 4; 200 controls).
• Scenario 3 evaluated the proposed models to predict radiographic progression (knees
with radiographic-only progression (Group 2; 102 progressors) compared to knees
without radiographic progression (Groups 3 and 4; 303 controls).
• Scenario 4 evaluated the proposed models to predict symptomatic progression (knees
with symptomatic-only progression (Group 3; 102 progressors) compared to knees
without radiographic progression (Groups 2 and 4; 303 controls).

3. Results
Table 1 represents the characteristics of the knees included in the current study for the
three FNIH sub-datasets: radio-symptomatic, radiographic-only, and symptomatic-only
progressors, as well as the non-progressor sub-dataset.

Table 1. Characteristics of the knees included in the current study.

Females Left Knee KL 1 = 1 KL = 2 KL = 3

N◦ of Knees
(%) (%) (%) (%) (%)
Complete dataset 597 59.0 46.2 12.6 51.1 36.3
Radio-symptomatic
192 56.8 48.4 12.5 43.2 44.3
progressors
Radiographic
102 45.1 47.1 13.7 46.1 40.2
progressors
Symptomatic
103 65.0 42.7 12.6 59.2 28.2
progressors
Non-progressors 200 65.0 45.5 12.0 57.0 31.0
1 Kellgren–Lawrence.

3.1. Primary Analysis: Radio-Symptomatic Progression

We first evaluated the performance of the different models for the prediction of radio-
symptomatic progression using data from Group 1 (192 progressors) and Groups 2, 3,
and 4 (405 controls). In this scenario, using solely TBT and ∆TBT (Model 4) statistically
significantly improved the performance of the prediction of radio-symptomatic progression
with an AUC = 0.658, compared to the reference model using baseline-only TBT (Model 1),
which achieved an AUC of 0.613 (p-value = 0.03; see Figure 2). In addition, as shown in
Biomedicines 2024,12,
Biomedicines2024, 12,666
x FOR PEER REVIEW 55 of
of 11
11

Table 2, adding
7: AUC = 0.577),BIO
or parameters
central-onlyto(Model
Model8:
4 very
AUCslightly
= 0.572)improved the2).
ROIs (Table balanced accuracy
More details are
(BACC = 0.59), positive predictive value (PPV
reported in Table S1 of the Supplementary File. = 0.52), and negative predictive value
(NPV = 0.73) values. However, they both achieved the same AUC.

Figure 2.
Figure 2. ROC
ROC curves
curvesobtained
obtainedbybythe different
the models
different modelsforfor
thethe
prediction of radio-symptomatic
prediction of radio-symptomaticpro-
gression using
progression TBT
using parameters
TBT of the
parameters complete
of the tibial
complete subchondral
tibial bonebone
subchondral structure, adjusted
structure, for clin-
adjusted for
ical, biochemical,
clinical, andand
biochemical, radiological parameters.
radiological parameters.

Table2.2.Primary
Table Primaryanalysis:
analysis:radio-symptomatic
radio-symptomatic progression
progression on
ontrabecular
trabecularbone
bonetexture
textureparameters
parameters
adjustedfor
adjusted forclinical,
clinical,biochemical,
biochemical,and
andradiographic
radiographicparameters.
parameters.

