Content uploaded by Gabriella Pravettoni
Author content
All content in this area was uploaded by Gabriella Pravettoni on Mar 22, 2017
Content may be subject to copyright.
© 2015 INTM, Italy. Published by Wichg Publishing
TJ
ISSN 0300-8916
Tumori 2016; 102(2): 217-221
CLINICAL TRIAL PROTOCOL
cells, but treatment can help achieve long-term remission
(3). In some paents, however, treatment appears to eec-
vely induce complete disease remission (4). In such cases,
disconnuaon of therapy may be an opon under certain
circumstances.
Disconnuaon of therapy can be proposed to CML
paents who reach a sustained deep molecular response
(DMR) for a minimum and stable period (normally 2 years)
(5). However, the choice may represent a considerable
challenge to paents, and not only clinical but also psycho-
logical and emoonal factors may inuence the medical
decision-making process. Emoonal and cognive aspects
can impact on medical choices, thereby interfering with
treatment, coping, risk evaluaon and paent-doctor rela-
onships, as shown in previous work by our group (6-10).
In this process, paents are asked to comtemplate a major
change in their usual treatment and this change can sig-
nicantly impact everyday-life acvies and health-related
quality of life (HRQL). For this reason, current scienc dis-
cussions on therapy and clinical trials are evaluang when
and how to propose the disconnuaon of TKI therapy
(11-13).
Currently only a small percentage of paents switch
to the new opon, while the majority prefer to connue
with the usual treatment even when blood tests reveal a
DOI: 10.5301/tj.5000451
Raonale and protocol of CML study: cognive and
emoonal impact of disconnuaon of therapy in
paents with chronic myeloid leukemia
Silvia Riva1,2, Ke Mazzocco1,2, Gabriella Praveoni1,2
1 Department of Oncology and Hemato-oncology, University of Milan, Milan - Italy
2 Applied Research Division for Cognive and Psychological Science, European Instute of Oncology (IEO), Milan - Italy
Introducon
Over the past decade, dierent types of drugs designed
to specically inhibit protein tyrosine kinases (tyrosine kinase
inhibitors or TKIs) have become a new treatment for cancer
paents, and the example of chronic myeloid leukemia (CML)
is one of the most resounding (1). In this disease, TKIs have
dramacally improved the long-term survival rate (95.2%) (2)
and they are now rmly established as an eecve therapy
for paents with CML (3).
Today, the most favorable outcome in CML treatment is
to eliminate the blood cells that contain the abnormal BCR-
ABL gene causing the overabundance of diseased blood cells.
In most paents it is not possible to eliminate all diseased
AbsTRACT
Introducon: Disconnuaon of therapy can be proposed to paents with chronic myelogenous leukemia (CML)
who reach a sustained deep molecular response (DMR) for a minimum and stable period. Today, a considerable
number of paents reach a sustained DMR, especially when they are treated with the latest drugs. Although new-
generaon treatments may provide signicant improvement in terms of paent health and health-related quality
of life, many paents are uncertain about disconnuaon and may refuse a treatment switch.
Methods: This study is an observaonal research project aimed at invesgang, from a psychological point of
view, possible cognive and emoonal components that can inuence treatment disconnuaon and treatment
decisions in a cohort sample of 120 CML paents.
Results: The expected results indicate that cognive and emoonal factors may inuence decision-making in this
seng and may prevent appropriate risk-and-benet evaluaon of new treatment approaches.
Conclusions: This is the rst study that will analyze in depth all possible psychological variables that can inter-
fere with the medical decision process of treatment disconnuaon in CML, providing new insights for clinical
pracce.
