Sindhu Ramchandren, M.D., M.SAssistant Professor of Neurology

Wayne State University-Detroit Medical Center sramchan@med.wayne.edu

Outcome Measurements asClinical Biomarkers

• Prevalence: 1 in 2500 people

• Childhood/adolescent onset; high-arched feet, distal atrophy, areflexic

• Financial burden unclear; but, for example, cost of Diabetic PN alone: 4 to 10 billion dollars annually

• Adverse Impact on Quality of Life (QOL): pain, disability, depression

Despite promising animal studies, no therapy that is curative or disease modifying exists in CMT

Impact of Charcot Marie Tooth Disease (CMT)

Treatments in Neuropathy

Disease Curative Rx

Modifying Rx

Inherited Neuropathies

- - - -

Chemotherapy-Neuropathies (CIPN)

- - - -

Diabetic Neuropathy - -Stringent control of blood sugar

Inflammatory Neuropathies

- -Steroids, IVIG,

pheresis

119 patients in community setting

Barriers to Developing Neuropathy Therapies: Multiple Etiologies

Huang, 1981; George and Twomey, 1986; Dyck et al., 1981

• Axonal degeneration

a. Distal axonal

b. Wallerian

• Demyelination

ab

NB: Functional Disability is a result of conduction block (acutely in inflammatory neuropathies) or axonal loss (chronically, in almost all other neuropathies)

Barriers to Developing Neuropathy Therapies: Variable Pathophysiology

Small myelinated

fiber

Unmyelinatedfibers

Large myelinated

fibers

Barriers to Developing Neuropathy Therapies: Variable Anatomy

Clinical-Pathophysiologic Correlates

• Large fibers: “Negative symptoms”– Weakness, wasting,

poor balance, sensory loss, areflexia

• Small fibers: “Positive symptoms”– Allodynia,

hyperesthesia, cramps, tingling, pins and needles, burning, etc.

Identifying Barriers to Developing Neuropathy Therapies

• Not one disease; multiple etiologies

• Variable pathophysiology:– Axonal vs. demyelinating processes– Small vs. Large fiber involvement

• Given the above, what outcome measure should we choose for a neuropathy trial?– Need to better understand natural history of

the disease, in order to select an outcome measure that can improve in a 1-2 year trial

Outcome Measures and the FDA

• FDA requirement for Rx approval: improvement in a Clinically Meaningful Outcome Measure

• In the absence of clinically meaningful OMs (or if trial cost of gathering such OM data is prohibitive), a Surrogate Measure can be used.– A Biomarker, which is a potential surrogate

measure, is a clinical or laboratory measure of disease activity

• Selecting the best outcome measure for your clinical trial requires careful study of the disease itself.

“Why should I care what outcome measure is chosen for a clinical trial? If the drug works, the patient will be

cured, wouldn’t it be obvious?”

Chronic Inflammatory Demyelinating Polyradiculoneuropathy (CIDP)

• Prevalence of 0.8- 1.9 per 100,000 adults• Aggressive: progression or recurrent relapses >

8 weeks from symptom onset• Large fiber (vibration, proprioception) involved

more than small (pain, temperature)• NCS: Patchy demyelinating neuropathy with

conduction block and temporal dispersion early on; later: secondary axonal loss

• Unfortunately- not much natural history data collected on effect on outcome measures in CIDP

TRIAL OUTCOME MEASURE Cross -over placebo trial of high-dose IVIG in CIDP (1-2 courses 8 days apart; 7 patients)

- van Doorn et al., Neurology 1990

1 point Δ in Rankin score

RCT of High-dose IVIG vs. placebo in CIDP (single treatment over 5 days) - Vermeulen et al., JNNP 1993

1 point Δ in Rankin score (score of at least 3 for inclusion

Cross-over placebo trial of high-dose IVIG in Rx-naïve CIDP (2 courses over 8 weeks) - Hahn et al., Brain 1996

20 point Δ in Neurological disability score

RCT of IVIG vs. Prednisolone in CIDP - Hughes et al., Ann Neurol 2001

11 point Δ in ODS

RCT of IVIG vs. placebo- Mendell et al., Neurology 2001

Δ in Average Muscle Score (modified MRC)

IVIG for CIDP (ICE trial)- Hughes et al., Lancet Neurol 2008

Difference betwn-groups in proportion of INCAT responders vs. non-responders (1 point Δ in INCAT = response)

IVIG in CIDP Clinical Trials

Escalating Costs of Clinical Trials

References: Eisentein et al., Clinical Trials 2008; DiMasi and Grabowski, 2007

Now, ~ $1 billion spent by industry on R&D, for each new drug approved in US

‘Dying back’ or ‘length-dependent’ process– Longest nerves affected

first– Ends of nerves

preferentially affected

Typical course– Indolent: time course of

months to years– Gradual, symmetric– Symptoms:

Predominantly sensory

Diabetic Peripheral Neuropathy (DPN)

Figure from: Jerry R, et al. Diagnosis and Management of Peripheral Nerve Disorders. Contemporary Neurology Series, 59.