N°N◦
Model
Model
BACC
BACC
PPVPPV
NPV
NPV
AUC (95%CI)
AUC (95%CI)
p-Value
p-Value
Model 1 TBT ← CLIN + KL + JSNM * 0.55 0.50 0.70 0.613 (0.565–0.662) -
Model 1 TBT ← CLIN + KL + JSNM * 0.55 0.50 0.70 0.613 (0.565–0.662) -
Model 2 ∆TBT ← CLIN + KL + JSNM 0.53 0.45 0.70 0.606 (0.558–0.653) 0.777
Model 2 ∆TBT ← CLIN + KL + JSNM 0.53 0.45 0.70 0.606 (0.558–0.653) 0.777
Model 3 TBT + ∆TBT ← CLIN + KL + JSNM 0.59 0.52 0.73 0.650 (0.603–0.697) 0.044
Model43
Model TBT + ∆TBT
TBT←+ CLIN
∆TBT+ KL + JSNM 0.59
0.58 0.510.52 0.73
0.72 0.650 (0.603–0.697)
0.658 (0.612–0.705) 0.044
0.030
Model54
Model TBT + ∆TBT ← BIO ∆TBT+ KL + JSNM
TBT++CLIN 0.58
0.59 0.520.51 0.72
0.73 0.658 (0.612–0.705)
0.649 (0.601–0.696) 0.030
0.054
Model
Model65 TBTM ∆TBT ← ←
TBT + ∆TBTM BIOCLIN + KL
+ CLIN + JSNM
+ KL + JSNM 0.53
0.59 0.520.52 0.69
0.73 0.594 (0.545–0.643)
0.649 (0.601–0.696) 0.456
0.054
Model
Model76 TBTL + ∆TBTL
TBTM ← CLIN
+ ∆TBTM ← CLIN+ KL + JSNM
+ KL + JSNM 0.51
0.53 0.460.52 0.68
0.69 0.577 (0.527–0.627)
0.594 (0.545–0.643) 0.173
0.456
Model
Model87 TBTC + ∆TBTC ←
TBTL + ∆TBTL ← CLIN + KL +JSNM
CLIN + KL + JSNM 0.51
0.51 0.460.46 0.68
0.68 0.572 (0.524–0.620)
0.577 (0.527–0.627) 0.131
0.173
* refers to the reference model. BACC, PPV, and NPV refer to balanced accuracy, positive predictive
Model 8 TBTC + ∆TBTC ← CLIN + KL + JSNM 0.51 0.46 0.68 0.572 (0.524–0.620) 0.131
value, and negative predictive value. TBT refers to the baseline trabecular bone texture parameters
* while
refers to the reference
∆TBT refers tomodel. BACC, PPV,
the variations and NPV
in TBT over refer to balanced
24 months. CLINaccuracy,
and BIO positive
refer predictive value, clin-
to the baseline and
negative predictive value. TBT refers to the baseline trabecular bone texture parameters while ∆TBT refers to the
ical (age, sex, and BMI) and biochemical (urine CTXII, Serum NTXI, and Serum HA) parameters,
variations in TBT over 24 months. CLIN and BIO refer to the baseline clinical (age, sex, and BMI) and biochemical
respectively.
(urine CTXII, SerumKL refers to the
NTXI, and baseline
Serum Kellgren–Lawrence
HA) parameters, respectively.scores. JSNM
KL refers to therefers to Kellgren–Lawrence
baseline the baseline joint
space JSNM
scores. narrowing scores
refers to in the medial
the baseline tibial
joint space plateau.scores
narrowing The model with the
in the medial descriptor
tibial plateau. Theonmodel
the left of (←)
with the
descriptor
is adjusted on for
the the
left of (←) is adjusted
descriptor(s) on for
thethe descriptor(s)
right on the TBTC,
of (←). TBTM, (←). TBTL
right of and TBTM,refer
TBTC,toand
theTBTL refer to
TBT param-
the TBTextracted
eters parameters extracted
from from the
the medial, medial,and
central, central, andplateaus,
lateral lateral plateaus, respectively.
respectively.

Regarding the influence of the regions of interest (ROIs) selected for the calculation of
3.2. Secondary Analysis
TBT, the use of the complete subchondral zone (Model 3) provided higher performance
3.2.1. Any
(AUC Progression
= 0.650) compared to using medial-only (Model 6: AUC = 0.594), lateral-only (Model
7: AUC In =
this scenario,
0.577), the performance
or central-only (Modelof8:the
AUC different models
= 0.572) was evaluated
ROIs (Table 2). More for the pre-
details are
diction ofinradiographic,
reported Table S1 of thesymptomatic, or both
Supplementary File. progressions using data from Groups 1, 2,
and 3 (397 progressors) and Group 4 (200 controls). The AUC score of Model 3, based on
baseline TBT and its 24-month variation, was significantly higher (AUC = 0.679; p-value =
Biomedicines 2024, 12, 666 6 of 11

3.2. Secondary Analysis

3.2.1. Any Progression
In this scenario, the performance of the different models was evaluated for the predic-
tion of radiographic, symptomatic, or both progressions using data from Groups 1, 2, and 3
(397 progressors) and Group 4 (200 controls). The AUC score of Model 3, based on baseline
TBT and its 24-month variation, was significantly higher (AUC = 0.679; p-value = 0.009)
than the one obtained by Model 1, based solely on baseline TBT (AUC = 0.628). Model
3 furthermore achieved the best BACC (0.60) and PPV (0.72) values, while Model 5 with
biochemical parameters, in addition, achieved slightly higher NPV (0.53) values (Table 3).
In this scenario, we noticed that using the whole subchondral zone (Model 3) provided
better performance than using medial-only (AUC = 0.614), lateral-only (AUC = 0.596), or
central-only ROI (AUC = 0.580) compartments. Further information can be found in Table
S2 of the Supplementary File.

