Neck Pain and Cervical Dystonia: Treatment Outcomes from CD PROBE (Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy)

Transcription

1 ORIGINAL ARTICLE Neck Pain and Cervical Dystonia: Treatment Outcomes from CD PROBE (Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy) P. David Charles, MD*; Aubrey Manack Adams, PhD ; Thomas Davis, MD*; Kathryn Bradley, MD*; Marc Schwartz, BS ; Mitchell F. Brin, MD, ; Atul T. Patel, MD *Vanderbilt University Medical Center, Nashville, Tennessee; Allergan, Inc., Irvine, California; MedNet Solutions, Inc., Minnetonka, Minnesota; University of California, Irvine, California; Kansas City Bone and Joint Clinic, Overland Park, Kansas U.S.A. & Abstract Background: Pain is a prevailing feature of cervical dystonia (CD), the most common form of focal dystonia. This analysis examined pain relief after onabotulinumtoxina treatment in CD subjects with moderate/severe pain from the Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy (CD PROBE). Methods: CD PROBE was a prospective, multicenter, observational registry of CD subjects who were na ıve to botulinum toxin (BoNT), new to physician, or had not received BoNT within 16 weeks if in a clinical trial. Subjects were eligible for 3 treatments, with variable session intervals. Descriptive and inferential statistics were utilized to evaluate the change in Address correspondence and reprint requests to: P. David Charles, MD, Vanderbilt University Medical Center, Department of Neurology, st Avenue South, Suite A-16 MCN, Nashville, TN 37232, U.S.A. david.charles@vanderbilt.edu. Submitted: April 4, 21; Revision accepted: September, 21 DOI..1111/papr The Authors. Pain Practice published by Wiley Periodicals, Inc. on behalf of World Institute of Pain, This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made., 13-78/16/$1. Pain Practice, Volume 16, Issue 8, pain scores in the population with moderate/severe neck pain at baseline (Pain Numeric Rating Scale [PNRS] score 4 to ). Results: Of 46 enrolled, 733 (7.7%) had moderate/severe neck pain at baseline. Postinjection pain questionnaire responses 4 to 6 weeks after each of the 3 treatments revealed that a majority of subjects (67.1%, 72.4%, and 76.4%) reported pain relief; mean time to pain relief was 7.1, 7.4, and 7.6 days. All pain scales showed significant improvements from baseline to final visit (all P <.1): PNRS, mean 6.6 to 3.8; CD Impact Profile-8 Pain and Discomfort subscale, mean 78.7 to 6.; and Toronto Western Spasmodic Torticollis Rating Scale Pain subscale, mean 12.6 to 8.. Multivariable regression models showed that initial pain score significantly contributed to the final pain score for all scales. Conclusion: Results from this real-world clinical registry indicate that a majority of CD subjects with moderate/ severe neck pain experience significant relief following onabotulinumtoxina treatment. & Key Words: botulinum toxin, cervical dystonia, pain, spasmodic torticollis INTRODUCTION Dystonia is a movement disorder characterized by involuntary, sustained, or intermittent muscle contractions that result in aberrant postures and/or

2 74 CHARLES ET AL. movements. 1,2 Cervical dystonia (CD) is the most common form of focal dystonia in adults and is associated with an abnormal head position due to involuntary contraction of muscles in the neck and upper thorax. 1,2 The prevalence of CD is estimated at 28 to 183 cases per million, with about twice as many women as men affected. 3 Pain in the cervical and upper thoracic area is one of the most prominent and disabling features of CD and is present in a vast majority of patients. 4, The diagnosis of CD is typically made by a neurologist following a careful differentiation of the associated pain and postural abnormalities from other potential orthopedic and myogenic disorders. 1,6,7 When CD is considered in the differential diagnosis, it is particularly important that an appropriate referral is made to a movement disorder specialist or other physician experienced in the treatment of CD so that appropriate treatment may be initiated to address all facets of the condition, including pain. Botulinum toxin (BoNT) injection of selected muscles is the treatment of choice for CD and has been shown to improve muscle control, increase range of motion, and relieve pain in patients diagnosed with CD. 8 CD PROBE was conducted to expand our knowledge on the characteristics of subjects with CD, as well as generate data on treatment outcomes in actual clinical practice, including physician- and subject-reported outcomes associated with pain. To date, it represents the largest clinical study of CD conducted. 