MildNormal

SevereModerate

Axonal Loss in Diabetic Neuropathy

Britland ST, et al. Pain. 1992 Mar;48(3):361-70.

0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1

0 50 100 150 200 250 300

1

0 50 100 150 200 250 300 0

.1

.2

.3

.4

.5

.6

.7

.8

.9

“Predominantly” small fiber neuropathy

Mixed small and large fiber sensory neuropathy

Sensorimotor neuropathy

Durations of neuropathy symptoms (months)

Fra

ctio

n o

f p

ati

en

ts

0 50 100 150 200 250 0

.1

.2

.3

.4

.5

.6

.7

.8

.9 1

300

Polydefkis M, et al. JAMA 2003;290:1371-1376

Hierarchy of Sensory Nerve Involvement

Past Clinical Trials- Diabetic Peripheral Neuropathy

• Natural history of DPN- vey slow progression• DPN Trials until the mid-90s may have been

inefficiently designed*:– Enrolled patients with mild and severe PN– Short trial duration– Looked for reversal of endpoints such as NCS

abnormalities that (a) did not have time to change and (b) were more likely to slow in progression than reverse

Need clinically meaningful end-points that can change within the allotted study time frame

- *Dyck PJ, et al. Diabetes Care 2007;30:2619–2625- *Ziegler D, et al. Int Rev Neurobiol 2002;50:431–463- *Carrington AL, et al. Diabetes Care 2001;25:2010–2015

Diabetic Neuropathy Trials: Future

• Validated Pain Scales have taken over as the patient reported outcome measure of choice

• Need to identify Clinical Outcome Measures– no consensus: measures used include NIS-LL, NCS…– Intraepidermal nerve fiber density is problematic:

• 20-50% of small fiber neuropathies are idiopathic• Clinical Exam not very sensitive or specific• Processes other than neuropathy cause small fiber

loss• IENF density has not been shown to improve clinical

management or patient health outcome– New research: early changes in axon excitability due

to ion channel dysfunction may precede long-term axonal loss (Sung et al; Diabetes 2012:1592-8)

• Natural history: variable, but for CMT1A, slow progression

• NCS in CMT1A: Symmetric demyelinating neuropathy, typically without conduction block or temporal dispersion

• Despite demyelination, disability in CMT1A correlates to secondary axonal loss (i.e. chronic denervation)

Charcot Marie Tooth Disease (CMT)

Demyelination and Secondary Axonal Loss

CHARCOT-MARIE-TOOTH 1A

Past Failures in CMT trials: Ineffective Drugs Vs. Trial Design Errors

HISTORICAL RATE OF CHANGE/TIME FOR OUTCOME MEASURE

Nerve Conduction VelocitiesNo change in 8 years(Sancho 1999; Sahenk 1999; Krajewski 2000; Shy 2008)

Charcot-Marie-Tooth Neuropathy Score (CMTNS)

1.4 points/2 years (CMT1A); 2.89 points/10 years (CMT1X) (Shy, 2008)

OUTCOME MEASURES USED IN CMT TRIALS

Myelinated fiber regeneration Sahenk et al., Neurology 2005

Myosin isoform changes Smith et al., Muscle Nerve 2006

2.5 m/sec/yr improvement in NCV Burns et al., Lancet Neurol 2009

10 m/sec improvement in NCV Verhamme et al., BMC Med 2009

5-point improvement in CMTNS Micallef et al., Lancet Neurol 2009

1.5 betwn-group in CMTNS in 2 yrs Pareyson et al., Lancet Neurol 2011

Need to Develop and Validate new Clinical Outcome Measures

• Inherited Neuropathy Consortium (NINDS-NS065712; PI Dr. Shy): identify reliable, sensitive and valid outcome measures for clinical trials in pediatric CMT.

• Clinical and related disease-quantifying measures are in development (ex. CMTPedS)

• There is currently no Patient-Reported Outcome (PRO) measure being developed for future trials.