Table 3. Secondary analysis based on TBT and ∆TBT adjusted for a set of symptomatic and radio-
graphic parameters.

Progression Model BACC PPV NPV AUC (95%CI) p-Value

TBT ← CLIN + KL + JSNM * 0.55 0.69 0.46 0.628 (0.582–0.675) -
Any
TBT + ∆TBT ← CLIN + KL + JSNM 0.60 0.72 0.52 0.679 (0.634–0.724) 0.009
(Scenario 1)
TBT + ∆TBT ← BIO + CLIN + KL + JSNM 0.60 0.72 0.52 0.678 (0.633–0.723) 0.011
TBT ← CLIN + KL + JSNM * 0.59 0.59 0.58 0.628 (0.574–0.682) -
All
TBT + ∆TBT ← CLIN + KL + JSNM 0.63 0.64 0.63 0.684 (0.632–0.736) 0.022
(Scenario 2)
TBT + ∆TBT ← BIO + CLIN + KL + JSNM 0.63 0.64 0.62 0.684 (0.632–0.735) 0.023
TBT ← CLIN + KL + JSNM * 0.57 0.45 0.78 0.709 (0.653–0.765) -
Radiographic
TBT + ∆TBT ← CLIN + KL + JSNM 0.65 0.51 0.82 0.779 (0.731–0.827) 0.012
(Scenario 3)
TBT + ∆TBT ← BIO + CLIN + KL + JSNM 0.65 0.51 0.82 0.779 (0.731–0.828) 0.011
TBT ← CLIN + KL + JSNM * 0.53 0.40 0.76 0.643 (0.583–0.703) -
Symptomatic
TBT + ∆TBT ← CLIN + KL + JSNM 0.60 0.49 0.79 0.710 (0.654–0.766) 0.027
(Scenario 4)
TBT + ∆TBT ← BIO + CLIN + KL + JSNM 0.60 0.49 0.79 0.710 (0.654–0.766) 0.027
* refers to the reference model. BACC, PPV, and NPV refer to balanced accuracy, positive predictive value, and
negative predictive value. The model with the descriptor on the left of (←) is adjusted for the descriptor(s) on the
right of (←).

3.2.2. All Progression

In this scenario, the performance of the aforementioned models was evaluated for the
prediction of radiographic or symptomatic progression using data from Groups 2 and 3
(205 progressors) and Group 4 (200 controls). Model 4, adjusted in addition to biochemical
parameters, obtained the best AUC score (AUC = 0.691, p-value = 0.018), significantly
higher than the one obtained by the reference model (AUC = 0.628). The best values were
obtained by Model 4, adjusted in addition to biochemical parameters and JSNM scores, for
the other statistical metrics (BACC = 0.64, PPV = 0.65, and NPV = 0.64). More details can be
found in Table S3 of the Supplementary File.

3.2.3. Radiographic Progression

In this scenario, the performance of the abovementioned models was evaluated for
the prediction of radiographic-only progression using data from Group 2 (102 progressors)
in addition to Groups 3 and 4 (303 controls). In this scenario, all knees with radiographic
and symptomatic progression were excluded. The best overall performance was achieved
by Model 4, adjusted in addition to clinical and biochemical parameters (Model 4B), with
an AUC of 0.783, BACC of 0.66, PPV of 0.52, and NPV of 0.82, while the reference model
achieved a significantly lower AUC of only 0.709 (p-value = 0.009), and lower values for
Biomedicines 2024, 12, 666 7 of 11

BACC, PPV, and NPV (0.55, 0.47, 0.77, respectively). In line with the other scenarios,
higher performance was achieved using the TBT parameters extracted from the complete
subchondral bone, compared to using the TBT parameters extracted from the medial,
lateral, or central compartments. Further information can be found in Table S4 of the
Supplementary File.

