Developing and validating equations to predict V̇O 2 peak from the 6MWT in Childhood ALL Survivors

Abstract

Introduction: The 6-Minute Walking Test (6MWT) is a safe, standardized and well utilized method to assess the functional capacity. Recently, it was reported that the published prediction equations cannot accurately predict a valid maximal oxygen consumption ( V̇O2 peak) value in cancer survivors. Thus, the aim of this study was to establish and to validate a new equation based on the 6MWT to predict V̇O2 peak in childhood acute lymphoblastic leukemia (ALL) survivors.
Methods: A total of 200 childhood ALL survivors were enrolled in this study, among which 168 participants underwent a cardiopulmonary exercise test and a 6MWT to assess their functional capacity and their cardiorespiratory fitness. In addition, participants completed a physical activity questionnaire. Participants were randomly divided in two groups to establish the equations (n = 118 (70%)) and to validate it (n = 50 (30%)). Multiple linear regression analyses were used to determine a new prediction equation for V̇O2 peak from 6MWT using clinical and specific variables related to the disease. The accuracy in between V̇O2 peak measured and V̇O2 peak predicted were assessed using the Bland and Altman method.
Results: The new establish clinical V̇O2 peak equation is: V̇O2 peak (mL.kg⁻¹.min⁻¹) = (−0.283*age(years)) − (0.099*weight(kg)) + (0.071*6MWD(meters)) −(0.135*HR end(bpm)) + 22.789 with a mean bias of 2.67 mL.kg⁻¹.min⁻¹ (95% CI (−9.64 to 14.98)). The new establish disease-specific V̇O2 peak equation is: V̇O2 peak (mL.kg⁻¹.min⁻¹) = (−0.236*age(years)) − (0.094*weight(kg)) −(0.120*HR end(bpm)) + (0.067*6MWD(meters)) + (0.065*MVLPA(min/day)) − (0.204*DT(years)) + 25.145 with a mean bias of 2.51 mL.kg⁻¹.min⁻¹ (95% CI (−9.98 to 15.01)).
Conclusion: This is the first study that predicted V̇O2 peak from a 6MWT using clinical and specific variables related to the disease of childhood ALL survivors. The availability of these newly established V̇O2 peak equations makes them an accurate tool to provide a better follow-up and better adapted physical training for survivors. We invite researchers to use our assessment procedures for their further studies.

• IMPLICATIONS FOR REHABILITATION
• It is critical to understand the cardiorespiratory fitness of the childhood ALL survivors
• The maximal oxygen consumption (i.e., V̇O2 peak) is recognized as the gold standard to measure the patient’s cardiorespiratory fitness in the field of exercise physiology
• This study is novel and reports the validation of two new VO2 peak equations, from 6MWT, by using clinical and disease-specific variables of childhood ALL survivors
• The availability of such validated equations can better facilitate the follow-up of survivors' cardiorespiratory fitness, by relevant health care professionals and exercise physiologists.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=idre20
Disability and Rehabilitation
ISSN: 0963-8288 (Print) 1464-5165 (Online) Journal homepage: https://www.tandfonline.com/loi/idre20
Developing and validating equations to predict
VO2 peak from the 6MWT in Childhood ALL
Survivors
Jennifer Labonté, Maxime Caru, Valérie Lemay, Nathalie Alos, Simon Drouin,
Laurence Bertout, Gregor Andelfinger, Maja Krajinovic, Caroline Laverdière,
Daniel Sinnett & Daniel Curnier
To cite this article: Jennifer Labonté, Maxime Caru, Valérie Lemay, Nathalie Alos, Simon Drouin,
Laurence Bertout, Gregor Andelfinger, Maja Krajinovic, Caroline Laverdière, Daniel Sinnett &
Daniel Curnier (2020): Developing and validating equations to predict VO2 peak from the 6MWT in
Childhood ALL Survivors, Disability and Rehabilitation, DOI: 10.1080/09638288.2020.1725159
To link to this article: https://doi.org/10.1080/09638288.2020.1725159
Published online: 11 Feb 2020.
Submit your article to this journal
View related articles
View Crossmark data

ORIGINAL ARTICLE
Developing and validating equations to predict
_
VO
2
peak from the 6MWT in
Childhood ALL Survivors
Jennifer Labont
e
a,b
, Maxime Caru
a,b,c
, Val
erie Lemay
a,b
, Nathalie Alos
b,d
, Simon Drouin
b
, Laurence Bertout
b
,
Gregor Andelfinger
b,d
, Maja Krajinovic
b,d
, Caroline Laverdi
ere
b,d
, Daniel Sinnett
b,d
and Daniel Curnier
a,b
a
Laboratory of Pathophysiology of EXercise (LPEX), School of Kinesiology and Physical Activity Sciences, Faculty of Medicine, University of
Montreal, Montreal, Canada;
b
Research Center, Sainte-Justine University Health Center, Montreal, Canada;
c
Laboratoire EA 4430 –Clinique
Psychanalyse Developpement (CliPsyD), Department of psychology, University of Paris Nanterre, Nanterre, France;
d
Department of Pediatrics,
University of Montreal, Montreal, Canada
ABSTRACT
Introduction: The 6-Minute Walking Test (6MWT) is a safe, standardized and well utilized method to
assess the functional capacity. Recently, it was reported that the published prediction equations cannot
accurately predict a valid maximal oxygen consumption ( _
VO
2
peak) value in cancer survivors. Thus, the
aim of this study was to establish and to validate a new equation based on the 6MWT to predict _
VO
2
peak in childhood acute lymphoblastic leukemia (ALL) survivors.
Methods: A total of 200 childhood ALL survivors were enrolled in this study, among which 168 partici-
pants underwent a cardiopulmonary exercise test and a 6MWT to assess their functional capacity and
their cardiorespiratory fitness. In addition, participants completed a physical activity questionnaire.