Patient-Reported Outcome (PRO) Measures

• PROs provide Direct measure of patient’s perspective on a given domain (depression, pain, function (ADLs), without clinician interpretation

• FDA defines treatment benefit as a Direct measure of improvementDirect measures include:- Survival- How patient feels, or- How patient functions

Surrogate measures of improvement include:- Blood pressure, Cholesterol, HbA1C, etc.

• Health-Related Quality of Life (QOL) is a specific PRO measure that assesses the patient’s perspective on the impact of chronic illnesses on their life’s quality or function.

Cancer Treat Rep. 1985 Oct;69(10):1155-9. Food and Drug Administration requirements for approval of new anticancer drugs.Johnson JR, Temple R.

AbstractFor approval of new drugs, the law requires a

minimum of two independent well-controlled studies for each indication. The preferred study design is prospective, randomized, and comparative. The use of historical controls, while sometimes acceptable, must be justified.

A favorable effect on survival and/or quality of life is generally required for approval.

PROs in Cancer Drug Approvals: ’95-’04

Product Year PRO Domains • Photofrin 1995 Dysphagia • Gemzar 1996 QOL (Pain/Weight)

• Novantrone 1996 Pain • Topotecan 1998 QOL (Symptoms) • Amifostine 1999 Xerostomia • Palifermin 2004 Mucositis

QOL as a PRO in Clinical Trials

• QOL instruments have prognostic value in many chronic illnesses, including cancer (Mapes 2003, Singh 2005; Cunningham 2005, Fang 2004; Gotay 2008; Movsas 2009; Quinten 2009; Sprangers 2010)

• In neuromuscular disorders, QOL may identify biological processes that affect long-term function earlier than traditional trial endpoints, because it is a more comprehensive assessment of one’s health status

• In addition, PROs provide invaluable data on toxicity and tolerability of a therapy in a clinical trial

• However, the decision on whether to use PROs as primary or secondary endpoints in clinical trial must be guided by their added value (ie, will the PRO data make a difference to study conclusions at the end of a trial?)

Utilization of QOL Instruments in Neuropathy Trials*

• Systematic literature review to assess the utilization of QOL in randomized controlled neuropathy trials

• Only 21% of randomized neuropathy trials utilized QOL as a primary or secondary outcome measure.

• Anonymous survey of 169 neuromuscular specialists in the United States to assess their awareness of validated QOL measures in neuropathy.

• Among the 25.4% survey respondents, 97.7% acknowledged the importance of assessing QOL, but less than a third could identify a validated QOL measure.

*-Ramchandren S and DeSousa C

Why is QOL Important in Pediatric CMT?

• QOL scores in the ascorbic acid trial were significantly reduced in children with CMT1A, compared to healthy norms (Ramchandren et al., Lancet Neurol 2009)

• These findings have been independently confirmed in children with varying types of CMT (Burns et al., J Child Neurol, 2010)

• Unfortunately, generic QOL measures are not very responsive to CMT

Study Goals and Methods*

1. Collaborate with the Inherited Neuropathy Consortium to develop and validate a disease-specific pediatric CMT quality of life instrument (pCMT-QOL) through an iterative process

2. Identify new clinical outcome measures that accurately reflects CMT disease progression in children, among composite neuropathy scores, electrophysiology and QOL data

* -Ramchandren, K23 NS072279

Systematic Review of Existing QOL MeasuresAnalysis of generic pediatric QOL measures (PEDSQL, TACQOL, CHQ)

Analysis of adult generic QOL measure (SF36):

Summary of domains from CMT literature:

Interaction in our CMT clinic:

Physical physical function, physical complaints, daily living, bodily pain, general health

Physical function, role physical, bodily pain, general health

physical disability, toe and heel walk, independent ambulation, forearm/hand intrinsic muscle strength, lower limb weakness, leg cramps, pain

Fatigue

Social cognitive, positive moods and negative moods (psychological function), physical role-social, family activities, family cohesion

vitality, social function, family, leisure items, community involvement/ productivity

Emotional Emotional role-social, mental health, self-esteem, parental emotion and time impact

role emotional, mental health, self-esteem

Psychosocial stressors (Arnold, 2005),depression

Other financial status, living situation, religion

Inadequate sleep

Cross-sectional studies correlating generic QOL domains to functional assessments*

• 70 children (40 boys, 30 girls) aged 5-16 years (mean age 9.1, SD 3.0 yrs)

• Independent determinants of reduced QOL in children with CMT1A were leg cramps, hand tremor, and reduced long jump distance

• Many standardized clinical and electrophysiologic measures currently used as end-points in clinical trials of CMT correlated poorly with QOL measured by the CHQ.