9 Baseline analyses in CD PROBE found that pain associated with CD negatively impacted quality of life, and a higher percentage of subjects with moderate-to-severe pain reported being disabled and were more likely to have stopped work due to their CD. Clinical trials of CD have demonstrated significant pain relief following treatment with onabotulinumtoxina. 11,12 Other studies have further supported the use of BoNT for relief of pain due to CD, but the few long-term trials did not test onabotulinumtoxina; most involved small patient numbers and/or were not prospective Herein, we report clinical outcomes related to pain relief in subjects treated with onabotulinumtoxina while participating in CD PROBE. METHODS Study Design and Subjects CD PROBE is an open-label, prospective, multicenter clinical registry for subjects with CD treated with onabotulinumtoxina (ClinicalTrials.gov: identifier NCT83617). 9 The goal of the study was to describe the patterns of onabotulinumtoxina treatment within this rare disease, so the planned sample size was the number of subjects who could be reasonably recruited within the time frame of the trial. Eighty-five centers with 88 physicians specializing in neurology, physical medicine and rehabilitation, or pain enrolled 1,46 CD subjects. A detailed description of the methods of CD PROBE has been previously published. 9 In brief, subjects with CD were eligible for enrollment if they were new to the physician s practice, were new to BoNT treatment, or had not received BoNT for 16 weeks if in a clinical trial testing a BoNT. Following the provision of informed consent, subjects could enroll in CD PROBE and receive up to 3 treatments, with followup assessments of pain conducted during physician visits and/or via telephone (Figure 1). Dosing and the muscles injected were at the full discretion of the treating physician. The duration of time between injections was determined by the physician and subject, based upon clinical necessity; thus, the intervals between assessments were also variable. All patients provided Baseline Visit 1 (office) Injection Weeks Injection 1 Follow-up Patient phone interview Visit 2 (office) Injection 2 Duration determined by clinician 4 6 Weeks Injection 2 Follow-up Patient phone interview Duration determined by clinician Visit 3 (office) Injection Weeks Final Visit 4 (office) Injection 3 Follow-up TWSTRS CDIP-8 CDIP-8 Postinjection pain CDIP-8 CDIP-8 Postinjection pain TWSTRS CDIP-8 TWSTRS CDIP-8 Postinjection pain Figure 1. Study design. CDIP-8, Cervical Dystonia Impact Profile; PNRS, Pain Numeric Rating Scale; TWSTRS, Toronto Western Spasmodic Torticollis Rating Scale. Figure is adapted with permission from Jankovic J, Adler CH, Charles PD, et al. Rationale and design of a prospective study: Cervical Dystonia Patient Registry for Observation of OnaBotulinumtoxinA Efficacy (CD PROBE). BMC Neurol. 211;11:14.

3 Neck Pain and Cervical Dystonia 7 written informed consent, and each investigator obtained institutional review board/ethics committee approval. Outcome Measures Pain outcomes were measured with the following instruments: Pain Numeric Rating Scale (PNRS; assessed at all visits and phone interviews), 16 Pain subscale of the Toronto Western Spasmodic Torticollis Rating Scale 17 (TWSTRS; measured at baseline and visits 3 and 4), Pain and Discomfort subscale of the Cervical Dystonia Impact Profile 18 (CDIP-8, assessed at all visits and phone interviews), and a postinjection pain questionnaire (measured at phone interviews 1 and 2 and visit 4, with each referring to the immediately prior treatment). This questionnaire included the following: (1) Did your CD cause you to feel neck pain or discomfort prior to your BoNT treatment?; (2) Was your neck pain or discomfort relieved during the first 2 weeks after your BoNT treatment?; and (3) If yes, about how many days did it take for you to feel this relief of neck pain or discomfort? The TWSTRS and CDIP-8 are both disease-specific, validated rating scales, while the PNRS is a recommended singleitem questionnaire that asks the subject to rank pain from ( no pain ) to ( pain as bad as you can imagine ), 16 modified to neck pain for this study. The TWSTRS Pain subscale ranges from to 2 17 and the CDIP-8 Pain and Discomfort subscale from to, 18 with higher scores indicating worse conditions for both. The postinjection pain questionnaire was created for this study. Statistical Analysis This analysis includes the as-treated