Participants were randomly divided in two groups to establish the equations (n¼118 (70%)) and to valid-
ate it (n¼50 (30%)). Multiple linear regression analyses were used to determine a new prediction equa-
tion for _
VO
2
peak from 6MWT using clinical and specific variables related to the disease. The accuracy in
between _
VO
2
peak measured and _
VO
2
peak predicted were assessed using the Bland and
Altman method.
Results: The new establish clinical _
VO
2
peak equation is: _
VO
2
peak (mL.kg
1
.min
1
)¼(0.283age(years))
(0.099weight(kg)) þ(0.0716MWD(meters)) (0.135HR end(bpm)) þ22.789 with a mean bias of
2.67 mL.kg
1
.min
1
(95% CI (9.64 to 14.98)). The new establish disease-specific _
VO
2
peak equation is:
_
VO
2
peak (mL.kg
1
.min
1
)¼(0.236age(years)) (0.094weight(kg)) (0.120HR end(bpm)) þ
(0.0676MWD(meters)) þ(0.065MVLPA(min/day)) (0.204DT(years)) þ25.145 with a mean bias of
2.51 mL.kg
1
.min
1
(95% CI (9.98 to 15.01)).
Conclusion: This is the first study that predicted _
VO
2
peak from a 6MWT using clinical and specific varia-
bles related to the disease of childhood ALL survivors. The availability of these newly established _
VO
2
peak equations makes them an accurate tool to provide a better follow-up and better adapted physical
training for survivors. We invite researchers to use our assessment procedures for their further studies.
äIMPLICATIONS FOR REHABILITATION
It is critical to understand the cardiorespiratory fitness of the childhood ALL survivors
The maximal oxygen consumption (i.e., _
VO
2
peak) is recognized as the gold standard to measure the
patient’s cardiorespiratory fitness in the field of exercise physiology
This study is novel and reports the validation of two new VO
2
peak equations, from 6MWT, by using
clinical and disease-specific variables of childhood ALL survivors
The availability of such validated equations can better facilitate the follow-up of survivors’cardio-
respiratory fitness, by relevant health care professionals and exercise physiologists.
ARTICLE HISTORY
Received 16 October 2019
Revised 27 January 2020
Accepted 29 January 2020
KEYWORDS
Acute lymphoblastic
leukemia; pediatric cancer
survivorship; cardiorespira-
tory fitness; 6-minute walk
test; exercise
testing; equation
Introduction
Acute lymphoblastic leukemia (ALL) is the most frequent type of
pediatric cancer diagnosed in Canada, with approximately 30% of
all the new cancer cases every year [1]. Medical progress and che-
motherapeutic treatments allow these children to overcome the
disease with a 5-year survival rate of 90% [2]. However, it has
been observed that these cancer treatments cause several long-
term adverse effects in childhood ALL survivors [3]. Additionally,
survivors suffer from at least one severe chronic disease even sev-
eral years after the end of treatments [3–5]. The most reported
long-term effects from survivors are cardiac toxicity and cardiovas-
cular issues; especially in those who had been treated with
anthracycline or irradiation in the cardiac region [6–9]. Some
studies also showed a decrease in the respiratory and physical
function of survivors [10,11] as well as a decrease of their cardio-
respiratory fitness [12–14].
In this sense, the maximal oxygen consumption (i.e., _
VO
2
peak)
is recognized as the gold standard in exercise physiology to
measure the patient’s cardiorespiratory fitness [15]. This allowed
some studies to measure the extent of the patient’s physical
CONTACT Daniel Curnier daniel.curnier@umontreal.ca Laboratory of Pathophysiology of EXercise (LPEX), School of Kinesiology and Physical Activity Sciences,
Faculty of Medicine, Universit
e de Montr
eal, CEPSUM, 2100, boulevard

Edouard Montpetit, Montr
eal, QC H3C 3J7, Canada
ß2020 Informa UK Limited, trading as Taylor & Francis Group
DISABILITY AND REHABILITATION
https://doi.org/10.1080/09638288.2020.1725159

deconditioning [11,16,17]. In cancer patients, the 6-Minute
Walking Test (6MWT) is used to assess the functional capacity. In
pediatric oncology, the 6MWT is well tolerated and inexpensive,
compared to a direct measure in laboratory [15,18,19]. Though
the 6MWT does not directly measure the _
VO
2
peak in these
patients, several published equations (in healthy population) exist
to predict the _
VO
2
peak, using weight, height, sex and age
[20,21]. However, Schumacher [22] reported that the currently
published equations cannot accurately predict a valid _
VO
2
peak
value in leukemia survivors. This aspect was verified mainly in leu-
kemia survivors where the _
VO
2
peak equations may underesti-
mate (25 ml.kg
1
.min
1
) or overestimate (<25 ml.kg
1
.min
1
)
the value [22]. Moreover, it is observed that none of the pub-
lished equations use disease-specific variables. Thus, there is a
need to establish a new _
VO
2
peak equation in childhood ALL sur-
vivors taking into consideration clinical and specific variables
related to the disease to provide a better follow-up and a better
adapted physical training.
Since treatments decreased cardiorespiratory fitness in child-
hood ALL survivors, there is an important need to understand
specifically the change in their cardiorespiratory fitness. Thus, the
aim of this study was to establish and to validate a new equation
based on the 6MWT to predict _
VO
2
peak with clinical and specific
variables related to the disease in childhood ALL survivors.
Methods
Participants
All 200 participating childhood ALL survivors were diagnosed
between 1987 and 2010 and treated according to the Dana-
Farber Cancer Institute (DFCI) protocols 87-01 to 05-01 for ALL
patients [23] at Sainte-Justine University Health Center (SJUHC),
Montreal (Quebec), Canada. They were recruited in the context of
the PETALE study, a multidisciplinary research project with the
goal to identify and to comprehensively characterize associated
predictive biomarkers of long-term treatment related complica-
tions in childhood ALL survivors [24]. These 200 participants were
the first to be recruited in the context of the PETALE study which
included 250 participants. They had no history of refractory or
recurrent disease and did not receive a hematopoietic stem cell
transplant. These participants were almost exclusively of French
Canadian descent (>95%), with predominantly European ancestry
[24]. In the current study, we restricted participants to those that
were less than 19 years old at diagnosis. Subjects who have suf-
fered from congenital bone disease, symptomatic avascular necro-
sis or who received osteotoxic drugs for non-ALL disease were
excluded. Written informed consent was obtained from every
patient or parent/legal guardian. The study was conducted in
accordance with the Declaration of Helsinki and the protocol was
approved by the Ethics Review Committee of SJUHC.