Keywords: Chronic myeloid leukemia, Decision-making, Disconnuaon, Medical choices, Personality, Risk
percepon
Accepted: October 20, 2015
Published online: December 11, 2015
Corresponding author:
Silvia Riva
Department of Health Sciences
University of Milan
Via A. Di Rudinì 8
20142 Milan, Italy
silvia.riva1@unimi.it
Raonale and protocol of CML study
218
© 2015 INTM, Italy. Published by Wichg Publishing
complete and stable remission of CML (14-16). However,
connuing therapy may no longer be appropriate in terms
of cost-eecveness, both from a welfare and paent per-
specve (12, 13). In fact, the prescripon of TKIs may weigh
on public health expenses. Moreover, the standard therapy
undoubtedly has an impact on paents’ HRQL in terms of
compliance with treatment and of the nontrivial side eects
of the drugs (17, 18). Nevertheless, the most eecve way
to support paents across this change has only marginally
been addressed (15, 17).
It is therefore crical to invesgate the risk atude
that could inuence the choices and decisional processing
of paents eligible to switch and/or disconnue therapy.
In this context, evaluang the psychological prole of CML
paents including personality traits, emoonal and cogni-
ve funcons can shed light on the factors that impact on
their risk atude and its inuence on treatment choice.
More specically, the atude toward risk behaviors, risk
percepon, and personality traits may reveal crical pieces
of informaon that subtend the willingness or rejecon to
switch to second-generaon TKIs and/or disconnuaon of
treatment.
Such informaon will be very useful to all clinicians who
treat paents with CML because it will make it clear which
type of paents can be proposed a switch to second-gener-
aon drugs and/or disconnuaon of therapy and how they
can be supported in this change.
Study objecves
The primary objecve of this study is to dene the risk
prole of CML paents at dierent stages of the disease using
the following measures:
• patients’ risk attitude
• patients’ risk preferences
• patients’ need for cognitive closure
• patients’ personality traits
The secondary objectives of this study are to assess which
determinants affect the risk profile, including
• patient history
• clinical parameters (e.g., diagnosis, treatment, status of
the disease)
• sociodemographic data (e.g., age, gender)
Study design
The study follows the STROBE (STrengthening the
Reporng of OBservaonal studies in Epidemiology) guide-
lines for observaonal studies and will adhere to the
STROBE checklist for cohort, case-control, and cross-seconal
studies (19).
This is a prospecve, non-intervenonal study (NIS) that
will describe the risk prole in CML paents with dierent
types of treatment and responses. The study is focused
on adults (>18 years of age) living in Italy. The NIS will be
coordinated in Milan at the Psycho-Oncology Unit of the
European Instute of Oncology (IEO), Milan, Italy. However,
also other Italian centers included in the Lombardy Hemato-
logical Network (Rete Ematologica Lombarda - hp://www.
rel-lombardia.net/) will be asked to parcipate in data col-
lecon under the supervision of the center in Milan. It is
expected that data from approximately 120 paents will be
collected.
Duraon
The study period will last 12 months. The rst 2 months
will be used for the regulatory process with approval of the
study protocol in all parcipang centers.
Premature disconnuaon of data capture
Paents will be followed as long as data from roune
assessments are available and the permission to document
the data has not been withdrawn. If a paent cannot be
followed, the reasons for this will be entered into the docu-
mentaon form.
Design and me frames
In this prospecve study, risk proles will be assessed
over a 12-month period in a cohort of adult CML paents
who are receiving roune medical care. In order to inves-
gate the individual cognive and psychological proles with
parcular aenon to the personal risk atude of paents,
each paent will be asked to parcipate in a 1-hour tesng
session that will take place in a hospital room assigned for
this purpose. The session will be conducted by sta who
have been specically trained by one of the authors of this
paper (GP), a trained cognive psychologist who is an ex-
pert in cognive and psychological assessment.
Paents
CML paents with dierent types of treatment and respons-
es as dened in Table I according to the European LeukemiaNet
guidelines of 2013 (5) will be included in this NIS.