population of subjects who reported whether they had previously received BoNT at baseline and received the initial injection. Data are presented for those who completed all assessments for a given outcome measure to avoid imputation of missing data. Subjects available for analysis were then further divided based on their neck pain reported at baseline as measured by the PNRS; those with moderate to-severe pain reported a PNRS score of 4 to. 22,23 Each of the above pain assessments Enrolled N = 46 Baseline (Visit 1) Injection 1 n = 41 Visit 2 Injection 2 n = 84 Visit 3 Injection 3 n = 636 Final Visit (Visit 4) Injection 3 Follow-up Completed n = 2 Discontinued 1 Lost to follow-up 2 Withdrew consent 2 Other 168 Discontinued (2.8%) 6 Adverse event 1 Death Financial 34 Lack of response 6 Lost to follow-up 6 Moved Non-study injector 8 Physician discretion 1 Pregnancy 4 Recovered/remission 1 Time burden 1 Travel burden 34 Withdrew consent 2 Other 237 Discontinued (22.8%) 26 Adverse event 2 Death Financial 49 Lack of response 72 Lost to follow-up 2 Moved 2 Non-study injector Other health issues 12 Physician discretion 1 Pregnancy 2 Recovered/remission 1 Time burden 1 Travel burden 2 Withdrew consent Other 134 Discontinued (21.1%) 1 Death 1 Financial 2 Lack of response 1 Lost to follow-up 2 Moved 1 Physician discretion 1 Pregnancy 3 Time burden Travel burden 7 Withdrew consent 1 Other Figure 2. Subject disposition. Reprinted from Jankovic J, Adler CH, Charles PD, et al. Primary results from the Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy (CD PROBE). J Neurol Sci. 21; 349:84 93 with permission from Elsevier.

4 76 CHARLES ET AL. was evaluated for the group reporting moderateto-severe pain at baseline because those with little to no pain at baseline would likely demonstrate limited change after treatment. The above pain measures were analyzed with descriptive and inferential statistics, including analysis of variance when appropriate. Reported statistical comparisons and associated P-values were adjusted for multiple comparisons. 24 Multivariable regression models were used to assess selected contributing factors to the changes in pain measures from their initial to final assessments. Least squares linear regression was used for continuous response measures, while logistic regression was used for dichotomous measures. In each model, the dependent variable was the final pain-related response, while prespecified independent variables included the initial pain-related measure, age, gender, cumulative units of onabotulinumtoxina, and physician-assessed severity at baseline. Continuous initial pain, age, and cumulative units of onabotulinumtoxina were modeled using cubic regression splines to allow for the possibility of curvilinear relationships to the final pain-related measure. RESULTS CD PROBE investigators enrolled 46 subjects, of which 733 (7.7%) had moderate-to-severe neck pain at baseline (PNRS, 4 to ) (Figure 2). The subgroup reporting moderate-to-severe pain at baseline was younger when compared with those with little to no pain ( vs years; P <.1) (Table 1). Subjects with moderate-to-severe pain at baseline were also rated as more severe by their physician and more commonly presented with an abnormal head position that included laterocollis or anterocollis (Table 1). Over 9% of subjects reported neck pain or discomfort prior to each onabotulinumtoxina treatment (Figure 3A). Subjects reported pain relief after each treatment on the postinjection pain questionnaire, ranging from 67.1% after the first treatment to 76.4% after the third treatment (P =.12; Figure 3B). For those reporting an improvement in pain, the average time to pain relief ranged from to days following the first and third treatments, respectively (Figure 3C). For subjects reporting moderate-to-severe neck pain at baseline, the mean TWSTRS Pain subscale significantly improved from at baseline to at visit 3 and at the final visit (P <.1 for each visit compared with baseline) Table 1. Subject Demographic and Disease Characteristics: Overall and By Pain Status at Baseline Moderate/ Total No/Mild Pain Severe Pain (N = 37) (n = 34) (n = 733) P-Value Age (years) <.1 Gender Female 772 (74.4) 226 (74.3) 46 (74.).968 Race/Ethnicity White 99 (92.) 28 (93.8) 674 (92.).2832 Nonwhite* 78 (7.) 19 (6.3) 9 (8.) BMI (kg/m 2 ) Data not available Severity.376 Mild 344 (33.2) 111 (36.) 233 (31.8) Moderate 46 (2.7) 161 (3.) 38 (2.6) Severe 146 (14.1) 32 (.) 