Measurements
Functional capacity
The 6-Minute walk test (6MWT) was performed twice in a 30
meters hallway according to a standardized protocol [25] before
the cardiopulmonary exercise test. Participants were allowed to
take a 10-min break between the two tests. Heart rate was meas-
ured during the second 6MWT with a heart rate monitor (Polar
FT60, Polar Electro Oy, Kempele, Finland). Blood pressure and oxy-
gen saturation (SpO
2
) were measured in a sitting position at the
beginning and at the end of 6MWT. Rating of perceived exertion
(RPE) was evaluated at the end of the 6MWT using the Borg scale
(0–10) [26]. No encouragements were given to the participants to
increase the walking speed. Verbal instructions were given to indi-
cate the remaining time. In order to predict survivors’
_
VO
2
peak,
Ross [21] equation was used for its strong correlation (R
2
¼0.58)
with the _
VO
2
peak value [22].
Cardiopulmonary exercise test
The McMaster incremental cycle protocol [27] was performed on
an electromagnetic cycle ergometer (Ergoline, ER900, Bitz,
Germany) on the same day as the 6MWT. The test consisted of a
standard incremental procedure at a pedaling cadence of
50–70rpm starting at 25 W, increasing the load by 25 W or 50 W
(depending on the height and sex of the subject) every 2 min. At
the end of the test, an active rest period at 25 W for 3 min was
followed by a passive rest period of 3 min. All subjects were moni-
tored continuously during the test with 12-leads electrocardio-
grams (CASE Exercise Testing, GE Marquette, Milwaukee, WI, USA).
Blood pressure was measured every 2 min (Tango M2, SunTech
Medical, Morrisville, NC, USA). Both the cardiologist and the exer-
cise physiologist determined whether the participants reached
two out of three of the following criteria to validate their _
VO
2
peak: a respiratory exchange ratio value >1.15, rated perceived
exertion scale >7 (OMNI scale/10), and HR
max
85% of the pre-
dicted value [28].
Physical activity assessment
The daily physical activities were measured from the Minnesota
Leisure Time Physical Activity Questionnaire [29,30] and the
Tecumseh Self-Administered Occupational Physical Activity
Questionnaire [31]. An experienced exercise physiologist first read
the 20 sports included in the questionnaire and then guided the
survivors to recall any other sports they might have practiced in
the last three months. Precision on frequency, duration and inten-
sity of the activities were asked. A metabolic equivalent value
(MET) from the Compendiums of Physical Activity for Adults and
for Youth [32] was used to quantify the intensity of each activity.
All activities with a MET value 3 were considered of moderate-
to-vigorous intensity. Total of weekly minutes of moderate-to-
vigorous leisure physical activities (MVLPA) was then calculated.
Statistical analysis
Statistical analysis was performed using IBM SPSS statistics, ver-
sion 24.0 (IBM Corp., Armonk, NY, USA). We perform descriptive
statistics to describe participants. Continuous variables were
reported as mean ± standard deviation (SD). For clinical character-
istics, a two-tailed Student t-test was performed between male
and female survivors with a significance level of 5% was per-
formed. The cohort of survivors was randomly divided in two
groups of survivors to predict the equations (n¼118 (70%) with
61 males and 57 females) and to validate them (n¼50 (30%) with
27 males and 23 females) in order to respect the statistical power
analysis and the sex distributions to ensure a correct sex repre-
sentation. The generating random samples function of SPSS was
used to randomly divided our cohort. For each step described in
the Figure 1, two multiple linear regression analyses were used to
determine two new prediction equations for _
VO
2
peak.
In the first step, we performed a first linear regression analysis
to establish the new clinical _
VO
2
peak equation including the fol-
lowing clinical variables in the statistical model: age (years),
weight (kg), height (cm), heart rate at rest (bpm), heart rate at the
end of the 6MWT (HRend) (bpm), rating of perceived exertion
2 J. LABONTÉ ET AL.

(RPE), moderate to vigorous leisure physical activities (MVLPA)
(min/day), and 6-min walk distance (6MWD) (m).
In the second step, we performed another linear regression
analysis where the following disease-specific variables were added
to the new clinical _
VO
2
peak equation previously determined to
establish the new disease-specific _
VO
2
peak equation: age at can-
cer diagnosis (years), duration of treatments (DT) (years), time
from the end of the treatments (years), time from the diagnosis
(years), vincristine cumulative dose (mg/m
2
), radiotherapy treat-
ments (0 ¼no and 1 ¼yes) and the prognostic risk group (i.e.,
standard risk (SR), high risk (HR)) [23]. The cumulative doxorubicin
dose for the SR group was 60 mg/m
2
and 360 mg/m
2
(DFCI-ALL
protocols 85-01, 87-01 and 91-01) or 300 mg/m
2
(DFCI-ALL proto-
cols 95-01) for the HR group.
In both statistical models, a backward elimination approach
was used to complete the regression models. For each multiple
linear regression performed, the coefficient of determination (R
2
)
was reported to assess the adjustment of the model. Spearman
correlation coefficient tests was realized to identify a possible rela-
tionship between the _
VO
2
peak predicted by the new equations
and measured by the cardiopulmonary exercise test. We used the
Bland and Altman method to assess the accuracy between _
VO
2
peak measured and the new predicted equations for _
VO
2
peak,
based on graphical techniques and simple calculations [33]. The
use of the Bland and Altman method allows better repeatability
and equation validation [33]. The first quantitative measure used
in the Bland Altman analyses was the difference between _
VO
2
peak measured and the new predicted equation and the second
quantitative measure was the mean between _
VO
2
peak measured
and the new predicted equation. Limits of Agreement (LOA) was
calculated for the Bland and Altman plots where LOA were
defined as the mean difference with a 95% LOA calculated (upper
LOA ¼(mean þ(1.96SD)) and lower LOA ¼(mean (1.96SD))).