3.2.4. Symptomatic Progression

In this scenario, all knees with radiographic and symptomatic progression were also
excluded. The performance of the different models was evaluated for the prediction of
symptomatic-only progression using data from Group 3 (103 progressors) and Groups
2 and 4 (302 controls). As found in the previous scenario, the best overall performance
was achieved by Model 4B with an AUC of 0.718, BACC of 0.61, PPV of 0.51, and NPV
of 0.79. As for the previously mentioned scenarios, the models using the TBT parameters
extracted from the complete subchondral bone achieved a higher overall performance
than the models using the TBT parameters extracted from the medial, lateral, or central
compartments. Further information can be found in Table S5 of the Supplementary File.

4. Discussion
One of the main contributions of this study is the evaluation of predictive performance
gained from utilizing longitudinal variations in TBT parameters adjusted by a set of clinical,
biochemical, and radiological biomarkers for predicting KOA progression. The results in the
present study show that integrating both baseline and longitudinal changes in radiographic
TBT descriptors plays an important role in predicting radio-symptomatic progression (best
AUC = 0.658), any (radiographic, symptomatic, or both) progression (best AUC = 0.679), all
(radiographic or symptomatic) progression (best AUC = 0.691), radiographic progression
(best AUC = 0.718), and symptomatic progression (best AUC = 0.783).
KOA progression is often considered to be slow; 12% to 23% of knees with radiographic
KOA experience radiographic progression over 5 years [26]. Hence, the findings in the
current study would help in better selecting participants in future structure-modifying
KOA trials.
The use of baseline, 12-month, and 24-month TBT parameters was previously evalu-
ated [7] for the prediction of 48-month radiographic and symptomatic progression in the
FNIH cohort. In that study, the medial subchondral tibial region only was investigated to
extract TBT parameters computed using the fractal signature analysis (FSA) method [20].
Introducing the time-integrated values (TIVs) of the TBT parameters over 24 months pro-
vided a benefit to the prediction of KOA radio-symptomatic progression (primary analysis),
with an AUC of 0.649, compared to the use of clinical covariates alone (AUC = 0.608) [7].
While the authors in [7,27] investigated the use of the 24-month TIVs of TBT parameters,
equivalent to the area under the curve defined by the baseline and 24-month TBT values,
they did not investigate the use of time-longitudinal changes, quantified as the difference
between the baseline and 24-month TBT values. In addition, in their study [7,27], six TBT
parameters were employed, extracted from the medial tibial plateau only.
This study also highlights the importance of exploiting the whole subchondral bone
of the tibia, rather than only the medial plateau or a limited part of the medial and lateral
plateaus [5,7], to extract radiographic TBT parameters as the performance of the prediction
models was lower in all the different scenarios using TBT parameters extracted from the
medial, lateral or central tibial plateaus alone (Tables S1–S5 of the Supplementary File).
The best AUC score was obtained for the prediction of radiographic progression
(AUC = 0.783). It is more difficult to predict symptomatic (pain) progression since it
is related to changes in pain scores, which are subjective due to differences in patient
tolerance. In addition, it has been demonstrated that pain scores can only be considered
modest markers in the prediction of KOA-related outcomes [6,11].
The results obtained by the present study confirm the interest in using both baseline
TBT parameters and their variations over 24 months, allowing a better prediction of radio-
Biomedicines 2024, 12, 666 8 of 11