*Burns J, Ramchandren S, Ryan MM, Shy M, Ouvrier R. Neurology 2010

QOL Domain Predictor variable

β weight

Multiple r2 model

Physical Health Summary

Long jump

Leg cramps

.456**

.359**

.322

Psychological Health Summary

Leg cramps

Hand tremor

.365**

.329**

.386

Developing questions pertinent to the domains: collaboration with Neuro-QOL

• Neuro-QOL: an offshoot of the NIH Roadmap initiative Patient-Reported Outcomes Measurement Information System (PROMIS)

• Benefits: – Large database of items pertinent to specific domains

(depression, anger, physical function), normative data, Rasch and Item response theory (IRT) analyses, CAT capability

– These item banks will be accessible to clinical researchers eventually through web-based computer adaptive testing systems

– PROMIS core group is interested in collaborating with investigators interested in developing disease-specific instruments using questions from their item bank.

Conceptual Scheme for thePediatric CMT QOL Instrument- Ratified by Expert Consensus*

• Physical symptoms- Fatigue/weakness, pain, sleep, cramps

• Physical function- ADL’s, Upper extremity and lower extremity functions, balance

• Social role-physical- Physical activities with peers, adults

• Mental symptoms- Stigma, anxiety/fear, depression, stress

• Mental function- Perceived cognitive function

• Social role-emotional- Self esteem, emotional bonding with peers, adults

Physical Domain

Mental Domain

*ENMC meeting, Naarden, September 2009

Focus Group Testing of Pediatric CMT QOL Instrument

• N= 31; 15 boys, 16 girls (39 evaluated, 6 boys and 2 girls classified as not CMT)

• 84% Caucasian• Ages 4-17 years (mean age 10.2

yrs, SD 4.3 yrs)• 10 identified genotypes• The physical function domains of

our instrument correlated well with the CMTNS, providing early content validity

• Issues that needed immediate changes: Age groups (5-18 changed to 8-18 years); recall period (“in past 7 days” to “lately”); “don’t know” option

Final Version of pCMT-QOL

Three versions:

1. A child version (for children ages 8-18)

2. A parent-proxy version (for children ages 8-18)

3. A parent-proxy version (for children ages 7 and younger)

PHYSICAL

Symptoms (10 questions)

Function (10 questions)

Social Activities (10 questions)

EMOTIONAL

Feelings (10 questions)

Cognition (10 questions)

Social Skills (10 questions)

Six Domains

Table 1: Summary of enrollment and demographic data

Sample size: (N; enrolled to date September 5th, 2012) = 131

Site breakdown:Children’s hospital of Philadelphia= 13National Hospital of Neurology /Neurosurgery, UK = 29University of Rochester = 5Wayne State University = 36Children’s Hospital at Westmead, Sydney, Australia = 43University of Iowa = 5

Gender: 65 males , 63 females 3 missing data

Ages 8-18: 99; Ages 7 and under: 30; 2 missing data

Patient Enrollment in pCMT-QOL

Other Measures being Investigated

• Disability Severity Index: An anchor for CMT– Categorization of disability severity based on

walking/movement aid used; both patient and physician perspectives

– This will serve as an Anchor in future, i.e. will help to determine the minimum change in our QOL measure that is functionally important to patient

– Status: 200 patients, 94 physicians (3 of whom have CMT) enrolled; data analysis in progress

Conclusions• There is a critical need for new therapies and new

clinical outcome measures in Peripheral Neuropathy

• QOL is a useful outcome measure since it is severely reduced in pediatric inherited neuropathy and therefore can be targeted for improvement in a therapeutic trial

• We have developed a disease-specific pediatric CMT QOL outcome measure and completed preliminary testing for content validity

• Future steps include psychometric testing of the final instrument through Dr. Shy’s RDCRC- Inherited Neuropathy Consortium study, as well as identification of other clinical outcome measures.

Acknowledgements

• Mentors- Dr. Michael Shy (U Iowa), Dr. Lawrence Lum (KCI), Dr. Jeff Loeb

(WSU), Dr. Deborah Ellis (WSU), Dr. Penner (WSU)

• INC Collaborators- Dr. Richard Finkel, Dr. Joshua Burns, Dr. Rosemary Shy, Dr.

Mary Reilly, Dr. Muntoni, Dr. Pareyson, Dr. Hermann

• Coordinators - Carly Siskind, Lindsey Miller, Shawna Feely, Stephanie Neal

• Biostatisticians- Dr. Templin (WSU), Dr. McDermott (U of Rochester), Dr.

Ingemar Merkies (Spaarne Hospital, The Netherlands)

• Grant support for Dr. Ramchandren: - NIH (NINDS):1K23NS072279