The differences were considered significant when p-value is
p<0.05 with an alpha risk of 5%.
Results
Our study included 200 survivors, among which 32 survivors were
excluded: 28 participants did not complete the functional capacity
test (n¼18) or the cardiopulmonary exercise test (n¼10) because
there were issues with the equipment or data collection and 4
participants received a different treatment than the DFCI-ALL pro-
tocols. A total of 168 survivors were included in our analyses and
their characteristics are presented in Table 1.
New validated clinical
_
VO
2
peak equation
To predict the new clinical VO
2
peak equation, the standardized
and non-standardized coefficients of each clinical variables from
the linear regression after the backward approach are presented
in Table 2. We observed a significant strong correlation between
the _
VO
2
peak value predicted from the new validated clinical _
VO
2
peak equation and the _
VO
2
peak measured by the cardiopulmon-
ary exercise test (R
2
¼0.56, Spearman correlation ¼0.695;
p<0.0001). The Bland and Altman validation method of this
model is presented in Figure 2(A) and shows that more than 95%
of values are between the LOA (9.64–14.98), while the mean
bias is 2.67 ml.kg
1
.min
1
. The new establish clinical _
VO
2
peak
equation from the 6MWT in childhood ALL survivors is: _
VO
2
peak
(mL.kg
1
.min
1
)¼(0.283age(years)) (0.099weight(kg)) þ
(0.0716MWD(meters)) (0.135HR end(bpm)) þ22.789.
New validated disease-specific
_
VO
2
peak equation
To predict the new disease-specific _
VO
2
peak equation, the stand-
ardized and non-standardized coefficients of each disease-specific
variables from the linear regression after the backward approach
are presented in Table 3. We observed a significant strong correl-
ation between the _
VO
2
peak value predicted from the new
validated disease-specific _
VO
2
peak equation and the _
VO
2
peak equation measured by the cardiopulmonary exercise test
Figure 1. Statistic plan of steps used for the prediction equations of VO
2
peak.
_
VO
2
peak, maximal oxygen consumption (mL.kg
1
.min
1
); HR rest (bpm), heart rate
at rest (bpm); RPE, rating of perceived exertion; MVLPA, moderate to vigorous leisure of physical activity (min/day); HR end (bpm), heart rate at the end of test
(bpm); 6MWD, distance at 6-Minute Walking Test (meters); DT, duration of treatments (years).
PREDICT V
˙O
2
PEAK FROM THE 6MWT IN ALL SURVIVORS 3

(R
2
¼0.61, Spearman correlation ¼0.695; p<0.0001). The Bland
and Altman validation method of this model is presented in
Figure 2(B) and shows that more than 95% of values are between
the LOA (9.98 to 15.01), while the mean bias is
2.51 ml.kg
1
.min
1
. The new establish disease-specific _
VO
2
peak
equation from the 6MWT in childhood ALL survivors is: _
VO
2
peak
(mL.kg
1
.min
1
)¼(0.236age(years)) (0.094weight(kg)) 
(0.120HR end(bpm)) þ(0.0676MWD(meters)) þ
(0.065MVLPA(min/day)) (0.204DT(years)) þ25.145.
Based on the Ross [21] data, we observed from the Bland and
Altman validation method a mean bias of 12.09 ml.kg
1
.min
1
when the _
VO
2
peak value predicted from the new validated dis-
ease-specific _
VO
2
peak is compared to the _
VO
2
peak value pre-
dicted from Ross [21] equation (Figure 2(C)).
Discussion
This is the first study that predicted _
VO
2
peak from a 6MWT using
clinical and disease-specific variables of childhood ALL survivors.
In this sense, the main result of this study is that our new vali-
dated _
VO
2
peak equations can predict more accurately _
VO
2
peak
from the 6MWT than the previously published equations in child-
hood ALL survivors. The need to evaluate survivors’cardiorespira-
tory fitness with as little expenses as possible is very important in
the care process to propose a better follow-up for the patient.
Our study provides two new validated equations to predict _
VO
2
peak from the 6MWT. The first validated equation used clinical
variables that are easy to use for every health care professional
and exercise physiologist, whereas the second validated equation
used disease-specific variables that offer better precision for exer-
cise and oncology researchers. The accessibility of these new vali-
dated equations is a major advance in the follow-up of survivors’
cardiorespiratory fitness.
The follow-up of the childhood ALL survivors, especially in
regard to their cardiorespiratory fitness is essential in an exercise
oncology context. Physical deconditioning being an important
issue in childhood ALL survivors [14], it is essential for the exercise
physiology researcher to develop new _
VO
2
peak equations to bet-
ter understand patients’cardiopulmonary status and thus better
prescribe exercise. Several studies have investigated the link
between 6MWT and _
VO
2
peak in order to try to find a specific
equation to predict _
VO
2
peak from 6MWT [22]. However, as
reported by Schumacher et al., 2018, no studies have validated
their equation and none of these equations accurately predicted
the _
VO
2
peak value. This aspect was verified mainly in leukemia
survivors where the _
VO
2
peak equations may underestimate
(25 ml.kg
1
.min
1
) or overestimate (<25 ml.kg
1
.min
1
) the
value [22]. Compared to our new validated equations, most of the
previously published equations have underestimated _
VO
2
peak
from 6MWT. It should be noted that a majority of equations in
previous models have used BMI, weight and height as principal
variables which have been recognized as the most important vari-
ables influencing 6MWT results [21,22,34]. However, since the BMI
is reported as inaccurate in the general population [35], it seems
essential to move forward and include, in the statistical model,
disease-specific variables related to pediatric cancer survivors. It is
in this aspect that our new equations outperform from others. For
example, our new validated clinical _
VO
2
peak equation did not
only include weight, age and 6MWD variables in order to predict
_
VO
2
peak, but also included the heart rate at the end of 6MWT.