symptomatic, radiographic, or symptomatic progression. When TBT descriptors were

excluded, all assessed models had failed (0.5 ≤ AUC < 0.6) to poor (0.6 ≤ AUC < 0.7)
values; however, when TBT descriptors were included, and their variations over 24 months
were taken into account, the evaluated models’ performances improved and they were able
to obtain acceptable (0.7 < AUC < 0.8) values [28,29].
Validated on the FNIH dataset, this research demonstrated the benefits of using
both baseline and longitudinal changes in TBT, calculated from standardized plain knee
radiographs, to improve the prediction of KOA progression within 48 months in patients
with mild KOA (knees with 1 ≤ KL ≤ 3) at baseline.
In the present study, adding molecular biomarkers into the model to the core set
of radiographic and clinical markers did not improve the performance of the reference
model. The AUC scores of the reference model (AUC = 0.709 and 0.779 using TBT and
TBT + ∆TBT, respectively) were similar to those obtained by including molecular biomark-
ers (AUC = 0.708 and 0.779 using TBT and TBT + ∆TBT, respectively) for the prediction
of radiological progression. This observation could be because the information given
by biochemical markers is already captured by data from subchondral bone (its texture),
osteophytes, and joint space width, which are known as strong predictors of KOA progres-
sion [4,10,11,28,30]. Similar results have been found for the other scenarios (Table 1). The
three molecular biomarkers used in the present study were selected based on their formerly
successful use in the literature concerning the prediction of KOA progression [8,13]. Other
relevant parameters might be used to improve KOA prediction models such as type II
collagen KOA formation [12] or inter-alpha trypsin inhibitor heavy chain 1 [31].
Our study has several strengths. The proposed prediction models are based on TBT
descriptors extracted from plain radiographs, widely used in clinical routine, while bio-
chemical parameters are not yet included in daily clinical routines and need much more time
to be extracted from blood or urine samples. Furthermore, data were selected in accordance
with each specific type of progression evaluated in the current study. For radiographic-only
progression, knees with symptomatic progression were not included, and vice versa; for
symptomatic-only progression, knees with radiographic progression were not included. In
addition, for radio-symptomatic progression, knees with symptomatic-only or radiographic-
only progression were not included. To avoid possible correlation between the TBT pa-
rameters of both the baseline and 24-month variations, which can lead to problems with
traditional logistic regression with respect to overfitting and convergence, the LASSO
method was used as an alternative regularization method [32].
A growing number of researchers are interested in evaluating the potential of imaging
biomarkers to enhance patient screening in phase III studies for KOA and determining
under which conditions they provide such enhancements [4,8]. From a clinical standpoint,
our model showed great precision in predicting false progressors. Incorporating such
progressors in a disease-modifying osteoarthritis drug (DMOAD) randomized clinical trial
could have counterproductive consequences.

5. Study Limitations
The limitations of this study include the absence of a femoral region of interest in our
TBT analysis. The femoral subchondral bone might also provide additional information [33].
In the current study, the included imaging-based biomarkers were limited to radiography.
Another limitation is the lack of investigation of the association between radiographic
KOA progression and 3D MRI bone texture [34] or shape [35], or other MRI-based features
such as bone marrow lesions [36]. Integrating such parameters into our model is worth
assessing. Conducting a comparative analysis of the results of prediction methods based
on logistic regression and those derived from alternative machine learning techniques may
provide valuable insights [37]. Lastly, combining descriptors extracted from both MRI and
XR images might help to improve the prediction of KOA progression [27,34].
Furthermore, there are several potentially interesting molecular markers for predicting
KOA progression [14]. Theoretically, there is no limit to the number of biochemical markers
Biomedicines 2024, 12, 666 9 of 11

that could be included in the model, although estimating their ability to enhance our
prediction model becomes increasingly challenging.

6. Conclusions
In conclusion, the combination of both baseline and 24-month radiographic TBT varia-
tions can increase modestly, but significantly, the ability to predict 48-month radiographic,
symptomatic, and radio-symptomatic KOA progressions. Adding the three aforementioned
studied molecular markers did not significantly improve the performance of the proposed
TBT∆TBT model.

Supplementary Materials: The following supporting information can be downloaded at: https://
www.mdpi.com/article/10.3390/biomedicines12030666/s1, Table S1: Primary analysis: Prediction
of radio-symptomatic progression; Table S2: Secondary analysis: Prediction of any (knees with
either radiographic or symptomatic progression, or both) progression; Table S3: Secondary analysis:
Prediction of all (knees with either radiographic or symptomatic progression) progression; Table S4:
Secondary analysis: Prediction of radiographic progression; Table S5: Secondary analysis: Prediction
of symptomatic progression.
Author Contributions: Conceptualization, A.A.-I. and E.L.; Methodology, A.A.-I. and E.L.; Formal
Analysis, A.A.-I. and E.L.; Data Curation, A.A.-I.; Writing—Original Draft Preparation, A.A.-I. and
E.L.; Writing—Review and Editing, all.; Supervision, E.L.; Funding Acquisition, H.T. All authors
have read and agreed to the published version of the manuscript.
Funding: The study was funded by the European Regional Development Fund (ERDF) and the City
of Orleans, France, under Project reference EX004579.
Institutional Review Board Statement: The OAI study was authorized by the institutional review
boards at each OAI clinical site and was performed in accordance with the Declaration of Helsinki
(approval code: 10-00532).
Informed Consent Statement: Participants involved in the OAI study provided informed consent.
Data Availability Statement: Publicly available datasets were analyzed in this study. This data can
be found online at https://nda.nih.gov/oai (accessed on 20 September 2023).
Acknowledgments: All authors would like to thank the participants and staff of the OAI study,
particularly the FNIH OA Biomarkers Consortium. They gratefully acknowledge the European
Regional Development Fund (ERDF) for financial support.
Conflicts of Interest: The authors declare no conflicts of interest. The funder had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or
in the decision to publish the results.