114 (1.6) Data not 1 1 available CD type.1 Anterocollis 9 (.7) 13 (4.3) 46 (6.3) Laterocollis 42 (38.8) 3 (33.9) 299 (4.8) Retrocollis (.3) 12 (3.9) 43 (.9) Torticollis 493 (47.6) 164 (3.9) 329 (44.9) Other 27 (2.6) 12 (3.9) 1 (2.) Data not 1 1 available Age at symptom onset (years) Time from CD onset to diagnosis (years) Time from CD diagnosis to treatment (years) Previously received BoNT treatment 378 (36.) 7 (3.2) 271 (37.).89 BMI, body mass index; BoNT, botulinum toxin; CD, cervical dystonia. Data are presented as mean standard deviation or n (%). Pain is defined by baseline score on the Pain Numeric Rating Scale: to 3 for no/mild pain and 4 to for moderate/severe pain. *Includes Asian, Black, Hispanic, Native American, and Other. Table is adapted with permission from Charles PD, Adler CH, Stacy M, et al. Cervical dystonia and pain: characteristics and treatment patterns from CD PROBE (Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy). J Neurol. 214;261: (Figure 4). Similarly, PNRS scores significantly improved from at baseline to upon the final assessment (P <.1) (Figure ). The Pain and Discomfort subscale of the CDIP-8 demonstrated similar improvement from at baseline to upon the final assessment (P <.1) (Figure 6). Multivariable regression models were generated to test the effects of the initial pain outcome score, age, gender, physician-assessed severity, and cumulative onabotulinumtoxina dose on final pain outcomes. For all pain measures, subjects initial pain score were significant predictors of the final score (P.; Table 2). For the TWSTRS Pain subscale and the PNRS,

5 Neck Pain and Cervical Dystonia 77 A B Subjects (%) Interview 1 Interview 2 Final Visit (Visit 4) Subjects (%) Interview 1 Interview 2 * Final Visit (Visit 4) C 16 Time to Relief (Days) Interview 1 Interview 2 Final Visit (Visit 4) Figure 3. Postinjection pain questionnaire in subjects with moderate/severe pain at baseline. (A) Percentage of subjects whose CD caused neck pain or discomfort prior to onabotulinumtoxina treatment. (B) Percentage of subjects whose neck pain or discomfort was relieved during the first 2 weeks after onabotulinumtoxina treatment. (C) Time to relief of neck pain or discomfort. CD, cervical dystonia; dot = mean; line in box = median; top & bottom of box = interquartile range; whiskers = minimum & maximum. *P <. vs. phone interview 1. Includes only subjects with baseline Pain Numeric Rating Scale score of 4 to. n = 33 in panel A, n = 31 in panel B, and n = 149 in panel C for all assessments, all with a baseline PNRS score of 4 to. TWSTRS Score 2 1 Baseline (Visit 1) Final Visit (Visit 4) age also was a significant predictor of the final score, while gender and physician-assessed severity were not significant predictors for any of the pain outcome * Visit 3 Figure 4. TWSTRS Pain subscale scores in subjects with moderate/ severe pain at baseline. Dot = mean; line in box = median; top & bottom of box = interquartile range; whiskers = minimum & maximum; TWSTRS, Toronto Western Spasmodic Torticollis Rating Scale. *P <.1 vs. baseline. Includes only subjects with baseline Pain Numeric Rating Scale score of 4 to. Scale ranges from to 2. n = 341 for all visits, all with a baseline PNRS score of 4 to. * PNRS Score Baseline (Visit 1) * * * * * Interview 1 Visit 2 Visit 3 Final Visit Interview 2 (Visit 4) Figure. Pain Numeric Rating Scale (PNRS) scores in subjects with moderate/severe pain at baseline. *P <.1 vs. baseline. Includes only subjects with baseline PNRS score of 4 to. Scale ranges from to. Data are shown as mean 9% confidence interval. n = 329 for all visits, all with a baseline PNRS score of 4 to. measures. As seen in Figure 7A, higher initial TWSTRS Pain scores contributed to higher final TWSTRS Pain scores, and age did not contribute to the final TWSTRS Pain score until about age, after which there was a slight trend in declining final pain score until about 7,

6 78 CHARLES ET AL. CIDP-8 Score Baseline (Visit 1) * Interview 1 * * * * Visit 2 Visit 3 Final Visit Interview 2 (Visit 4) Figure 6. CDIP-8 Pain and Discomfort score in subjects with moderate/severe pain at baseline. CDIP-8, Cervical Dystonia Impact Profile. *P <.1 vs. baseline. Scale ranges from to. Includes only subjects with baseline Pain Numeric Rating Scale score of 4 to. Data are shown as mean 9% confidence interval. n = 326 for all visits, all with a baseline PNRS score of 4 to. and then, the trend leveled off. Higher initial PNRS scores contributed to higher final PNRS scores until an initial PNRS score of about 8, after which the trend leveled off and then decreased (Figure 7B). Increasing age contributed to a