Inclusion criteria
• CML patients with one of the characteristics described in
Table I
• Age >18 years
• Ability to read and understand the study materials (pa-
tient information and data protection form, patient-related
questionnaires)
• Signed data protection form
Exclusion criteria
The following paents will not be included in the study:
• Patients affected by neurological and/or severe psychiat-
ric disorders (i.e., patients suffering from psychosis and/
or personality disorders)
• Patients with mild cognitive impairment
• Patients treated with drugs other than TKIs
Riva et al 219
© 2015 INTM, Italy. Published by Wichg Publishing
Variables assessed
Dierent types of variables will be assessed in the frame-
work of the current study including risk atude, risk prefer-
ences, cognive variables and personality traits related to risk,
as well as informaon about sociodemographic and clinical
data. Most of the variables will be derived from validated
quesonnaires.
All quesons will be presented in a paent booklet for
each single paent. Administraon of the quesonnaire will
take about 20 minutes per paent.
Measures of psychological aributes, risk atude, risk
preferences and need for cognive closure
The ALGA quesonnaire
ALGA is a quesonnaire developed to invesgate dierent
psychological areas and is mainly aimed at creang a paent
prole that can help physicians and other health-care provid-
ers to beer interact with paents (20). In this quesonnaire,
the 3 macro-areas under consideraon are (a) cognive, (b)
physical-related and (c) psychological, all invesgated in dif-
ferent subdimensions.
The benets of the ALGA paent proles are many:
• improved information delivery to the patient: with the
help of a graphical summary of the patient profile, doc-
tors can adjust the content and the level of verbal in-
formation to the patient’s perceived needs and level of
understanding.
• identification of important trials for enrolment: besides
being used to adjust the information provided, the patient
profile can also enable automatic identification of possible
clinical trials in which the patient could be enrolled. His/
her clinical information is compared against the eligibility
criteria of several trials and possible matches are identified.
This matching uses advanced algorithms over a semanti-
cally enriched clinical trial repository.
The Smulang-Instrumental Risk Inventory (SIRI)
The SIRI enables in-depth analysis of 2 characteriscs of risk-
taking behavior: (i) smulang risk taking (SRT), which is uncon-
trollable, impulsive, unconcerned with the magnitude of poten-
al losses, and dominated by emoonal processes and posive
arousal; it emphasizes potenal gains and focuses on the short
term; (ii) instrumental risk taking (IRT), which is controllable,
reecve, concerned with the magnitude of potenal losses,
and dominated by cognive processes and negave arousal; it
emphasizes potenal losses and focuses on the long term (21).
The quesonnaire comprises 17 items and uses 4-point rangs
(1 = does not describe me at all to 4 = describes me very well).
The Need for (Cognive) Closure Scale (NFCS)
The need for closure reects the desire for “an answer on
a given topic, any answer, as compared to confusion and am-
biguity” (22). A person characterized by a high need for closure
needs an immediate resoluon of the problem. The search for a
soluon makes the person hypervigilant and somemes panick-
ing looking for a soluon and tends to give the choice respon-
sibility to others. The need for cognive closure aects what
informaon individuals seek out, and how they process such
informaon to make a decision. The quesonnaire comprises
42 items and uses 6-point rangs (1 = strongly disagree to 6 =
strongly agree). The scale includes 5 subscales pertaining to
• desire for predictability
• preference for order and structure
• discomfort with ambiguity
• decisiveness
• close-mindedness
The Passive Risk Taking Scale (PRT)
The PRT measures the “passive risk” – risk brought on or
magnied by inacon. We dene passive risk taking as fore-
going an opportunity to act in order to reduce outcome vari-
ance (23). The quesonnaire is composed of 25 items and
uses a 7-point rang scale indicang to what extent people
are likely to behave in the manner described in each item. It
measures 3 subscales:
• risks that involve resources
• risks that involve medical issues
• risks that involve ethical issues
Measures of personality traits related to risk
The Resistance to Change (RTC) quesonnaire
The RTC is a personality quesonnaire specically focused
on evaluang traits related to risk. “Resistance to change”
can be dened as the acon taken by individuals when they
perceive a change that is occurring as a threat to them (24).