References
1. Attur, M.; Krasnokutsky, S.; Zhou, H.; Samuels, J.; Chang, G.; Bencardino, J.; Rosenthal, P.; Rybak, L.; Huebner, J.L.; Kraus, V.B.;
et al. The combination of an inflammatory peripheral blood gene expression and imaging biomarkers enhance prediction of
radiographic progression in knee osteoarthritis. Arthritis Res. Ther. 2020, 22, 208. [CrossRef]
2. Cai, G.; Otahal, P.; Cicuttini, F.; Wu, F.; Munugoda, I.P.; Jones, G.; Aitken, D. The association of subchondral and systemic bone
mineral density with osteoarthritis-related joint replacements in older adults. Osteoarthr. Cartil. 2020, 28, 438–445. [CrossRef]
3. Felson, D.T.; Gale, D.R.; Elon Gale, M.; Niu, J.; Hunter, D.J.; Goggins, J.; Lavalley, M.P. Osteophytes and progression of knee
osteoarthritis. Rheumatol. Oxf. Engl. 2005, 44, 100–104. [CrossRef] [PubMed]
4. Almhdie-Imjabbar, A.; Podsiadlo, P.; Ljuhar, R.; Jennane, R.; Nguyen, K.-L.; Toumi, H.; Saarakkala, S.; Lespessailles, E. Trabecular
bone texture analysis of conventional radiographs in the assessment of knee osteoarthritis: Review and viewpoint. Arthritis Res.
Ther. 2021, 23, 208. [CrossRef]
5. Woloszynski, T.; Podsiadlo, P.; Stachowiak, G.W.; Kurzynski, M.; Lohmander, L.S.; Englund, M. Prediction of progression of
radiographic knee osteoarthritis using tibial trabecular bone texture. Arthritis Rheum. 2012, 64, 688–695. [CrossRef] [PubMed]
6. Almhdie-Imjabbar, A.; Toumi, H.; Harrar, K.; Pinti, A.; Lespessailles, E. Subchondral tibial bone texture of conventional X-rays
predicts total knee arthroplasty. Sci. Rep. 2022, 12, 8327. [CrossRef] [PubMed]
Biomedicines 2024, 12, 666 10 of 11