slightly higher (about 1 point) final PNRS score until about age 4, ages 4 to 6 contributed to a slightly lower final PNRS score, and ages > 6 did not appear to contribute to final PNRS score. For the CDIP-8, a higher initial score contributed to a higher final score in a linear fashion (Figure 7C), with a similar trend seen for the time to relief of neck pain (Figure 7D). DISCUSSION CD PROBE is the largest prospective clinical trial conducted for subjects with CD treated with BoNT. Outcomes of this real-world subject registry demonstrate that in a practice setting, physicians of differing medical specialties achieve significant pain relief for their patients who had moderate-to-severe pain at baseline. Even though the TWSTRS does not have a defined minimal clinically important difference for improvement, registration trials for BoNTs have used an improvement of 3% and/or points in the TWSTRS Total score to represent a clinical response. 1,2 TWSTRS Pain results for subjects with moderate-to-severe pain in CD PROBE met this criterion, with a mean change from baseline of 32.%. This exceeds or is comparable to the change in TWSTRS Pain observed in randomized controlled trials of onabotulinumtoxina, although dosing and timing of assessments differed from those of CD PROBE Table 2. Multivariable Regression Models: Impact of Initial Pain Outcomes, Age, Gender, Physician-Assessed Severity, and Cumulative OnabotulinumtoxinA Dose on Final Pain Outcomes P-value Final TWSTRS Pain score Initial TWSTRS Pain score <.1 Age.3 Gender.426 Physician-assessed severity.976 Cumulative dose.237 Final PNRS score Initial PNRS score.4 Age.119 Gender.642 Physician-assessed severity.6884 Cumulative dose.79 Final CDIP-8 Pain and Discomfort score Initial CDIP-8 Pain and Discomfort score <.1 Age.88 Gender.4649 Physician-assessed severity.9992 Cumulative dose.64 Postinjection pain questionnaire Neck pain/discomfort prior to final treatment session Neck pain/discomfort prior to initial treatment session.49 Age.6494 Gender.633 Physician-assessed severity.6423 Cumulative dose.227 Neck pain/discomfort relieved within 2 weeks of final treatment session Neck pain/discomfort relieved within <.1 2 weeks of initial treatment session Age.483 Gender.788 Physician-assessed severity.819 Cumulative dose.248 Time to neck pain/discomfort relief after final treatment session Time to neck pain/discomfort relief.4 after initial treatment session Age.4891 Gender.761 Physician-assessed severity.8367 Cumulative dose.371 CDIP-8, Cervical Dystonia Impact Profile; PNRS, Pain Numeric Rating Scale; TWSTRS, Toronto Western Spasmodic Torticollis Rating Scale. While these findings were similar to those from clinical trials, CD PROBE also expands the body of knowledge by providing results from clinical practice settings over a longer duration of time. Dosing and injection schema in this study were at the physician s discretion, not according to a defined protocol, further showing the effectiveness of onabotulinumtoxina in a real-world clinical setting. CD PROBE also included a wider range of subjects than the fairly homogeneous populations enrolled in clinical trials, as well as a greater number of outcome measures. Indeed, results from the postinjection pain questionnaire, PNRS, and Pain and Discomfort subscale of the CDIP-8 were consistent with the

7 Neck Pain and Cervical Dystonia 79 A TWSTRS Pain B PNRS 8 Final TWSTRS Pain 1 Final PNRS Initial TWSTRS Pain Initial PNRS 8 Final TWSTRS Pain 1 Final PNRS Age (Years) Age (Years) C CDIP-8 Pain and Discomfort D Time to Relief of Neck Pain or Discomfort Final CDIP-8 Pain and Discomfort Final Time to Relief Initial CDIP-8 Pain and Discomfort Initial Time to Relief Figure 7. Multivariable regression models to assess the contributions to the final scores in subjects with moderate/severe pain at baseline for (A) TWSTRS Pain, (B) PNRS, (C) CDIP-8 Pain and Discomfort, and (D) postinjection pain questionnaire assessment of time to relief. Fitted regression lines (solid) and 9% confidence intervals (dotted). Includes only subjects with baseline PNRS score of 4 to. For the value that varies (horizontal axis), the other values are held constant where appropriate: gender = male; severity = moderate; median age = 7; median cumulative onabotulinumtoxina dose = 9; median initial TWSTRS Pain = 13 (panel A); median initial PNRS = 7 (panel B); median initial CDIP-8 Pain & Discomfort = 8 (panel C). CDIP-8, Cervical Dystonia Impact Profile; PNRS, Pain Numeric Rating Scale; TWSTRS, Toronto Western Spasmodic Torticollis Rating Scale.