Therefore, the quesonnaire was designed to assess individu-
als’ tendencies “to resist or avoid making changes, to devalue
change generally, and to nd change aversive across diverse
contexts and types of change” (24).
TAbLE I - Hematological characteriscs of paents included in the
study
Type of response Denion
MCyR Major cytogenec
response
0-35% Ph+ marrow
metaphases
CCyR Complete cy togenec
response
0% Ph+ marrow
metaphases
MMR Major molecular
response
BCR-ABL/ABL
≤0.1% (IS)
MR4.0 BCR-ABL/ABL <0.01%
(IS) “4-log reducon”
MR4.5 or more BCR-ABL/ABL <0.003%
(IS) “4.5-log reducon”
IS = internaonal scale; MR = molecular response; Ph+ = Philadelphia chro-
mosome posive.
Raonale and protocol of CML study
220
© 2015 INTM, Italy. Published by Wichg Publishing
The quesonnaire is composed of 17 items and uses
6-point rangs (1 = strongly disagree to 6 = strongly agree). It
measures 4 subscales pertaining to
• routine seeking (RS): the behavioral component of
resistance to change, “the inclination to adopt rou-
tines”
• emotional reaction (ER): the affective component of re-
sistance to change, “the amount of stress and uneasi-
ness” induced by change
• short-term focus (SF): the affective component of re-
sistance to change, “the extent to which individuals are
distracted by the short-term inconveniences” associated
with change
• cognitive rigidity (CR): the cognitive component of resis-
tance to change, “frequency and ease with which people
change their minds”
Addional parameters
Addional variables will be assessed with regard to dif-
ferent parameters such as paent history and will be col-
lected by health-care professionals (8 quesons). A set of
ad hoc items (7 closed quesons) will be developed in or-
der to analyze the paent’s knowledge of and preferences
about the disease: diagnosis, treatment, and disease status.
Finally, clinical and sociodemographic data will be collected
as described in Table II. Regular appointments are expected
to be performed at the medical center, starng with the
appointment where the paent’s wrien consent to data
collecon is obtained.
The paent idener consists of 2 parts by which the
paents will be uniquely idened: (i) the center (2 digits),
and (ii) the paent idencaon number within the center
(2 numbers). Each center will be provided with a parcipa-
on list where each enrolled paent will be consecuvely
documented.
Stascal analysis
Sample size determinaon
This study is an NIS and the stascal approach is explor-
atory in nature. A sample size of 120 paents aged >18 has
been esmated. This sample size was chosen on the basis of
the following consideraons:
1. The available literature on discontinuation of therapy in
CML patients is very limited.
2. In the absence of preliminary data, sample size criteria
will be based on subjective experience.
3. A sample size of 120 patients seems to be reasonable
in view of the fact that no dropout is to be considered
as this is a cross-sectional study. It has to be taken into
account that CML is a relatively rare disease with an
even lower prevalence of those eligible to discontinue
therapy.
4. A possible reconsideration of the sample size is expected
as the first 10 patients will have been recruited and pre-
liminary results will be available.
Stascal data
The study objecve is addressed in an exploratory man-
ner to idenfy the risk prole of CML paents at dierent
disease stages. Descripve stascs will be used to summa-
rize sociodemographic, clinical and psychological/cognive
data (paent preferences, cognive measures of risk a-
tude, risk preferences and need for cognive closure, and
measures of personality traits related to risk) by means of
frequency, median, mean, SD, range, and condence inter-
vals. Descripve stascs will give an overview of the study
populaon. This will be synthesized with a graph for distri-
buon of the study populaon, esmates of means, medi-
ans and standard errors. In addion, a scaerplot will be
generated to idenfy and examine possible outliers in the
sample. The data will be analyzed using the SPSS soware,
version 22 or higher.