7. Kraus, V.B.; Collins, J.E.; Charles, H.C.; Pieper, C.F.; Whitley, L.; Losina, E.; Nevitt, M.; Hoffmann, S.; Roemer, F.; Guermazi, A.;
et al. Predictive Validity of Radiographic Trabecular Bone Texture in Knee Osteoarthritis: The Osteoarthritis Research Society
International/Foundation for the National Institutes of Health Osteoarthritis Biomarkers Consortium. Arthritis Rheumatol. 2018,
70, 80–87. [CrossRef] [PubMed]
8. Kraus, V.B.; Collins, J.E.; Hargrove, D.; Losina, E.; Nevitt, M.; Katz, J.N.; Wang, S.X.; Sandell, L.J.; Hoffmann, S.C.; Hunter, D.J.;
et al. Predictive validity of biochemical biomarkers in knee osteoarthritis: Data from the FNIH OA Biomarkers Consortium. Ann.
Rheum. Dis. 2017, 76, 186–195. [CrossRef] [PubMed]
9. Janvier, T.; Jennane, R.; Toumi, H.; Lespessailles, E. Subchondral tibial bone texture predicts the incidence of radiographic knee
osteoarthritis: Data from the Osteoarthritis Initiative. Osteoarthr. Cartil. 2017, 25, 2047–2054. [CrossRef]
10. Janvier, T.; Jennane, R.; Valery, A.; Harrar, K.; Delplanque, M.; Lelong, C.; Loeuille, D.; Toumi, H.; Lespessailles, E. Subchondral
tibial bone texture analysis predicts knee osteoarthritis progression: Data from the Osteoarthritis Initiative: Tibial bone texture &
knee OA progression. Osteoarthr. Cartil. 2017, 25, 259–266.
11. Almhdie-Imjabbar, A.; Nguyen, K.-L.; Toumi, H.; Jennane, R.; Lespessailles, E. Prediction of knee osteoarthritis progression using
radiological descriptors obtained from bone texture analysis and Siamese neural networks: Data from OAI and MOST cohorts.
Arthritis Res. Ther. 2022, 24, 66. [CrossRef]
12. Bay-Jensen, A.C.; Manginelli, A.A.; Karsdal, M.; Luo, Y.; He, Y.; Michaelis, M.; Guehring, H.; Ladel, C. Low levels of type II
collagen formation (PRO-C2) are associated with response to sprifermin: A pre-defined, exploratory biomarker analysis from the
FORWARD study. Osteoarthr. Cartil. 2022, 30, 92–99. [CrossRef] [PubMed]
13. Bihlet, A.R.; Bjerre-Bastos, J.J.; Andersen, J.R.; Byrjalsen, I.; Karsdal, M.A.; Bay-Jensen, A.-C. Clinical and biochemical factors
associated with risk of total joint replacement and radiographic progression in osteoarthritis: Data from two phase III clinical
trials. Semin. Arthritis Rheum. 2020, 50, 1374–1381. [CrossRef] [PubMed]
14. Liu, M.; Haque, N.; Huang, J.; Zhai, G. Osteoarthritis year in review 2023: Metabolite and protein biomarkers. Osteoarthr. Cartil.
2023, 31, 1437–1453. [CrossRef]
15. Lotz, M.; Martel-Pelletier, J.; Christiansen, C.; Brandi, M.-L.; Bruyère, O.; Chapurlat, R.; Collette, J.; Cooper, C.; Giacovelli, G.;
Kanis, J.A.; et al. Value of biomarkers in osteoarthritis: Current status and perspectives. Ann. Rheum. Dis. 2013, 72, 1756–1763.
[CrossRef]
16. Hosnijeh, F.S.; Runhaar, J.; van Meurs, J.B.J.; Bierma-Zeinstra, S.M. Biomarkers for osteoarthritis: Can they be used for risk
assessment? A systematic review. Maturitas 2015, 82, 36–49. [CrossRef]
17. Nelson, A.E.; Fang, F.; Arbeeva, L.; Cleveland, R.J.; Schwartz, T.A.; Callahan, L.F.; Marron, J.S.; Loeser, R.F. A machine learning
approach to knee osteoarthritis phenotyping: Data from the FNIH Biomarkers Consortium. Osteoarthr. Cartil. 2019, 27, 994–1001.
[CrossRef]
18. Ying, G.-S.; Maguire, M.G.; Glynn, R.; Rosner, B. Tutorial on Biostatistics: Statistical Analysis for Correlated Binary Eye Data.
Ophthalmic Epidemiol. 2018, 25, 1–12. [CrossRef] [PubMed]
19. Hunter, D.J.; Collins, J.E.; Deveza, L.; Hoffmann, S.C.; Kraus, V.B. Biomarkers in osteoarthritis: Current status and outlook—The
FNIH Biomarkers Consortium PROGRESS OA study. Skeletal Radiol. 2023, 52, 2323–2339. [CrossRef]
20. Lynch, J.A.; Hawkes, D.J.; Buckland-Wright, J.C. A robust and accurate method for calculating the fractal signature of texture in
macroradiographs of osteoarthritic knees. Med. Inform. Med. Inform. 1991, 16, 241–251. [CrossRef]
21. Lee, C.Y. Nested logistic regression models and ∆AUC applications: Change-point analysis. Stat. Methods Med. Res. 2021, 30,
1654–1666. [CrossRef]
22. Bradley, A.P. The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognit. 1997,
30, 1145–1159. [CrossRef]
23. Hosmer, D.W.; Lemeshow, S.; Sturdivant, R.X. Assessing the Fit of the Model. In Applied Logistic Regression; John Wiley & Sons:
New York, NY, USA, 2013.
24. Sharma, D.K.; Chatterjee, M.; Kaur, G.; Vavilala, S. 3-Deep learning applications for disease diagnosis. In Deep Learning for Medical
Applications with Unique Data; Academic Press: New York, NY, USA, 2022; pp. 31–51. [CrossRef]
25. DeLong, E.R.; DeLong, D.M.; Clarke-Pearson, D.L. Comparing the areas under two or more correlated receiver operating
characteristic curves: A nonparametric approach. Biometrics 1988, 44, 837–845. [CrossRef]
26. Leyland, K.M.; Hart, D.J.; Javaid, M.K.; Judge, A.; Kiran, A.; Soni, A.; Goulston, L.M.; Cooper, C.; Spector, T.D.; Arden, N.K. The
natural history of radiographic knee osteoarthritis: A fourteen-year population-based cohort study. Arthritis Rheum. 2012, 64,
2243–2251. [CrossRef]
27. Hunter, D.J.; Deveza, L.A.; Collins, J.E.; Losina, E.; Nevitt, M.C.; Roemer, F.W.; Guermazi, A.; Bowes, M.A.; Dam, E.B.; Eckstein, F.;
et al. Multivariable modeling of biomarker data from the phase 1 Foundation for the NIH Osteoarthritis Biomarkers Consortium.
Arthritis Care Res. 2022, 74, 1142–1153. [CrossRef] [PubMed]
28. Almhdie-Imjabbar, A.; Toumi, H.; Lespessailles, E. Radiographic Biomarkers for Knee Osteoarthritis: A Narrative Review. Life
2023, 13, 237. [CrossRef] [PubMed]
29. Nahm, F.S. Receiver operating characteristic curve: Overview and practical use for clinicians. Korean J. Anesthesiol. 2022, 75, 25–36.
[CrossRef] [PubMed]
30. van der Kraan, P.M.; van den Berg, W.B. Osteophytes: Relevance and biology. Osteoarthr. Cartil. 2007, 15, 237–244. [CrossRef]
[PubMed]
Biomedicines 2024, 12, 666 11 of 11

