Effect of Estrogen Depletion on Pain Sensitivity in Aromatase Inhibitor–Treated Women With Early-Stage Breast Cancer

The Journal of Pain, Vol -, No - (-), 2014: pp 1-8Available online at www.jpain.org and www.sciencedirect.com

Effect of Estrogen Depletion on Pain Sensitivity in Aromatase

Inhibitor–Treated Women With Early-Stage Breast Cancer

N. Lynn Henry,* Anna Conlon,y Kelley M. Kidwell,y Kent Griffith,y

Jeffrey B. Smerage,* Anne F. Schott,* Daniel F. Hayes,* David A. Williams,z

Daniel J. Clauw,z and Steven E. Hartez

*Breast Oncology Program, University of Michigan Comprehensive Cancer Center, Ann Arbor, Michigan.yDepartment of Biostatistics, University of Michigan School of Public Health, Ann Arbor, Michigan.zDepartment of Anesthesiology, Chronic Pain and Fatigue Research Center, University of Michigan Medical School,Ann Arbor, Michigan.

ReceivedResearchsupporte(grant nMichiganFashion(D.F.H.).UL1RR02nent ofMedicaland do nResearchN.L.H. reand sanoNovartis,and gran

Abstract: Aromatase inhibitors (AIs), which are used to treat breast cancer, inhibit estrogen produc-

tion in postmenopausal women. AI-associated musculoskeletal symptoms occur in approximately half

of treatedwomen and lead to treatment discontinuation in 20 to 30%. The etiology may be due in part

to estrogen deprivation. In premenopausal women, lower estrogen levels have been associated with

increased pain aswell aswith impairment of descending pain inhibitory pathways, whichmay be a risk

factor for developing chronic pain. We prospectively tested whether AI-induced estrogen deprivation

alters pain sensitivity, thereby increasing the risk of developing AI-associated musculoskeletal symp-

toms. Fifty postmenopausal breast cancer patients underwent pressure pain testing and conditioned

pain modulation (CPM) assessment prior to AI initiation and after 3 and 6months. At baseline, 26 of 40

(65%) assessed patients demonstrated impaired CPM, which was greater in those who had previously

received chemotherapy (P = .006). No statistically significant change in pressure pain threshold or CPM

was identified following estrogen deprivation. In addition, there was no association with either

measure of pain sensitivity and change in patient-reported pain with AI therapy. AI-associated muscu-

loskeletal symptoms are not likely due to decreased pain threshold or impaired CPM prior to treatment

initiation, or to effects of estrogen depletion on pain sensitivity.

Perspective: This article presents our findings of the effect of estrogen deprivation on objective

measures of pain sensitivity. In postmenopausal women, medication-induced estrogen depletion

did not result in an identifiable change in pressure pain threshold or CPM. Impaired CPM may be

associated with chemotherapy.

ª 2014 by the American Pain Society

Keywords: Pain threshold, conditionedpainmodulation, breast cancer, estrogendeprivation, aromatase

inhibitor.

Aromatase inhibitors (AIs) inhibit the conversion ofandrogens to estrogens in postmenopausalwomen, thereby reducing circulating estrogen

December 5, 2013; Accepted January 4, 2014.funding: N.L.H. is a Damon Runyon-Lilly Clinical Investigatord (in part) by the Damon Runyon Cancer Research Foundationumber CI-53-10). This work was also supported in part by theInstitute for Clinical and Health Research (N.L.H.) and the

Footwear Association of New York/QVC Presents Shoes on SaleThis publication was made possible by Grant Number4986 from the National Center for Research Resources, a compo-the National Institutes of Health (NIH), and NIH Roadmap forResearch. Its contents are solely the responsibility of the authorsot necessarily represent the official views of National Center forResources or National Institutes of Health.ceives research funding from AstraZeneca, Lilly Pharmaceuticals,fi aventis. D.F.H. received research funding from AstraZeneca,Pfizer, and Veridex LLC. D.J.C. has received consulting incomet support from Pfizer, Forest, Wyeth, Cerephex, and Nuvo, and

levels to approximately 1/10th of normal levels.8 Thisclass of medication has been shown to significantlyimprove breast cancer outcomes in postmenopausal

consulting income from Lilly, UCB, Cerephex, Tonix, and Purdue Labs.S.E.H. has received grant support from Cerephex, Merck, and Forest Lab-oratories.No other authors reported conflicts of interest.Supplementary data accompanying this article are available online atwww.jpain.org/ and www.sciencedirect.com/.Address reprint requests to N. Lynn Henry,MD, PhD, Breast Oncology Pro-gram, University of Michigan Comprehensive Cancer Center, 1500 EastMedical Center Drive, Med Inn Building C450, Ann Arbor, MI 48109-5843. E-mail: [email protected]

1526-5900/$36.00

ª 2014 by the American Pain Society

http://dx.doi.org/10.1016/j.jpain.2014.01.487

1

2 The Journal of Pain Effect of Estrogen Depletion on Pain Sensitivity

women with hormone receptor (HR)-positive breastcancer compared to the prior standard of care, tamox-ifen.5 Treatment with an AI has therefore been recom-mended as part of the treatment regimen for allpostmenopausal women with HR-positive early-stagebreast cancer.2