A documentaon form will be completed for each pa-
ent. The forms will be reviewed for accuracy and com-
pleteness by the principal invesgator. Data will be entered
directly into SPSS. Computer-aided data validaons will be
performed on an ongoing basis. Coding of medical terms
and medicaon will be done using the current versions
of coding diconaries (Medical Diconary for Regulatory
Acvies [MedDRA] and World Health Organizaon Drug
Diconary Enhanced [WHO-DDE]) as specied in the data
management plan. Quesonnaire data transformed into
SPSS will undergo further detailed analysis by the principal
invesgator, who is an expert in psychological analysis.
Conclusions
Stopping treatment with TKIs appears interesng and at
the same me challenging to many paents. This choice has
the potenal to drascally modify clinical pracce in the man-
agement of CML in the near future. Analyzing how and when
paents would consider this opon is an important aspect in
the care of this condion. Understandably, as the few avail-
able studies and reports have discussed, many paents re-
port fear about the choice because it makes them feel they
lose control over the disease. Moreover, they may be fright-
ened by the possibility of disease resistance upon restarng
TKI therapy aer treatment interrupon.
TAbLE II - Clinical and sociodemographic data
Clinical data Demographic data
Treatment history Marital status (past, current)
Disease history Partnership
Concomitant diseases Living situaon (e.g., household
size, geographical locaon)
Treatment modality Professional situaon (e.g.,
employment status)
Concomitant treatment Educaon
Medical visits
Riva et al 221
© 2015 INTM, Italy. Published by Wichg Publishing
This study is unique in that it will analyze in depth all pos-
sible psychological, cognive and emoonal variables that can
inuence the medical decision of treatment disconnuaon
in CML, giving insights that can be used in clinical pracce for
informing future discussions with paents who consider stop-
ping treatment with TKIs. Moreover, the study will provide new
informaon to plan future clinical trials on CML treatment dis-
connuaon.
Acknowledgment
The study group want to thank Prof. Paolo Corradini (Univer-
sity of Milan) and all the other REL (Rete Ematologica Lombarda-
Hematological Network of Lombardy) members who have expressed
interest for this study.
Disclosures
Financial support: The study is supported by the University of
Milan, which has received a limited grant from Novars for supporng
regulatory procedures.
Conict of interest: None of the authors has any nancial interest
related to this study to disclose.
References
1. Mahon FX. Disconnuaon of tyrosine kinase therapy in CML.
Ann Hematol. 2015;94(Suppl 2):S187-S193.
2. Kimura S, Ashihara E, Maekawa T. New tyrosine kinase inhibi-
tors in the treatment of chronic myeloid leukemia. Curr Pharm
Biotechnol. 2006;7(5):371-379.
3. Jabbour E, Cortes JE, Giles FJ, O’Brien S, Kantarjian HM. Current
and emerging treatment opons in chronic myeloid leukemia.
Cancer. 2007;109(11):2171-2181.
4. Ross DM, Branford S, Seymour JF, et al. Paents with chronic
myeloid leukemia who maintain a complete molecular response
aer stopping imanib treatment have evidence of persistent
leukemia by DNA PCR. Leukemia. 2010;24(10):1719-1724.
5. Baccarani M, Deininger MW, Ros G, et al. European Leuke-
miaNet recommendaons for the management of chronic my-
eloid leukemia: 2013. Blood. 2013;122(6):872-884.
6. Riva S, Camerini AL, Allam A, Schulz PJ. Interacve secons
of an Internet-based intervenon increase empowerment of
chronic back pain paents: randomized controlled trial. J Med
Internet Res. 2014;16(8):e180.
7. Riva S, Schulz P, Staoni L, Schoeb V. Paent parcipaon in dis-
charge planning decisions in the frame of primary nursing ap-
proach: a conversaon analyc study. Stud Commun Sci. 2014;
14(1):61-67.