31. Lourido, L.; Balboa-Barreiro, V.; Ruiz-Romero, C.; Rego-Pérez, I.; Camacho-Encina, M.; Paz-González, R.; Calamia, V.; Oreiro, N.;
Nilsson, P.; Blanco, F.J. A clinical model including protein biomarkers predicts radiographic knee osteoarthritis: A prospective
study using data from the Osteoarthritis Initiative. Osteoarthr. Cartil. 2021, 29, 1147–1154. [CrossRef] [PubMed]
32. Akalin, A. Computational Genomics with R: Logistic Regression and Regularization; CRC Press: New York, NY, USA; Taylor & Francis
Group: Boca Raton, FL, USA, 2023.
33. Hirvasniemi, J.; Thevenot, J.; Immonen, V.; Liikavainio, T.; Pulkkinen, P.; Jämsä, T.; Arokoski, J.; Saarakkala, S. Quantification of
differences in bone texture from plain radiographs in knees with and without osteoarthritis. Osteoarthr. Cartil. 2014, 22, 1724–1731.
[CrossRef] [PubMed]
34. MacKay, J.; Kapoor, G.; Driban, J.; Lo, G.; McAlindon, T.; Toms, A.; McCaskie, A.; Gilbert, F. Association of subchondral bone
texture on magnetic resonance imaging with radiographic knee osteoarthritis progression: Data from the Osteoarthritis Initiative
Bone Ancillary Study. Eur. Radiol. 2018, 28, 4687–4695. [CrossRef]
35. Chang, G.H.; Park, L.K.; Le, N.A.; Jhun, R.S.; Surendran, T.; Lai, J.; Seo, H.; Promchotichai, N.; Yoon, G.; Scalera, J.; et al.
Subchondral bone length in knee osteoarthritis: A deep learning derived imaging measure and its association with radiographic
and clinical outcomes. Arthritis Rheumatol. 2021, 73, 2240–2248. [CrossRef] [PubMed]
36. Felson, D.T.; McLaughlin, S.; Goggins, J.; LaValley, M.P.; Gale, M.E.; Totterman, S.; Li, W.; Hill, C.; Gale, D. Bone marrow edema
and its relation to progression of knee osteoarthritis. Ann. Intern. Med. 2003, 139, 330–336. [CrossRef] [PubMed]
37. Widera, P.; Welsing, P.; Ladel, C.; Loughlin, J.; Lafeber, F.; Petit Dop, F.; Larkin, J.; Weinans, H.; Mobasheri, A.; Bacardit, J. Multi-
classifier prediction of knee osteoarthritis progression from incomplete imbalanced longitudinal data. Sci. Rep. 2020, 10, 8427.
[CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.