Despite positive oncologic benefits, AIs can causebothersome long-term side effects.4 These toxicities cannegatively impact quality of life, which can lead todecreased adherence to AI therapy.19 In particular, AI-associated musculoskeletal symptoms (AIMSS) (eg, painand stiffness) can affect up to half of treated patientsand is the primary reason for discontinuation reportedby about 75% of those who discontinue treatment early.Overall, 20 to 30% of patients stop taking their initiallyprescribed AI medication because of toxicity.The mechanism of AIMSS is poorly understood but is

thought to be related to the profound depletion ofestrogen, either systemically or locally in joints.6,21 Abetter understanding of the mechanism(s) thatunderlie AIMSS, and why some women develop themand others do not, may lead to interventions toprevent or treat them. Such interventions might resultin higher levels of adherence and, in theory, evenbetter outcomes for women with estrogen receptor–positive breast cancer.Centrally mediated descending control over pain is

thought to utilize at least 2 pathways in humans: thenorepinephrine–serotonin pathway and the opioidergicpathway. Estrogen may play a role in the modulation ofthe latter pathway. During a pain stressor, women withlow levels of estrogen have been shown to havedecreased activation of opioid neurotransmission, as as-sessed by positron emission tomography resulting in hy-peralgesic responses.35 In addition, a few studies inpremenopausal women have reported an association be-tween patient-reported pain severity and phase of themenstrual cycle,with lower levels of estrogen being asso-ciated with greater pain sensitivity.18,23 Despite thesefew studies, the relationship between estrogenconcentration and pain sensitivity remains complex andpoorly understood.Quantitative sensory testing (QST) permits the stan-

dardized assessment of pain sensitivity.15 Pressure painthreshold can be quantitated using an evoked pain stim-ulus at the thumbnail or other sites.9-11,14,26,27 Inaddition, the magnitude of endogenous descendinginhibition of neurons in response to noxious stimuli canbe assessed using conditioned pain modulation (CPM)studies.39 CPM defines the physiologic response torepeated or chronic pain stimuli. Deficient or inefficientCPM is observed in patients with chronic pain syndromesof a variety of types. These QST methods, including bothpressure pain threshold assessment and CPM studies,have been used successfully to demonstrate deficienciesin central pain processing and modulation in patientswith many different chronic pain conditions comparedto healthy controls. Such testing can also be used todifferentiate underlying mechanisms within groups ofindividuals having nociceptive, neuropathic, or central-ized forms of chronic pain.28,37,38 QST findings are

highly correlated with changes in functional imagingactivation patterns and clinical pain in patients withmultiple different chronic pain conditions.7,24,25

Based on these considerations, we hypothesized thatpreexisting low pain threshold and/or deficiencies inCPM would predispose women to AIMSS. We also hy-pothesized that AI-induced estrogen deprivation woulddecrease pain threshold and/or impair CPM over time,thereby increasing the risk of subsequently developingAIMSS. We therefore enrolled a cohort of women withearly-stage HR-positive breast cancer who were initi-ating AI therapy into a prospective study in which weapplied QST methods to evaluate the effect of estrogendepletion on pain sensitivity and the predictive role oflow pressure pain threshold and impaired CPM for devel-opment of AIMSS.

Methods

PatientsEligible patientswere recruited to this prospective clin-

ical study from June 2009 until January 2012 at a singleinstitution (Clinicaltrials.gov NCT01814397). Patientswere eligible if they were postmenopausal womenwith stage 0-III HR-positive breast cancer who were plan-ning to receive a standard dose of AI therapy (ie, anastro-zole 1 mg orally daily, exemestane 25 mg orally daily, orletrozole 2.5 mg orally daily). Surgical resection, chemo-therapy, and radiation therapy, as indicated, werecompleted prior to study enrollment. Patients wereineligible if they had received prior AI therapy, hadpreexisting grade 2 or higher sensory neuropathy, hadchemotherapy-induced fingernail changes thought tointerfere with QST, had a history of schizophrenia, orhad a history of suicidal ideation or attempt within the2 years prior to enrollment. Patients who reported anaverage pain of at least 8 out of 10 at baseline wereexcluded. The protocol was approved by the Universityof Michigan Institutional Review Board, and all enrolledpatients provided written informed consent.

Study DesignPatients completed all evaluations in the University of

Michigan Chronic Pain and Fatigue Research Center. Priorto initiation of AI therapy, patients underwent phlebot-omy and QST and completed self-report questionnaires,as described below. Following the baseline visit, patientsinitiated treatment with 1 of the 3 AI medications asdirected by their treating medical oncologist. After3 months of AI therapy, patients again underwent phle-botomy andQST, and after 6months theyunderwentQST.