8. Schulz PJ, Hartung U, Riva S. Causes, coping, and culture: a
comparave survey study on representaon of back pain in
three Swiss language regions. PLoS ONE. 2013;8(11):e78029.
9. Baldi PL, Iannello P, Riva S, Antonie A. Cognive reecon
and socially biased decisions. Stud Psychol (Brasl). 2013;5(4):
265-271.
10. Riva S, Mon M, Iannello P, Praveoni G, Schulz PJ, Antonie A.
A preliminary mixed-method invesgaon of trust and hidden
signals in medical consultaons. PLoS ONE. 2014; 9(3):e90941.
11. Guastaerro S, Falcone U, Celentano M, Coppola M, Ferrara
MG, Sica A. Is it possible to disconnue imanib mesylate ther-
apy in chronic myeloid leukemia paents with undetectable
BCR/ABL? A case report and a review of the literature. Leuk
Res. 2009;33(8):1079-1081.
12. Mahon FX, Fort MP, Eenne G, et al. Interferon alpha alone is able
to cure chronic myeloid leukemia in a small subset of paents
despite the persistence of leukemic cells: experience of long fol-
low up aer treatment disconnuaon. Blood. 2010;116:2299
(abstract). ASH Annual Meeng Abstracts.
13. Mahon FX, Réa D, Guilhot J, et al; Intergroupe Français des
Leucémies Myéloïdes Chroniques. Disconnuaon of imanib
in paents with chronic myeloid leukaemia who have main-
tained complete molecular remission for at least 2 years: the
prospecve, mulcentre Stop Imanib (STIM) trial. Lancet On-
col. 2010;11(11):1029-1035.
14. Cortes J, Goldman JM, Hughes T. Current issues in chronic my-
eloid leukemia: monitoring, resistance, and funconal cure. J
Natl Compr Canc Netw. 2012;10(Suppl 3):S1-S13.
15. Cornelison M, Jabbour EJ, Welch MA. Managing side eects of
tyrosine kinase inhibitor therapy to opmize adherence in pa-
ents with chronic myeloid leukemia: the role of the midlevel
praconer. J Support Oncol. 2012;10(1):14-24.
16. Sweet K, Oehler V. Disconnuaon of tyrosine kinase inhibi-
tors in chronic myeloid leukemia: when is this a safe opon
to consider? Hematology (Am Soc Hematol Educ Program).
2013;2013(1):184-188.
17. Baccarani M, Ecace F, Ros G. Moving towards paent-cen-
tered decision-making in chronic myeloid leukemia: assessment
of quality of life and symptom burden. Haematologica. 2014;
99(2):205-208.
18. Cortes J, O’Brien S, Kantarjian H. Disconnuaon of imanib
therapy aer achieving a molecular response. Blood. 2004;
104(7):2204-2205.
19. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC,
Vandenbroucke JP; STROBE Iniave. The Strengthening the
Reporng of Observaonal Studies in Epidemiology (STROBE)
statement: guidelines for reporng observaonal studies. J
Clin Epidemiol. 2008;61(4):344-349.
20. Gorini A, Mazzocco K, Gandini S, Munzone E, McVie G, Pravet-
toni G. Development and psychometric tesng of a breast can-
cer paent-proling quesonnaire. Breast Cancer (Dove Med
Press). 2015;7:133-146.
21. Zaleskiewicz T. Beyond risk seeking and risk aversion: person-
ality and the dual nature of economic risk taking. Eur J Pers.
2001;15(S1):S105-S122.
22. Webster DM, Kruglanski AW. Individual dierences in need for
cognive closure. J Pers Soc Psychol. 1994;67(6):1049-1062.
23. Keinan R, Bereby-Meyer Y. “Leaving it to chance”— Passive
risk taking in everyday life. Judgment and Decision Making.
2012;7(6):705-715.
24. Oreg S. Resistance to change: developing an individual dier-
ences measure. J Appl Psychol. 2003;88(4):680-693.
