With Kalorama Information estimating the worldwide marketfor pain management drugs and devices at about $35 billion last year, thestakes for bringing therapies for pain conditions to market are higher thanever—and clinical trial sponsors in this space are working to address some ofthe challenges inherently involved in investigating what can be highlysubjective patient outcomes.
In response to these concerns, diverse players across thedrug development spectrum are working to achieve greater clarity and soundermeasurement of patient response. Of equal concern to trial sponsors is stayingup-to-date on the growing body of industry-authored literature as well asregulator guidance on the subject, all with the aim of differentiating theiranalgesia products in the competitive pharmaceutical marketplace.
And the conditions that may be affected by their efforts arejust as diverse, with both chronic and acute pain conditions likeosteoarthritis, lower back pain, diabetic neuropathy, fibromyalgia, whiplashinjury and complex regional pain syndrome all standing to benefit from industryconsensus on the way trials in this space are conducted.
"Pain is subjective. It's what the patient tells you it is,and there is no objective way of measuring that," says Dr. Robert H. Dworkin, aprofessor of anesthesiology, neurology, oncology and psychiatry at the Universityof Rochester School of Medicine and Dentistry.
In 2002, Dworkin co-founded the Initiative on Methods,Measurement, and Pain Assessment in Clinical Trials, colloquially known asIMMPACT. Seeking to develop consensus reviews and recommendations for improvingthe design, execution and interpretation of clinical trials for paintreatments, the interdisciplinary group, IMMPACT's participants includedacademia, regulatory agencies, consumer support and advocacy groups as well asindustry.
"We met to discuss a wide range of issues involving design,analysis, conduct and interpretation of clinical trials," Dworkin says. "Wethought about the heterogeneity of evaluating pain treatments and doingmeta-analysis, and that it would help the field if there was some consensus ondealing with outcomes in clinical trials. We went into that first meetingthinking it would be a one-off kind of thing, but the group quickly decided tomeet on a regular basis."
Subsequent meetings yielded a set of standards by which trialscould be more effectively conducted for a wide range of disciplines such asanesthesiology, clinical pharmacology, internal medicine, law, neurology,nursing, oncology, outcomes research, psychology, rheumatology and surgery.Now, coming off its 16th meeting in June, IMMPACT's recommendationsand systematic reviews have been widely cited and have guided the design ofclinical trials, other types of clinical research and a national survey.Additional consensus meetings are planned, and research initiatives involvingthe assessment of pain and the design and interpretation of clinical trials areongoing.
The main outcome of these efforts has been the developmentof Patient Reported Outcomes, or PROs, which is an umbrella term covering arange of consequences of a disease condition and/or its treatment as reportedby the patient, including symptom experience, sense of wellbeing, functionalstatus, treatment adherence and treatment satisfaction.
"For the assessment of pain, PRO measures are often the onlyviable clinical endpoints because there are no objective physical orpsychological markers of disease or treatment activity that can be observed ormeasured," writes Michael Kuss, vice president of analgesia at Premier Researchin a recent white paper, "Patient-Reported Outcomes in Analgesia ClinicalTrials." "Evaluation of analgesic efficacy requires the assessment of multiplePRO domains in order to adequately characterize treatment impact."
"PROs need to be primary endpoint, and PROs should also bevery well represented in secondary endpoints," Dworkin says. "One of the thingspeople have thought about is more objective measures of physical function, suchas patient activity levels during the day, and even while sleeping. If a drugis efficacious and reduces pain more than the placebo, does the drug alsoincrease a patient's activity level over the course of the trial? That would bean interesting secondary outcome."
Dworkin notes that data generated by a PRO instrument canprovide evidence of a treatment benefit from the patient perspective, but forthis data to be meaningful, there should be evidence that the PRO instrumenteffectively measures the particular concept that is studied. Generally,findings measured by PRO instruments may be used to support claims in approvedproduct labeling if the claims are derived from adequate and well-controlledinvestigations that use PRO instruments that reliably and validly measure thespecific concepts at issue.
The U.S. Food and Drug Administration (FDA) and European MedicinesAgency (EMA) have imposed standards that call for a higher degree of scientificrigor in the development and evaluation of PRO measures, as well as the levelof documentation required to support use and content validity of these measuresin confirmatory clinical trials.
"If you're going to bring novel PROs to the FDA, you betterbe sure to have crossed your Ts and dotted your Is. Seek guidance from the FDAas soon as possible," Dworkin advises.
Many sponsors seeking expert consultation from contractresearch organizations (CROs) with expertise in PRO best practices andinstrument validation—and in fact, Kuss' white paper advises companies to dojust that. One such CRO, Chapel Hill, N.C.-based Rho, recently published itsown white paper, "The Minimal Clinically Important Difference in EfficacyTrials of Analgesics," on the subject.
"Various PROs will likely remain the gold-measures inclinical trials for analgesics … but they do not indicate whether the magnitudeof the improvement would be meaningful to patients," writes Dr. BrianBoehlecke, medical officer at Rho. "Small differences may be statisticallysignificant, but not indicative of a change in pain intensity which manyindividuals would consider important and worth the expense and possible risksassociated with taking that medication. Therefore, for trials of the efficacyof analgesics, it is necessary to determine the smallest change in a PRO painscale score that would be considered important by patients—i.e., the minimal clinically important difference (MCID)."
The most commonly used quantitative scales for a patient'srating of pain intensity are numeric rating scales (NRS) and linear visualanalog scales (VAS). The former often has choices ranging from 0 to 10representing "none" to "the worst pain imaginable, respectively. The latterrequires a respondent to mark a point along a 100-mm line indicating painintensity, with the extremes having the same descriptive anchors as the NRS.
There are two general approaches to estimating the MCID:anchor-based and distribution-based. Anchor-based methods compare the changesin pain scale scores with an independent external criterion or anchor. Thelatter is usually some type of global assessment rating in which the patientsrate the change in their pain status under treatment into one of severalsubjective descriptive categories such as "very much worse," "slightly worse,""no change," "slightly better," "much better" or "very much better."
"You want to establish a result that is meaningful for thepatient, something that is anchored to some other thing a patient is reporting—achange from baseline improvement to the more meaningful improvement of, 'yes, Ifeel better.' You want a tight enough estimate that you have statisticalsignificance," explains Ben Vaughn, a statistical scientist at Rho.
The average change in pain scale score for the subjects whorated their change in pain into the chosen category (e.g., "much better") is considered to be the MCID.
"Although reductions of 30 to 35 percent from baselinescores on an 11-point NRS or a 100-mm VAS have been found to be reasonableestimates for the MCID in a variety of studies of both acute and chronic pain,there is no 'one-size-fits-all' level of MCID in all settings and for allpurposes. Patients' assessments of the importance of a given reduction in painintensity may be influenced by whether it has allowed them to reach a tolerablelevel of residual pain," concludes Boehlecke.