QSTQST was performed using a standardized proto-

col.10,12,13,16 In brief, discrete pressure stimuli weredelivered by a custom-built apparatus that eliminateddirect patient contact by the examiner (SupplementaryFigure 1). This device employed a hydraulic system toapply pressure to the thumbnail bed via a 1-cm2 hardrubber circular probe. The probe was positioned over

Henry et al The Journal of Pain 3

the center of the patient’s nondominant thumbnail by ahand-held plastic housing, and the hydraulic system wasactivated by placing calibrated weights on a moveableplatform and adjusting valves to control stimulus timing.The probe was lowered to apply pressure consistent withthe weight on the moveable platform. The combinationof valves and calibrated weights produced controlledand repeatable stimulation. All patients underwent de-vice familiarization and training prior to testing.To assess pressure pain threshold, the testing sequence

consisted of a series of ascending pressure stimuli deliv-ered at 25-second intervals beginning at .5 kg/cm2 andincreasing in .5-kg/cm2 increments until tolerance or toa maximum of 10 kg/cm2 (Supplementary Figure 2A).The duration of each pressure was 5 seconds. Patientsrated the intensity of each pressure sensation using a0 to 100 numerical rating scale (0 = no pain, 100 = worstpain imaginable). These pain ratings were used to inter-polate via a regressed function each individual’s Pain50.Pain50, defined as the amount of applied pressure inkilograms per square centimeter that evoked a pain in-tensity rating of 50 out of 100, served as a measure ofsuprathreshold pressure pain sensitivity.Endogenous pain modulation was then evaluated

using the pressure delivery apparatus described aboveand following a standardized CPM paradigm(Supplementary Figure 2B).39 CPM procedures use aconditioning stimulus (a noxious stimulus that activatespain modulatory systems) and a test stimulus (a noxiousstimulus used to evaluate the analgesic response to thetest stimulus). Pressure equivalent to Pain50 for the indi-vidual patient was applied via a probe to the nondomi-nant thumbnail for 30 seconds as a test stimulus, andthe patient rated the intensity of the pressure at10-second intervals. Ten minutes later, a pressure condi-tioning stimulus was then continuously applied to thedominant thumbnail for 60 seconds at the same Pain50intensity as the test stimulus. After 30 seconds of condi-tioning stimulation, the test stimulus was again appliedto thenondominant thumbnail for 30 secondsand thepa-tient rated the intensity of the pressure every 10 seconds.CPMmagnitude was calculated as the difference (secondminus first) in the mean of the 3 pain ratings to the teststimulus applied prior to and during the conditioningstimulus. Higher CPM values indicate less efficient CPM.

Patient-Reported OutcomesPatients completed self-report questionnaires at each

time point. Pain, the primary symptom of interest, wasassessed using an 11-point Likert-type scale. Each patientrecorded her average pain daily for 7 consecutive daysprior to each time point, and values were averaged toobtain an average pain value for each time point. Addi-tional patient-reported outcomes that were assessed atbaseline included fatigue (Multidimensional FatigueIndex),33 sleep problems (Medical Outcomes Study–Sleepquestionnaire),17,36 cognitive dysfunction (MultipleAbilities Self-Report Questionnaire),31 and depression(Center for Epidemiologic Studies–Depression Scale[CESD]).29

Estradiol AssessmentSerum was collected prior to AI initiation and after

3 months of therapy. Serum estradiol quantitation wasperformed using an ultrasensitive gas chromatographytandem mass spectroscopy–based assay as previouslydescribed (inVentiv Health Clinical).30 Lower limit ofquantitation was 2 pg/mL.

Statistical MethodsThe primary endpoint of the study was change in pain

threshold (Pain50) and CPM between the baseline and3-month assessments. Change over time in patient-reported pain, Pain50, and CPM were assessed usingrepeated measures mixed models using the square rootof pain, the natural log of Pain50, and CPM to evaluatefor normality of outcomes, without covariates. Forassessment of change in pain50, with 50 patients wehad 80% power to detect a change of .6 to .73 over3 months, assuming a standard deviation of 1.5 to 1.8.For CPM, we had 80% power to detect a change of 7.3,assuming a standard deviation of 18.Associations between patient-reported symptoms and

pain sensitivity scores were assessed using Spearmanrank correlation because of the nonnormal distributionof the symptom scores. Associations between baselinepatient-reported symptoms and treatment discontinua-tion were assessed using a 2-sample Wilcoxon Mann-Whitney test. Linear regression models were used toassess associations between baseline estradiol levelsand the baseline QST measures. Linear regressionmodels were also used to assess associations betweenbaseline QST measures with the clinical factors of priorchemotherapy use, prior tamoxifen use, and bodymass index, as well as treatment discontinuation. Forall statistical tests, P values #.05 were considered statis-tically significant.

Results

Baseline CharacteristicsA total of 50 postmenopausal patients enrolled and

underwent baseline evaluation (Fig 1). Baseline charac-teristics are listed in Tables 1 and 2. Twenty-eightpercent of patients had received prior tamoxifen, and46% had been treated with chemotherapy. Eightypercent of patients received anastrozole. Baselinepatient-reported pain was 1.7 (standard deviation [SD]= 1.3) on a 0 to 10 scale.At baseline, therewas no statistically significant associ-

ation between patient-reported pain and depression.The average pain reported by the 5 patients with CESDscores within the possibly or probably depressed range(CESD $ 16) was 2.5 (SD = 1.9), compared to an averageof 1.6 (SD = 1.3) for those with CESD scores lower than16 (P = .19). Baseline patient-reported painwas also asso-ciatedwith both poor sleep quality (r = .56, P < .0001) andgeneral fatigue (r = .49, P = .0006), but not with poorcognitive function (r = .14, P = .41).There was considerable interpatient variability in mea-

sures of pain sensitivity prior to AI initiation (Fig 2).Mean

Figure 1. Patient flow diagram.