The addition of disease-specific variables allowed to increase the
accuracy of our validated equation to predict _
VO
2
peak. The
choice of variables to be included in the model seems to be one
of the most important aspects of the validation process of an
equation and needs to be discussed.
According to Schumacher [22], the most important variables
included in the previously published equations are weight, height,
Table 1. Clinical characteristics of childhood acute lymphoblastic leukemia survivors.
Survivors Females survivors Males survivors pValue
Total (n) 168 80 88
Age at visit (y) 22.2 ± 6.3 22.2 ± 6.4 22.2 ± 6.2 0.99
Weight (kg) 65.4 ± 15.4 62.7 ± 15.3 68.2 ± 15.0 0.02
Height (cm) 1.7 ± 0.1 1.6 ± 0.1 1.7 ± 0.1 <0.001
Age at cancer diagnosis (y) 6.2 ± 4.5 6.3 ± 4.4 6.1 ± 4.6 0.81
Time from diagnosis (y) 16.1 ± 5.2 16.1 ± 5.5 16.0 ± 5.0 0.95
Time from the end of the treatment (y) 13.9 ± 5.3 13.9 ± 5.5 13.9 ± 5.0 0.98
Duration of treatments (y) 2.2 ± 0.3 2.2 ± 0.4 2.1 ± 0.3
DOX (mg/m
2
) 183.8 ± 122.1 188.0 ± 126.3 180.0 ± 118.8 0.68
DEX (mg/m
2
) 2834.9± 417.0 2902.9 ± 368.0 2766.8 ± 459.6 0.30
Survivors who received radiotherapy treatments 98 38 60 0.03
Cardiopulmonary exercise test
_
VO
2
peak during exercise (mL.kg
1
.min
1
) 32.1 ± 8.1 27.2 ± 6.5 36.7 ± 6.7 <0.001
Heart rate at
_
VO
2
peak (bpm) 188.2 ± 10.2 187.3± 10.7 188.9 ± 9.8 0.33
RPE
max
9.0 ± 1.3 9.1 ± 1.2 8.8 ± 1.3 0.17
Functional capacity test (6MWT)
Total distance (m) 611.2 ± 77.3 584.5± 68.6 635.5 ± 77.2 <0.001
Maximal heart rate (bpm) 150.0 ± 21.5 153.8± 20.6 146.6 ± 21.8 0.03
RPE
max
3.6 ± 1.9 3.9 ± 1.8 3.3 ± 1.9 0.06
ALL: acute lymphoblastic leukemia; DOX: cumulative dose of doxorubicin; DEX: cumulative dose of dexrazoxane;
_
VO
2
peak: maximum oxygen
consumption; RPE: Rating of perceived exertion.
Data are expressed as mean ± SD.
Significant difference between female and male survivors.
Table 2. Standardized and non-standardized coefficients from multiple linear
regression analysis to predict
_
VO
2
peak from clinical variables.
Survivors (R
2
¼0.56)
Non-standardized
coefficients
Standardized
coefficients (95% CI) pValue
Constant 22.789
Age (y) 0.283 0.219 (0.46; 0.11) 0.002
Weight (kg) 0.099 0.194 (0.17; 0.03) 0.005
Walk distance (m) 0.071 0.695 (0.06; 0.08) <0.0001
Heart rate at the end
of the 6MWT (bpm)
0.135 0.333 (0.19; 0.08) <0.0001
_
VO
2
peak: maximal oxygen consumption.
4 J. LABONTÉ ET AL.

age and 6MWD. One of the main challenges to discuss our
findings with previous studies [21,22,34,36] is that previously pub-
lished equations did not include standardized and non-standar-
dized coefficients in their results, making the validation of their
equations difficult. Thus, the presentation of these parameters in
our results is the strength of our study. Moreover, it appeared
during the statistical analysis process that the reproduction of
previous studies’statistical analyses is not possible, especially
Figure 2. Bland–Altman plots of the difference between
_
VO
2
peak predicted and
_
VO
2
peak measured compared to the means. (a) Bland–Altman plots of the differ-
ence between
_
VO
2
peak predicted from standard equation and
_
VO
2
peak measured compared to the means; (b) Bland–Altman plots of the difference between
_
VO
2
peak predicted from specific equation and
_
VO
2
peak measured compared to the means; (c) Bland–Altman plots of the difference between
_
VO
2
peak predicted from
Ross equation and
_
VO
2
peak measured compared to the means.
PREDICT V
˙O
2
PEAK FROM THE 6MWT IN ALL SURVIVORS 5

when we want to apply this process in a disease-specific popula-
tion, such as childhood ALL survivors. In order to explain the dif-
ferences between our findings and those of previous studies, the
use of the correlation coefficient to validate and determine the
strength of the equations has been used in several studies.
However, it has been shown that the use of correlations to com-
pare a new measurement, like a new predicted equation, with an
already established one can be misleading [33]. It is recom-
mended to perform an alternative approach based on graphical
techniques and simple calculations. In this sense, the use of the
Bland and Altman method allows better repeatability and equa-
tions’validation [33]. By this method, our results showed that
without disease-specific variables (i.e., clinical _
VO
2
peak equation)
in the validated _
VO
2
peak equation, the mean bias was of
2.67 ml.kg
1
.min
1
, whereas with disease-specific variables (i.e.,
disease-specific _
VO
2
peak equation), the mean bias was of
2.51 ml.kg
1
.min
1
. This approach again shows the importance of
adding the disease-specific variables that can reduce the mean
bias by 6% (0.16 ml.kg
1
.min
1
). The absence of a significant dif-
ference between _
VO
2
peak values from the clinical _
VO
2
peak
equation and the disease-specific _
VO
2
peak equation allows
health care professionals and exercise physiologists to use the
equation of their choice. This is reinforced by the fact that both
_
VO
2
peak equations are below the minimum clinical value (i.e.,
_
VO
2
peak) generally used in cardiac rehabilitation by the
American College of Sports Medicine [37]. Also, we showed that
both of our new validated equations are more accurate than the
Ross equation [38] since we reported a difference of mean bias of
14.6 ml.kg
1
.min
1
. The use of data dispersion from the Bland
and Altman method to validate both our new equations was the
strength of our study, since it has been shown that statistical
methods to assess agreement between two methods of clinical
measurement are more adequate in maintaining data validity [33].