8 8 CHARLES ET AL. TWSTRS Pain subscale findings of a significant decrease in pain following treatment with onabotulinumtoxina. Finally, although there are some differences in the slopes and curvature of the lines, which may be due to differences among the subject- and physician-reported outcomes, the regression analyses indicate that subjects with higher initial pain scores on the TWSTRS, PNRS, and CDIP-8 show a greater improvement in pain than those with lower initial scores. The results of the postinjection pain questionnaire indicate that pain does return prior to the next treatment in the majority of subjects; this suggests that pain may be one of the driving factors for subjects to seek retreatment. The questionnaire does not ask whether the returning pain is less severe than the pain prior to the previous treatment; however, the PNRS and CDIP-8 Pain and Discomfort results indicate that pain does not return to baseline levels at the time of the next treatment. In addition to examining the outcomes in the population with moderate-to-severe neck pain at baseline (PNRS, 4 to ), we also analyzed outcomes in the subpopulation who were na ıve to BoNT at baseline, as well as in those with any neck pain at baseline (PNRS, 1 to ) in the overall and na ıve populations. As expected, pain scores were slightly lower in the groups with any pain compared with those with moderate-to-severe pain, but the outcomes exhibited similar and significant trends. Subjects with moderate-to-severe pain who were na ıve to toxin at baseline exhibited nearly identical pain outcome scores as the overall population with moderate-to-severe pain. This finding may be explained by the inclusion criteria, in that subjects who have been previously treated with BoNT but are new to a physician s practice are often treated as though they are na ıve to BoNT. OnabotulinumtoxinA can ameliorate pain by disrupting motor and sensory mechanisms, both of which require intracellular vesicular trafficking. In the former, inhibition of acetylcholine release decreases excessive contractions and likely the spasms that contribute to pain in joints and tendons. 29 In the latter, onabotulinumtoxina blocks neuropeptide release from nociceptive terminals of primary afferent nerve fibers (eg, CGRP, substance P) 3 32 and downregulates surface expression of pain receptors and ion channels (eg, TRPV1, TRPA1) on nerve membranes. 3,33 Consistent with a primary effect on pain, a previous study showed that treatment of CD subjects with onabotulinumtoxina led to significant pain relief prior to improvement in muscle relaxation, indicating that the antinociceptive and muscle relaxation mechanisms can be distinct. 34 Prior work reported that subjects with CD frequently requested re-injection of onabotulinumtoxina when pain returned but before the return of their postural symptoms, and some patients had pain relief disproportionate to motor benefit. 3 A recent study in upper and lower limb poststroke spasticity specifically assessed an association between muscle tone and pain, and although pain was significantly improved, the correlation coefficient was weak (r =.146), supporting the notion that pain improvement was distinguished from improvement in muscle tone. 36 CD PROBE has both strengths and limitations. The frequency of discontinuation was relatively high, which reduced the sample size available for pain outcomerelated analyses. As CD PROBE is an observational, naturalistic study, rather than a randomized clinical trial, this is not unexpected. Registries generally have higher discontinuation rates, as they often enroll a broader subject population, have a longer study duration, and do not have a protocol-defined treatment schedule or provide compensation for treatment and travel. 37 For a detailed exploration of analyses to best understand the reasons for discontinuation (Figure 1) and the potential impact on the study conclusions, we refer to the reports detailing the methods and primary results. 