4 The Journal of Pain Effect of Estrogen Depletion on Pain Sensitivity

baseline Pain50 was 4.3 kg/cm2 (SD = 1.5). During assess-ment for CPM, mean scores for the test stimulusincreased from 44.2 (SD = 20.2) to 50.7 (SD = 20.6) withthe conditioning stimulus, which equates to a meanbaseline CPMmagnitude of 7.9 (SD = 14.7). Of the 40 pa-tients who underwent CPM assessment at baseline, 26(65%) had impaired (>0) levels. The mean baseline CPMfor those who received (n = 19) and did not receive(n = 21) chemotherapy was 14.6 and 2.0, respectively(P = .006; Fig 3). No differences were noted accordingto prior chemotherapy for baseline Pain50 (4.4 vs 4.2,P = nonsignificant) or patient-reported pain (1.8 vs 1.6,P = nonsignificant).

Table 1. Baseline Patient Characteristics

CHARACTERISTIC

TOTAL COHORT

(N = 50)

Median age, years (range) 60 (38–77)

Race

White 47 (94%)

Black 2 (4%)

Other 1 (2%)

Mean weight, kg (SD) 80.7 (17.8)

Mean body mass index (SD) 30.0 (6.5)

AI

Anastrozole 40 (80%)

Exemestane 1 (2%)

Letrozole 9 (18%)

Prior chemotherapy 23 (46%)

Prior taxane 20/23 (87%)

Time since chemotherapy completion, year (range) .3 (.1–3.6)

Prior tamoxifen 14 (28%)

Mean baseline quantitative sensory testing

measures (SD)

Pain50, kg/cm2 4.3 (1.5)

Baseline CPM 7.9 (14.7)

Mean baseline serum estradiol concentration, pg/mL (SD) 6.0 (7.6)

No associations were detectedwith either prior tamox-ifen or body mass index and baseline patient-reportedpain, Pain50, or CPM. Baseline general fatigue was asso-ciated with baseline CPM (r = .33, P = .04) but not withbaseline Pain50 (r = –.06, P = nonsignificant). No associa-tions were identified between baseline depression, sleepquality, or cognitive function and either baseline Pain50or CPM.

Change in Pain Sensitivity WithAI-Induced Estrogen DeprivationAverage baseline serum estradiol concentration was

6.0 pg/mL (SD = 7.6; Supplementary Figure 3). After3 months of AI therapy, 38 of 43 (88%) assessed patientshad undetectable serum estradiol concentrations (<.625pg/mL). The estradiol concentrations of the other 5 pa-tients at the 3-month time point ranged from .67 to3.38 pg/mL, which were all within the lower portion ofthe postmenopausal range (<10 pg/mL).No change with AI therapy in mean patient-reported

pain, Pain50, or CPM was identified (Fig 2, Table 3).There were no associations detected between changein mean patient-reported pain, Pain50, or CPM andeither prior chemotherapy or prior tamoxifen. In addi-tion, no associations were detected between changein patient-reported pain and change in either Pain50or CPM between baseline and 3 months (data notshown).

Association Between Pain Sensitivity andDiscontinuation of AI Therapy Because ofMusculoskeletal SymptomsSeven patients (14%) discontinued therapy by

6months because ofmusculoskeletal symptoms. No asso-ciation was detected between baseline Pain50 or CPMand discontinuation of therapy because of musculoskel-etal toxicity (data not shown). Patients who discontinuedAI therapy by 6 months had an average Pain50 of 2.25(SD = .61) at the 3-month time point, whereas thosewho continued treatment had an average Pain50 of4.40 (SD = 1.77) at 3 months (P = .01) (Table 3,Supplementary Figure 4). No association was identifiedbetween average CPM at 3 months and treatmentdiscontinuation by 6 months.

Association Between Patient-ReportedSymptoms and Discontinuation of AITherapy Because of MusculoskeletalSymptomsMean patient-reported pain in this patient cohort was

1.7 (SD = 1.3) at baseline, 2.1 (SD = 1.8) at 3 months, and1.7 (SD = 1.4) at 6 months (Table 3). Patients who discon-tinued AI therapy by 6 months reported an average painof 4.17 (SD = 1.31) at the 3-month time point, whereasthose who continued treatment reported an averagepain of 1.85 (SD = 1.76) at 3 months (P = .021). No associ-ation was detected between baseline patient-reportedpain and discontinuation of therapy because ofmusculo-skeletal toxicity.