Finally, in a clinical context, it was a priority in this study to give
the same access to _
VO
2
peak equations to every health care pro-
fessional and exercise physiologist, as well as exercise and oncol-
ogy researchers. The new validated clinical _
VO
2
peak equation
offers more access to the patient’s cardiorespiratory fitness follow-
up for health care professionals and exercise physiologists.
Clinical implications
This study highlights major clinical implications for every health
care professional, as well as exercise physiologist. As discussed in
this study, the need to evaluate survivors’cardiorespiratory fitness
is very important in the care process, especially in regard to their
cardiorespiratory follow-up. Both new validated equations, clinical
and disease-specific _
VO
2
peak equations, offer better access to
patient’s health since the equations use clinical and specific varia-
bles related to the disease. Also, the new validated equations take
into consideration the age of the patient which makes them
accurate, valid and representative equations of the predicted _
VO
2
peak value in childhood ALL survivors’population (age range of
the cohort was 11 to 40 years old). This is an important aspect of
this research because not all hospitals or rehabilitation centers
have the financial resources to administer a maximal cardiopul-
monary exercise test in laboratory. Thus, 6MWT remains more
advantageous for this population since it is a sub-maximal exer-
cise test that is well tolerated, safe and less expensive for those
who administer the test. With these new equations for childhood
ALL survivors, we can estimate accurately _
VO
2
peak. This will
allow a better follow-up and better adapted physical training
among survivors. It will also significantly help exercise physiolo-
gists in monitoring the progression of their patients. Moreover,
the 6MWT may be performed more frequently than a maximal
cardiopulmonary exercise test with much lower costs. The use of
our new validated disease-specific _
VO
2
peak equation could allow
future studies to suggest clinical thresholds according to a model
that is similar to the one used to determine clinical thresholds in
chronic heart failure [38,39] and chronic obstructive pulmonary
disease patients [40,41].
Limitations
Our study identifies some limitations. The first concerns our
method because both _
VO
2
peak equations were predicted from a
maximal exercise test on ergocycle. It would have been possible
to evaluate the cardiorespiratory fitness with aerobic field tests
[42] but reaching a maximum level of exercise effort in survivors
has been reported to be limited by symptoms [43]. Also, perform-
ing a maximal exercise test on a treadmill in our cancer survivor
population was not possible. Indeed, cancer treatments induce, in
a clinical context, imbalance, bone fragility and modification of
the walking pattern which lead to significant risks, including the
risk of falling [44–46]. Moreover, in a research context, the use of
a treadmill test does not contribute to achieving maximum effort
(i.e., limited by symptoms) [43] and increases the number of
physiological artifacts on the electrocardiogram [47] which makes
the interpretation of the data and the follow-up of the patient
during the test difficult. In order to ensure that survivors had
reached their _
VO
2
peak, confirmed by clinical recommendations
[28], they performed cardiopulmonary exercise tests on ergocycle.
In this sense, the use of valid physiological measurements was
the strength of our study, especially regarding their implication in
predicting the _
VO
2
peak value, which is considered as the gold
standard to assess cardiorespiratory fitness in exercise physiology
[48]. Another limitation of this study was that the survivors’prog-
nostic risk group was included in the model, before the backward
elimination, to establish the new disease-specific equation. This
could have limited the use of the new validated disease-specific
_
VO
2
peak equation to the children who had been treated accord-
ing to DFCI-ALL protocols. However, this potential limitation can-
not be considered a bias since this disease-specific variable was
excluded from the backward elimination approach. The last limita-
tion of this study was that the physical activity levels were meas-
ured by questionnaires, which may overestimate or underestimate
survivors’physical activity levels, especially in low-intensity activ-
ities [49]. In this sense, moderate to vigorous physical activity
data were our variables of interest and must be considered in the
prediction of _
VO
2
peak, as demonstrated in our study. Also, it
Table 3. Standardized and non-standardized coefficients from multiple linear
regression analysis to predict
_
VO
2
peak from disease-specific variables.
Survivors (R
2
¼0.61)
Non-
standardized
coefficients
Standardized
coefficients (95% CI) pValue
Constant 25.145
Age (y) 0.236 0.183 (0,41; 0.07) 0.007
Weight (kg) 0.094 0.184 (0.16; 0.03) 0.006
Heart rate at the end
of the 6MWT (bpm)
0.120 0.294 (0.17; 0.07) <0.0001
Walk distance (m) 0.067 0.657 (0.05; 0.08) <0.0001
MVLPA 0.065 0.204 (0.03; 0.11) 0.002
Duration of Treatments (y) 0.204 0.118 (0.410; 0) <0.05
_
VO
2
peak: maximal oxygen consumption; MVLPA: moderate to vigorous leisure
of physical activity.
6 J. LABONTÉ ET AL.

should be noted that the use of an accelerometer could have
been an interesting assessment tool to measure survivors’physical
activity levels. However, the complexity of its administration and
the time required to obtain the data is an important limitation in
a clinical oncology context.
Conclusion
This study established two new validated _
VO
2
peak equations
from the 6MWT using clinical and disease-specific variables in
childhood ALL survivors. The availability of these newly estab-
lished _
VO
2
peak equations makes them an accurate tool to pro-
vide a better follow-up and better adapted physical training for
survivors. We invite researchers to use our assessment procedures
for their further studies. This study addresses an important need
in oncology and will help researchers and clinicians to increase
the knowledge regarding the survivor’s physiological parameters.