9,38 The strength of CD PROBE is that it is a large clinical registry of prospectively followed subjects that provides data on the clinical nuances of treatment that are not generally obtainable from randomized controlled trials. CONCLUSION Pain associated with CD is one of the most common and disabling features of the condition. The pain relief offered by onabotulinumtoxina for subjects with CD is robust and can be achieved by physicians treating patients in a real-world setting. ACKNOWLEDGEMENTS Jennifer L. Giel, PhD, CMPP, of Evidence Scientific Solutions, Philadelphia, PA, provided medical editorial assistance that was funded by Allergan, Inc. The authors would like to thank all of the investigators in the CD PROBE Study Group for their dedication to this study. This study and its analysis were funded by Allergan, Inc., which was responsible for the design and conduct of the study, management and analysis of the data, and

9 Neck Pain and Cervical Dystonia 81 preparation of the manuscript. Data were collected by the investigators at the study centers and were monitored and analyzed by Allergan, Inc. All authors were involved in data analysis and interpretation, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors provided final approval to submit the manuscript. CONFLICT OF INTEREST Vanderbilt University receives income from grants and contracts with Allergan, Ipsen, Merz, and Medtronic for research or educational programs led by Dr Charles. Dr Charles receives income for Allergan, Concert, Ipsen, Merz, and Medtronic for education or consulting services. Dr Davis has no conflict of interests to report. Dr Bradley does not have any conflict of interest to disclose. Dr Patel has been an investigator and/or participated in trials for Allergan, Merz, Ipsen, Myoscience, and Pfizer; and he is on the speaker s bureau for Allergan, Inc. Mr Schwartz is an employee of MedNet Solutions, Inc., which was contracted by Allergan, Inc. to provide statistical support. Drs Manack Adams and Brin are employees of Allergan, Inc. and receive salary, stock, and stock options from Allergan, Inc. REFERENCES 1. Albanese A, Bhatia K, Bressman SB, et al. Phenomenology and classification of dystonia: a consensus update. Mov Disord. 213;28: Phukan J, Albanese A, Gasser T, Warner T. Primary dystonia and dystonia-plus syndromes: clinical characteristics, diagnosis, and pathogenesis. Lancet Neurol. 211;: Defazio G, Jankovic J, Giel JL, Papapetropoulos S. Descriptive epidemiology of cervical dystonia. Tremor Other Hyperkinet Mov (N Y). 213;3: Chan J, Brin MF, Fahn S. Idiopathic cervical dystonia: clinical characteristics. Mov Disord. 1991;6: Jankovic J, Leder S, Warner D, Schwartz K. Cervical dystonia: clinical findings and associated movement disorders. Neurology. 1991;41: Pilge H, Prodinger PM, B urklein D, Holzapfel BM, Lauen J. Nontraumatic subluxation of the atlanto-axial joint as rare form of acquired torticollis: diagnosis and clinical features of the Grisel s syndrome. Spine (Phila Pa 1976). 211;36:E747 E Jankovic J. Camptocormia, head drop and other bent spine syndromes: heterogeneous etiology and pathogenesis of Parkinsonian deformities. Mov Disord. 2;2: Comella CL, Jankovic J, Brin MF. Use of botulinum toxin type A in the treatment of cervical dystonia. Neurology. 2;(12 Suppl. ):S1 S Jankovic J, Adler CH, Charles PD, et al. Rationale and design of a prospective study: Cervical Dystonia Patient Registry for Observation of OnaBotulinumtoxinA Efficacy (CD PROBE). BMC Neurol. 211;11:14.. Charles PD, Adler CH, Stacy M, et al. Cervical dystonia and pain: characteristics and treatment patterns from CD PROBE (Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy). J Neurol. 214;261: Charles D, Brashear A, Hauser RA, Li HI, Boo LM, Brin MF.; for the CD 14 Study Group. Efficacy, tolerability, and immunogenicity of onabotulinumtoxina in a randomized, double-blind, placebo-controlled trial for cervical dystonia. Clin Neuropharmacol. 212;3: Greene P, Kang U, Fahn S, Brin M, Moskowitz C, Flaster E. Double-blind, placebo-controlled trial of botulinum toxin injections for the treatment of spasmodic torticollis. Neurology. 199;4: Evidente VG, Fernandez HH, LeDoux MS, et al. A randomized, double-blind study of repeated incobotulinumtoxina (Xeomin((R))) in cervical dystonia. J Neural Transm. 