Table 2. Baseline Patient-Reported Symptoms by Treatment-Discontinuation Status

CHARACTERISTIC

TOTAL (N = 50)DISCONTINUED AI BY 6 MONTHS

DUE TO MSK SYMPTOMS (N = 7) ALL OTHERS (N = 43)

P VALUE*MEAN SD MEAN SD MEAN SD

Pain diary (0–10) 1.7 1.3 1.9 1.2 1.6 1.3 .52

Depression (CESD) 7.6 6.9 14.0 9.5 6.6 5.9 .033

Sleep (MOS-Sleep, SPDX2) 31.2 18.8 47.1 17.9 28.6 17.8 .013

Fatigue (MFI)

General fatigue 11.5 4.5 16.9 3.6 10.6 4.1 .003

Physical fatigue 10.4 4.3 14.9 2.7 9.7 4.1 .004

Reduced activity 9.3 4.4 12.9 3.9 8.7 4.3 .021

Reduced motivation 7.6 4.0 11.2 2.3 7.0 4.0 .006

Mental fatigue 8.3 4.3 9.0 3.7 8.1 4.4 .49

Cognitive function (MASQ)

Language 14.0 3.4 15.0 4.9 14.4 3.8 .86

Visual-perception ability 10.7 3.1 10.3 2.0 10.8 3.2 .97

Verbal memory 16.3 3.6 17.3 1.6 16.1 3.8 .42

Visual-spatial memory 14.0 3.3 14.7 2.1 13.9 3.4 .31

Attention/concentration 15.5 3.4 16.8 2.9 15.4 3.5 .31

Abbreviations: MSK, Musculoskeletal; MFI, Multidimensional Fatigue Index; MOS, Medical Outcomes Study; SPDX2, Sleep Problems Index II; MASQ, Multiple Abilities

Self-Report Questionnaire.

*From 2-sample Wilcoxon Mann-Whitney test.

Henry et al The Journal of Pain 5

Patients were also assessed for nonpain symptomspresent prior to initiation of AI therapy using validatedquestionnaires (Table 2). Patients who discontinued AItherapy by 6 months were more likely to report higherbaseline levels of depression (CESD score = 14.0 [SD =9.5] vs 6.6 [SD = 5.9], P = .033), poor sleep quality (SleepProblems Index II = 47.1 [SD = 17.9] vs 28.6 [SD = 17.8],P = .013), and fatigue (Multidimensional Fatigue Indexgeneral fatigue = 16.9 [SD = 3.6] vs 10.6 [SD = 4.1],P = .003) compared to those who continued treatment

Figure 2. Effect of estrogen depletion on quantitative sensory testpain threshold (Pain50) and (C)CPMwithAI therapy. Individual patie(mo), and triangles at 6 months. Points above the horizontal dotted

beyond 6 months. No associations were noted be-tween baseline cognitive function and treatmentdiscontinuation.

DiscussionIn this prospective study utilizing QST, we were unable

to confirm our hypothesis that preexisting high painsensitivity or impaired descending pain inhibitory path-ways predispose women to AIMSS. In addition, we failed

ing measures. Change in (A) patient-reported pain, (B) pressurents are representedby circles at baseline (BL), squares at 3monthsline in C reflect impaired CPM.

Figure 3. Mean CPM for those who were or were not treatedwith chemotherapy (chemo). Circles represent individual pa-tients who received chemotherapy, squares represent individualpatients who did not receive chemotherapy, and horizontallines represent mean values.

6 The Journal of Pain Effect of Estrogen Depletion on Pain Sensitivity

to confirm our second hypothesis that interpatient dif-ferences in estrogen deprivation leads to an increase inpain sensitivity or impairment in CPM in AI-treated post-menopausal breast cancer survivors. Therefore, it is un-likely that estrogen deprivation within the centralnervous system leading to a generalized increase inevoked pain sensitivity is contributing substantially tothe increase in musculoskeletal pain, which is experi-enced by about half of AI-treated women.A smaller thanexpectednumberofpatients inour study

discontinued treatment by 6 months because of new orworsened musculoskeletal symptoms. Baseline pain,pain sensitivity, and CPM were not predictive of earlytreatment discontinuation. As expected, averagepatient-reported pain was higher at the 3-month timepoint in those patients who subsequently discontinuedAI therapy by 6months because ofmusculoskeletal symp-toms. Interestingly, average pain threshold was signifi-cantly lower at the 3-month time point in those patientswho discontinued AI therapy by 6 months. Although thesample size is limited, these findings suggest that in

Table 3. Patient-Reported Pain and QuantitativeSensory Testing Measures at All 3 Time Points

MEASURE

ALL

PATIENTSP

VALUE

STOPPED

EARLYCONTINUED

AIP

VALUE

Patient-reported pain

Baseline 1.7 (1.3) .75 1.9 (1.2) 1.6 (1.3) .52

3 months 2.1 (1.8) 4.2 (1.3) 1.9 (1.8) .021

6 months 1.7 (1.4) n/a 1.7 (1.4)

Pain50, kg/cm2

Baseline 4.3 (1.5) .41 3.5 (1.3) 4.4 (1.6) .24

3 months 4.2 (1.8) 2.3 (.6) 4.4 (1.8) .01

6 months 4.2 (1.6) n/a 4.2 (1.6)

CPM

Baseline 7.9 (14.7) .66 14.2 (8.0) 7.2 (15.2) .25

3 months 6.3 (18.4) 15.0 (20.3) 5.1 (18.2) .52

6 months 10.4 (18.9) n/a 10.4 (18.9)

NOTE. Values are mean (SD).