Future replication studies are needed to improve our knowledge
of the physiological parameters of other cancers. The accessibility
of these new validated equations is a major advance in the fol-
low-up of the survivors’cardiorespiratory fitness, especially for
every health care professionals and exercise physiologists.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This work was supported by the Institute of Cancer Research (ICR)
of the Canadian Institutes of Health Research (CIHR), in collabor-
ation with C17 Council, Canadian Cancer Society (CCS), Cancer
Research Society (CRS), Garron Family Cancer Centre at the
Hospital for Sick Children, Ontario Institute for Cancer Research
(OICR) and Pediatric Oncology Group of Ontario (POGO). This
research was also supported in part by PhD study grants from
Cole Foundation, Fonds de Recherche du Qu
ebec –Sant
e (FRQS),
Sainte-Justine University Hospital Center Foundation and
Foundation of Stars. The funders had no role in the study design,
data collection and analysis, decision to publish, or preparation of
the manuscript. We appreciate the assistance of Ariane Levesque
(McGill University) for her review of the article in the
English language.
ORCID
Maxime Caru http://orcid.org/0000-0003-2904-9185
Daniel Curnier http://orcid.org/0000-0001-9717-192X
References
[1] Ellison L, Janz T. Incidence du cancer et mortalit
e par can-
cer chez les enfants au Canada. Statistique Canada, Coup
D’oeil Sur la Sant
e; 2015. p. 82–624.
[2] Hunger SP, Lu X, Devidas M, et al. Improved survival for
children and adolescents with acute lymphoblastic leuke-
mia between 1990 and 2005: a report from the children’s
oncology group. J Clin Oncol. 2012;30(14):1663–1669.
[3] Hudson MM, Ness KK, Gurney JG, et al. Clinical ascertain-
ment of health outcomes among adults treated for child-
hood cancer. JAMA. 2013;309(22):2371–2381.
[4] Oeffinger KC, Mertens AC, Sklar CA, et al. Chronic health
conditions in adult survivors of childhood cancer. N Engl J
Med. 2006;355(15):1572–1582.
[5] Lemay V, Caru M, Samoilenko M, et al. Prevention of long-
term adverse health outcomes with cardiorespiratory fit-
ness and physical activity in childhood acute lymphoblastic
leukemia survivors. J Pediatr Hematol Oncol. 2019;41(7):
e450–e458.
[6] Fresneau B, Fayech C, Butel T, et al. Treatment-related car-
diotoxicity in childhood cancer survivors: risk factors and
follow-up. Rev Med Interne. 2017;38(2):125–132.
[7] Lipshultz SE, Adams MJ, Colan SD, et al. Long-term cardio-
vascular toxicity in children, adolescents, and young adults
who receive cancer therapy: pathophysiology, course, mon-
itoring, management, prevention, and research directions:
a scientific statement from the American Heart Association.
Circulation. 2013;128(17):1927–1995.
[8] Raj S, Franco VI, Lipshultz SE. Anthracycline-induced cardio-
toxicity: a review of pathophysiology, diagnosis, and treat-
ment. Curr Treat Options Cardiovasc Med. 2014;16(6):315
[9] Caru M, Corbin D, Perie D, et al. Doxorubicin treatments
induce significant changes on the cardiac autonomic ner-
vous system in childhood acute lymphoblastic leukemia
long-term survivors. Clin Res Cardiol. 2019;108(9):
1000–1008.
[10] Lipshultz SE, Lipsitz SR, Sallan SE, et al. Chronic progressive
cardiac dysfunction years after doxorubicin therapy for
childhood acute lymphoblastic leukemia. J Clin Oncol.
2005;23(12):2629–2636.
[11] Hung SH, Rankin A, Virji-Babul N, et al. Associating physical
activity levels with motor performance and physical func-
tion in childhood survivors of acute lymphoblastic leuke-
mia. Physiother Can. 2017;69(1):57–64.
[12] van Brussel M, Takken T, Lucia A, et al. Is physical fitness
decreased in survivors of childhood leukemia? A systematic
review. Leukemia. 2005;19(1):13–17.
[13] Tonorezos ES, Snell PG, Moskowitz CS, et al. Reduced car-
diorespiratory fitness in adult survivors of childhood acute
lymphoblastic leukemia. Pediatr Blood Cancer. 2013;60(8):
1358–1364.
[14] Caru M, Samoilenko M, Drouin S, et al. Childhood acute
lymphoblastic leukemia survivors have a substantially lower
cardiorespiratory fitness level than healthy Canadians des-
pite a clinically equivalent level of physical activity. J
Adolesc Young Adult Oncol. 2019;8(6):674–683.
[15] Powell AW, Jefferies JL, Hopkin RJ, et al. Cardiopulmonary
fitness assessment on maximal and submaximal exercise
testing in patients with Fabry disease. Am J Med Genet A.
2018;176(9):1852–1857.
[16] Ness KK, Hudson MM, Pui CH, et al. Neuromuscular impair-
ments in adult survivors of childhood acute lymphoblastic
leukemia: associations with physical performance and
chemotherapy doses. Cancer. 2012;118(3):828–838.
[17] Hooke MC, Garwick AW, Neglia JP. Assessment of physical
performance using the 6-minute walk test in children
receiving treatment for cancer. Cancer Nurs. 2013;36(5):
E9–E16.
[18] Bartels B, de Groot JF, Terwee CB. The six-minute walk test
in chronic pediatric conditions: a systematic review of
measurement properties. Phys Ther. 2013;93(4):529–541.
[19] Rostagno C, Gensini GF. Six minute walk test: a simple and
useful test to evaluate functional capacity in patients with
heart failure. Intern Emerg Med. 2008;3(3):205–212.
PREDICT V
˙O
2
PEAK FROM THE 6MWT IN ALL SURVIVORS 7

[20] Jalili M, Nazem F, Sazvar A, et al. Prediction of maximal
oxygen uptake by six-minute walk test and body mass
index in healthy boys. J Pediatr. 2018;200:155–159.