213;12: Jabbari B, Machado D. Treatment of refractory pain with botulinum toxins an evidence-based review. Pain Med. 211;12: Truong D, Duane DD, Jankovic J, et al. Efficacy and safety of botulinum type A toxin (Dysport) in cervical dystonia: results of the first US randomized, double-blind, placebo-controlled study. Mov Disord. 2;2: Dworkin RH, Turk DC, Farrar JT, et al. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain. 2;113: Consky E, Basinski A, Belle L, Ranawaya R, Lang AE. The Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS): assessment of validity and inter-rate reliability [abstract]. Neurology. 199;4(Suppl. 1): Cano SJ, Warner TT, Linacre JM, et al. Capturing the true burden of dystonia on patients: the Cervical Dystonia Impact Profile (CDIP-8). Neurology. 24;63: Albanese A, Sorbo FD, Comella C, et al. Dystonia rating scales: critique and recommendations. Mov Disord. 213;28: Comella CL, Stebbins GT, Goetz CG, Chmura TA, Bressman SB, Lang AE. Teaching tape for the motor section of the Toronto Western Spasmodic Torticollis Scale. Mov Disord. 1997;12: Cano SJ, Hobart JC, Edwards M, et al. CDIP-8 can measure the impact of botulinum toxin treatment in cervical dystonia. Neurology. 26;67: Fejer R, Jordan A, Hartvigsen J. Categorising the severity of neck pain: establishment of cut-points for use in clinical and epidemiological research. Pain. 2;119:

10 82 CHARLES ET AL. 23. Zelman DC, Dukes E, Brandenburg N, Bostrom A, Gore M. Identification of cut-points for mild, moderate and severe pain due to diabetic peripheral neuropathy. Pain. 2;11: Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 1988;7: Comella CL, Jankovic J, Truong DD, Hanschmann A, Grafe S.; on behalf of the U.S. XEOMIN Cervical Dystonia Study Group. Efficacy and safety of incobotulinumtoxina (NT 21, XEOMIN â, botulinum neurotoxin type A, without accessory proteins) in patients with cervical dystonia. J Neurol Sci. 211;38: Comella CL, Jankovic J, Shannon KM, et al. Comparison of botulinum toxin serotypes A and B for the treatment of cervical dystonia. Neurology. 2;6: Quagliato EM, Carelli EF, Viana MA. A prospective, randomized, double-blind study comparing the efficacy and safety of type a botulinum toxins botox and prosigne in the treatment of cervical dystonia. Clin Neuropharmacol. 2;33: Pappert EJ, Germanson T.; for The Myobloc/Neurobloc European Cervical Dystonia Study Group. Botulinum toxin type B vs. type A in toxin-na ıve patients with cervical dystonia: randomized, double-blind, noninferiority trial. Mov Disord. 28;23: Colhado OC, Boeing M, Ortega LB. Botulinum toxin in pain treatment. Rev Bras Anestesiol. 29;9: Aoki KR, Francis J. Updates on the antinociceptive mechanism hypothesis of botulinum toxin A. Parkinsonism Relat Disord. 211;17(Suppl. 1):S28 S da Silva LB, Kulas D, Karshenas A, et al. Time course analysis of the effects of botulinum neurotoxin type A on pain and vasomotor responses evoked by glutamate injection into human temporalis muscles. Toxins (Basel). 214;6: Pavone F, Luvisetto S. Botulinum neurotoxin for pain management: insights from animal models. Toxins (Basel). 2;2: Burstein R, Zhang X, Levy D, Aoki KR, Brin MF. Selective inhibition of meningeal nociceptors by botulinum neurotoxin type A: therapeutic implications for migraine and other pains. Cephalalgia. 214;34: Relja M, Telarovic S. Botulinum toxin type-a and pain responsiveness in cervical dystonia: a dose response study. Mov Disord. 2;2:S Brin MF, Fahn S, Moskowitz C, et al. Localized injections of botulinum toxin for the treatment of focal dystonia and hemifacial spasm. Mov Disord. 1987;2: Wissel J, Ganapathy V, Ma YJ, et al. OnabotulinumtoxinA improves pain in poststroke spasticity patients: findings from a randomized controlled trial. Toxicon. 21;93:S64 S Manack A, Turkel CC, Kaplowitz H. Role of epidemiological data within the drug development lifecycle: a chronic migraine case study. In: Lunet N, ed. Epidemiology Current Perspectives on Research and Practice. Rijeka, Croatia: InTech; 212: Jankovic J, Adler CH, Charles PD, et al. Primary results from the Cervical Dystonia Patient Registry for Observation of OnabotulinumtoxinA Efficacy (CD PROBE). J Neurol Sci. 21;349:84 93.