some patients, there may be an association betweenincreased sensitivity to pain and intoleranceofAI therapy.Prior chemotherapy has been identified as a risk factor

for development of AI-associated musculoskeletal symp-toms in multiple clinical trials.3,19,32 In this study, wefound an association between prior treatment withchemotherapy and impairment of the descendinginhibitory pain system at baseline, prior to AI therapy.This finding suggests that chemotherapy-induced nervedamage might ‘‘prime’’ an individual for the subsequentdevelopment of pain in part by leading to impairedCPM. The recent studies showing that beneficial responseto duloxetine in diabetic neuropathic pain can be pre-dicted by less efficient baseline CPM, as well as the evi-dence that duloxetine is effective in treatingchemotherapy-induced pain, offer additional supportfor this hypothesis.34,40 In addition, we previouslydemonstrated that duloxetine has activity in treatmentof AI-associated musculoskeletal pain.20

Furthermore, prior research has demonstrated thatsome women who are unable to tolerate one AI medica-tion are able to tolerate a different AI medication.Because the degree of estrogen level suppression issimilar for all AI medications, these data suggest that es-trogen deprivation alone is not the cause of AI-associated arthralgias.1,19 Therefore, it remains unclearwhy somewomen experience AI-associatedmusculoskel-etal pain and others do not.Our results also have important implications for pain

research more generally. Previously reported studies inpremenopausal womenwith chronic pain have suggesteda linkbetween low-estrogenperiodsduring themenstrualcycle and higher reported pain.18,23 In contrast, weidentified no change in pain sensitivity, CPM, or patient-reported pain with the profound 1 logarithm reductionin estradiol associated with aromatase inhibition in thiscohort of postmenopausal women. Although there wasconsiderable intrapatient variability in QST measures inpatients over time, the mean values during estrogendeprivation did not differ from the pretreatment values.The lowerbaseline serumestrogen concentrations inpost-menopausal women in this study, coupledwith the imme-diate effect of AIs on estrogen, are in contrast to the slowand intermittent variations that occur during the men-strual cycle. Thus, our study is a different test of whetherestrogen levels directly affect pain processing than the de-signs used in these previous studies.In addition to evaluating the effect of pain sensitivity

measures on treatment discontinuation, we also investi-gated associations between persistence with therapyand baseline patient-reported symptoms that werealready present at the time of AI initiation. We failedto identify an association between preexisting pain andtreatment discontinuation due to pain, which is similarto previously reported findings in a different trial.19

However, in this patient cohort we identified statisticallysignificant associations between baseline depression,poor sleep quality, and fatigue and early discontinuationof AI therapy because of musculoskeletal symptoms. Thefinding that increased global symptom burden, ratherthan just pain-related symptoms, may impact persistence

Henry et al The Journal of Pain 7

with adjuvant endocrine therapy suggests that manage-ment of the symptom cluster may be more effective thanfocusing specifically on analgesia.One limitation of our study is relatively small sample

size, which could limit our power to detect differencesin measures of pain sensitivity in AI-treated women.However, despite the small sample size, we had 80%power to detect relatively small changes in pain sensi-tivity and were unable to detect changes of that magni-tude. In addition, given the stability of Pain50 and CPMvalues between the baseline and 3-month time points(Fig 2), in conjunction with the biochemically verifiedsuppression of circulating estradiol, it is therefore un-likely that a clinically significant change in pain sensi-tivity parameters would be identified in a largerpatient cohort.The unexpectedly low incidence of development of

moderate or severe pain in our study population (17%at 3 months) could also limit our power to detect associ-ations between pain sensitivity measures and patient-reported symptoms. Finally, the findings could be biasedtoward the null if patients with data missing eitherbecause of early treatment discontinuation due to symp-toms or because of patient request to discontinue partic-ipation in the overall study or to undergo CPMassessment at the 3-month time point had substantial de-creases in their pain threshold.Another challenge is the lack of an established defini-

tion of AIMSS in the literature. This is in part due to thedifficulty accounting for variability in patient-reported

pain symptoms with AI therapy (eg, arthralgia, myalgia,tendonitis), the impact of concomitant over-the-counterand prescription medications, and the need to parse outchange in pain in patients who may have preexistingpain from prior surgery or chemotherapy or from comor-bidities common in this postmenopausal population,especially osteoarthritis.We therefore chose to use treat-ment discontinuation due to musculoskeletal symptomsas a surrogate endpoint for AIMSS, as we have done inprevious publications.19,22

In summary, estrogen deprivation with AI therapydid not impact experimental pain sensitivity or de-scending pain modulation. Additional studies areneeded to better understand how effects of priorchemotherapy might contribute to the pain and intol-erance to therapy that a substantial proportion ofbreast cancer survivors, including those treated withadjuvant AI therapy, develop. These studies mightthen lead to better interventions designed to increasedpersistence with these life-saving medications. In thisregard, we recently initiated a large placebo-controlled trial of duloxetine for patients with AIMSS,which should shed further light on this frequent andproblematic toxicity.

Supplementary DataSupplementary data related to this article can be

found at http://dx.doi.org/10.1016/j.jpain.2014.01.487.