[21] Ross RM, Murthy JN, Wollak ID, et al. The six minute walk
test accurately estimates mean peak oxygen uptake. BMC
Pulm Med. 2010;10(1):31.
[22] Schumacher AN, Shackelford D, Brown JM, et al. Validation
of the six-minute walk test for predicting peak VO2 in can-
cer survivors. Med Sci Sports Exerc. 2018;51(2):271–277.
[23] Silverman LB, Stevenson KE, O’Brien JE, et al. Long-term
results of Dana-Farber Cancer Institute ALL Consortium
protocols for children with newly diagnosed acute lympho-
blastic leukemia (1985-2000). Leukemia. 2010;24(2):
320–334.
[24] Marcoux S, Drouin S, Laverdiere C, et al. The PETALE study:
late adverse effects and biomarkers in childhood acute
lymphoblastic leukemia survivors. Pediatr Blood Cancer.
2017;64(6):1–8.
[25] Butland RJ, Pang J, Gross ER, et al. Two-, six-, and 12-
minute walking tests in respiratory disease. Br Med J (Clin
Res Ed). 1982;284(6329):1607–1608.
[26] Thiel TA. The validity and reliability of the Borg rating of
perceived exertion scale during aerobic exercise. Boston
(MA): Northeastern University; 1986.
[27] Bar-Or O, Rowland TW. Pediatric exercise medicine: from
physiologic principles to health care application.
Champaign (IL): Human Kinetics; 2004.
[28] Guazzi M, Adams V, Conraads V, et al. EACPR/AHA scientific
statement. Clinical recommendations for cardiopulmonary
exercise testing data assessment in specific patient popula-
tions. Circulation. 2012;126(18):2261–2274.
[29] Taylor HL, Jacobs DR Jr., Schucker B, et al. A questionnaire
for the assessment of leisure time physical activities. J
Chronic Dis. 1978;31(12):741–755.
[30] Kriska AM, Caspersen CJ. Introduction to a collection of
physical activity questionnaires. Med Sci Sports Exerc. 1997;
29(6):5–9.
[31] Montoye HJ. Estimation of habitual physical activity by
questionnaire and interview. Am J Clin Nutr. 1971;24(9):
1113–1118.
[32] Ridley K, Ainsworth BE, Olds TS. Development of a com-
pendium of energy expenditures for youth. Int J Behav
Nutr Phys Act. 2008;5(1):45–48.
[33] Bland JM, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
Lancet. 1986;1(8476):307–310.
[34] Enright PL, Sherrill DL. Reference equations for the six-
minute walk in healthy adults. Am J Respir Crit Care Med.
1998;158(5):1384–1387.
[35] Tomiyama AJ, Hunger JM, Nguyen-Cuu J, et al.
Misclassification of cardiometabolic health when using
body mass index categories in NHANES 2005–2012. Int J
Obes. 2016;40(5):883–886.
[36] Cahalin LP, Mathier MA, Semigran MJ, et al. The six-minute
walk test predicts peak oxygen uptake and survival in
patients with advanced heart failure. Chest. 1996;110(2):
325–332.
[37] American College of Sports Medicine. ACSM’s guidelines
for exercise testing and prescription. Baltimore (MD):
Lippincott Williams & Wilkins; 2013.
[38] Guazzi M, Dickstein K, Vicenzi M, et al. Six-minute walk test
and cardiopulmonary exercise testing in patients with
chronic heart failure: a comparative analysis on clinical and
prognostic insights. Circ Heart Fail. 2009;2(6):549–555.
[39] Prescher S, Schoebel C, Koehler K, et al. Prognostic value of
serial six-minute walk tests using tele-accelerometry in
patients with chronic heart failure: a pre-specified sub-
study of the TIM-HF-Trial. Eur J Prev Cardiolog. 2016;
23(2_suppl):21–26.
[40] Celli BR, Cote CG, Marin JM, et al. The body-mass index, air-
flow obstruction, dyspnea, and exercise capacity index in
chronic obstructive pulmonary disease. The. N Engl J Med.
2004;350(10):1005–1012.
[41] Wise RA, Brown CD. Minimal clinically important differences
in the six-minute walk test and the incremental shuttle
walking test. Copd. 2005;2(1):125–129.
[42] Hartman A, Hop W, Takken T, et al. Motor performance
and functional exercise capacity in survivors of pediatric
acute lymphoblastic leukemia. Pediatr Blood Cancer. 2013;
60(3):494–499.
[43] Kim S, Song IC, Jee S. Cardiopulmonary exercise test in leu-
kemia patients after chemotherapy: a feasibility study. Ann
Rehabil Med. 2017;41(3):456–464.
[44] Fulbright JM, Raman S, McClellan WS, et al. Late effects of
childhood leukemia therapy. Curr Hematol Malig Rep.
2011;6(3):195–205.
[45] Nathan PC, Wasilewski-Masker K, Janzen LA. Long-term out-
comes in survivors of childhood acute lymphoblastic leuke-
mia. Hematol Oncol Clin North Am. 2009;23(5):1065–1082,
vi-vii.
[46] Campbell G, Wolfe RA, Klem ML. Risk factors for falls in
adult cancer survivors: an integrative review. Rehabil Nurs.
2018;43(4):201–213.
[47] Stone DB, Tamburro G, Fiedler P, et al. Automatic removal
of physiological artifacts in EEG: the optimized fingerprint
method for sports science applications. Front Hum
Neurosci. 2018;12:96
[48] Shephard RJ, Allen C, Benade A, et al. The maximum oxy-
gen intake: an international reference standard of cardio-
respiratory fitness. Bull World Health Organ. 1968;38(5):757.
[49] Skender S, Ose J, Chang-Claude J, et al. Accelerometry and
physical activity questionnaires –a systematic review. BMC
Public Health. 2016;16(1):515.
8 J. LABONTÉ ET AL.