References

1. Briot K, Tubiana-Hulin M, Bastit L, Kloos I, Roux C: Effectof a switch of aromatase inhibitors on musculoskeletalsymptoms in postmenopausal women with hormone-receptor-positive breast cancer: The ATOLL (Articular Toler-ance of Letrozole) study. Breast Cancer Res Treat 120:127-134, 2010

2. Burstein HJ, Prestrud AA, Seidenfeld J, Anderson H,Buchholz TA, Davidson NE, Gelmon KE, Giordano SH,Hudis CA, Malin J, Mamounas EP, Rowden D, Solky AJ,Sowers MR, Stearns V, Winer EP, Somerfield MR, Griggs JJ:American Society of Clinical Oncology clinical practiceguideline: Update on adjuvant endocrine therapy forwomen with hormone receptor-positive breast cancer. JClin Oncol 28:3784-3796, 2010

3. Crew KD, Greenlee H, Capodice J, Raptis G, Brafman L,Fuentes D, Sierra A, Hershman DL: Prevalence of joint symp-toms in postmenopausal women taking aromatase inhibi-tors for early-stage breast cancer. J Clin Oncol 25:3877-3883, 2007

4. Dixon JM, Renshaw L, Young O, Murray J, Macaskill EJ,McHugh M, Folkerd E, Cameron DA, A’Hern RP,Dowsett M: Letrozole suppresses plasma estradiol andestrone sulphate more completely than anastrozole in post-menopausal women with breast cancer. J Clin Oncol 26:1671-1676, 2008

5. Dowsett M, Cuzick J, Ingle J, Coates A, Forbes J, Bliss J,Buyse M, Baum M, Buzdar A, Colleoni M, Coombes C,Snowdon C, Gnant M, Jakesz R, Kaufmann M, Boccardo F,

Godwin J, Davies C, Peto R: Meta-analysis of breast canceroutcomes in adjuvant trials of aromatase inhibitors versustamoxifen. J Clin Oncol 28:509-518, 2010

6. Felson DT, Cummings SR: Aromatase inhibitors and thesyndrome of arthralgias with estrogen deprivation. ArthritisRheum 52:2594-2598, 2005

7. Foerster BR, Petrou M, Harris RE, Barker PB, Hoeffner EG,Clauw DJ, Sundgren PC: Cerebral blood flow alterations inpain-processing regions of patients with fibromyalgia usingperfusion MR imaging. AJNR Am J Neuroradiol 32:1873-1878, 2011

8. Geisler J, Haynes B, Anker G, Dowsett M, Lonning PE: In-fluence of letrozole and anastrozole on total body aromati-zation and plasma estrogen levels in postmenopausal breastcancer patients evaluated in a randomized, cross-over study.J Clin Oncol 20:751-757, 2002

9. Geisser ME, Glass JM, Rajcevska LD, Clauw DJ,Williams DA, Kileny PR, Gracely RH: A psychophysicalstudy of auditory and pressure sensitivity in patientswith fibromyalgia and healthy controls. J Pain 9:417-422,2008

10. Geisser ME, Gracely RH, Giesecke T, Petzke FW,Williams DA, Clauw DJ: The association between experi-mental and clinical pain measures among persons with fi-bromyalgia and chronic fatigue syndrome. Eur J Pain 11:202-207, 2007

11. Giesecke J, Reed BD, Haefner HK, Giesecke T, Clauw DJ,Gracely RH: Quantitative sensory testing in vulvodynia pa-tients and increased peripheral pressure pain sensitivity. Ob-stet Gynecol 104:126-133, 2004

8 The Journal of Pain Effect of Estrogen Depletion on Pain Sensitivity

12. Giesecke T, Gracely RH, Grant MA, Nachemson A,Petzke F, Williams DA, Clauw DJ: Evidence of augmentedcentral pain processing in idiopathic chronic low backpain. Arthritis Rheum 50:613-623, 2004

13. Giesecke T, Gracely RH, Williams DA, Geisser ME,Petzke FW, Clauw DJ: The relationship between depression,clinical pain, and experimental pain in a chronic pain cohort.Arthritis Rheum 52:1577-1584, 2005

14. Giesecke T, Williams DA, Harris RE, Cupps TR, Tian X,Tian TX, Gracely RH, Clauw DJ: Subgrouping of fibromyalgiapatients on the basis of pressure-pain thresholds and psy-chological factors. Arthritis Rheum 48:2916-2922, 2003

15. Gracely RH, Lota L, Walter DJ, Dubner R: A multiplerandom staircase method of psychophysical pain assess-ment. Pain 32:55-63, 1988

16. Hassett AL, Epel E, Clauw DJ, Harris RE, Harte SE,Kairys A, Buyske S, Williams DA: Pain is associated with shortleukocyte telomere length in women with fibromyalgia. JPain 13:959-969, 2012

17. Hays RD, Stewart AL: Sleep measures, in Stewart AL,Ware JE (eds): Measuring Functioning and Well-Being: TheMedical Outcomes Study Approach. Durham, NC, Duke Uni-versity Press, 1992, pp 235-259

18. Hellstrom B, Anderberg UM: Pain perception across themenstrual cycle phases in women with chronic pain. PerceptMot Skills 96:201-211, 2003

19. Henry NL, Azzouz F, Desta Z, Li L, Nguyen AT, Lemler S,Hayden J, Tarpinian K, Yakim E, Flockhart DA, Stearns V,Hayes DF, Storniolo AM: Predictors of aromatase inhibitordiscontinuation due to treatment-emergent symptoms inearly-stage breast cancer. J Clin Oncol 30:936-942, 2012

20. Henry NL, Banerjee M, Wicha M, Van Poznak C,Smerage JB, Schott AF, Griggs JJ, Hayes DF: Pilot study of du-loxetine for treatment of aromatase inhibitor-associatedmusculoskeletal symptoms. Cancer 117:5469-5475, 2011

21. Henry NL, Giles JT, Stearns V: Aromatase inhibitor-associated musculoskeletal symptoms: Etiology and strate-gies for management. Oncology 22:1401-1408, 2008

22. Henry NL, Skaar TC, Dantzer J, Li L, Kidwell K, Gersch C,Nguyen AT, Rae JM, Desta Z, Oesterreich S, Philips S,Carpenter JS, Storniolo AM, Stearns V, Hayes DF,Flockhart DA: Genetic associations with toxicity-relateddiscontinuation of aromatase inhibitor therapy for breastcancer. Breast Cancer Res Treat 138:807-816, 2013

23. LeResche L,Mancl L, Sherman JJ, Gandara B, Dworkin SF:Changes in temporomandibular pain and other symptomsacross the menstrual cycle. Pain 106:253-261, 2003

24. Napadow V, LaCount L, Park K, As-Sanie S, Clauw DJ,Harris RE: Intrinsic brain connectivity in fibromyalgia is asso-ciated with chronic pain intensity. Arthritis Rheum 62:2545-2555, 2010

25. Napadow V, Lee J, Kim J, Cina S, Maeda Y, Barbieri R,Harris RE, Kettner N, Park K: Brain correlates of phasic auto-nomic response to acupuncture stimulation: An event-related fMRI study. Hum Brain Mapp 34:2592-2606, 2013

26. Petzke F, Clauw DJ, Ambrose K, Khine A, Gracely RH:Increased pain sensitivity in fibromyalgia: Effects of stimulustype and mode of presentation. Pain 105:403-413, 2003

27. Petzke F, Harris RE, Williams DA, Clauw DJ, Gracely RH:Differences in unpleasantness induced by experimentalpressure pain between patients with fibromyalgia andhealthy controls. Eur J Pain 9:325-335, 2005

28. Phillips K, Clauw DJ: Central pain mechanisms in therheumatic diseases: Future directions. Arthritis Rheum 65:291-302, 2013

29. Radloff LS: The CES-D scale: A self-report depressionscale for research in the general population. J Appl PsycholMeas 1:385-401, 1977

30. Santen RJ, Demers L, Ohorodnik S, Settlage J,Langecker P, Blanchett D, Goss PE, Wang S: Superiority ofgas chromatography/tandem mass spectrometry assay (GC/MS/MS) for estradiol for monitoring of aromatase inhibitortherapy. Steroids 72:666-671, 2007

31. Seidenberg M, Haltiner A, Taylor MA, Hermann BB,Wyler A: Development and validation of a Multiple AbilitySelf-Report Questionnaire. J Clin Exp Neuropsychol 16:93-104, 1994

32. Sestak I, Cuzick J, Sapunar F, Eastell R, Forbes JF,Bianco AR, Buzdar AU: Risk factors for joint symptoms in pa-tients enrolled in the ATAC trial: A retrospective, exploratoryanalysis. Lancet Oncol 9:866-872, 2008

33. Smets EM, Garssen B, Bonke B, De Haes JC: The Multidi-mensional Fatigue Inventory (MFI) psychometric qualities ofan instrument to assess fatigue. J Psychosom Res 39:315-325,1995

34. Smith EM, Pang H, Cirrincione C, Fleishman S, Paskett ED,Ahles T, Bressler LR, Fadul CE, Knox C, Le-Lindqwister N,Gilman PB, Shapiro CL: Effect of duloxetine onpain, function,and quality of life among patients with chemotherapy-induced painful peripheral neuropathy: A randomized clin-ical trial. J AmMed Assoc 309:1359-1367, 2013

35. Smith YR, Stohler CS, Nichols TE, Bueller JA, Koeppe RA,Zubieta JK: Pronociceptive and antinociceptive effects ofestradiol through endogenous opioid neurotransmissionin women. J Neurosci 26:5777-5785, 2006

36. Spritzer KL, Hays RD: MOS Sleep Scale: A Manual for Useand Scoring, Version 1.0. Los Angeles, CA, Rand, 2003

37. Tracey I: Can neuroimaging studies identify pain endo-phenotypes in humans? Nat Rev Neurol 7:173-181, 2011

38. Woolf CJ: Central sensitization: Implications for thediagnosis and treatment of pain. Pain 152:S2-S15, 2011

39. Yarnitsky D, Crispel Y, Eisenberg E, Granovsky Y, Ben-Nun A, Sprecher E, Best LA, Granot M: Prediction of chronicpost-operative pain: Pre-operative DNIC testing identifiespatients at risk. Pain 138:22-28, 2008

40. Yarnitsky D, Granot M, Nahman-Averbuch H,Khamaisi M, Granovsky Y: Conditioned pain modulationpredicts duloxetine efficacy in painful diabetic neuropathy.Pain 153:1193-1198, 2012