Biomarkers for Prediction of Analgesic Efficacy in Knee OA.

Knee, Osteoarthritis (OA), Biomarkers for Prediction of OA Treatment Efficacy

With high NNTs for indiscriminative use in chronic pain, treatment unavoidably entails frustrating long trial and errors. It is timely to identify biomarkers that can predict analgesic efficacy for the individual patient. The investigators propose a framework of interrelations between patient's pain modulation profile (PMP) and the drug's mode of action (MOA) based on two principles: (1) 'fix the dysfunction', relevant for drugs whose main mode of action is to modulate central pain processing; the more the dysfunctional the better the modulating drug efficacy. For example, patients with pro-nociceptive PMP due to reduced endogenous pain inhibition, as expressed by less efficient CPM will benefit from drugs that fix this dysfunction such as SNRIs, relative to patients whose pain inhibitory capacity is well functioning. Thus, for the modulating drugs, pro-nociceptivity predicts better efficacy. (2) 'bear with the dysfunction', relevant for drugs which are mostly non-modulating, acting mainly in the periphery; the more dysfunctionalת the less the non-modulating drug efficacy. This is since efficacy is limited by the dysfunctional modulation system, despite the drug's MOA-like reduction of peripheral pain mediators. Thus, for the non-modulating drugs, for example NSAIDs, pro-nociceptivity predicts less good efficacy. The likely protocol suggests that patients with anti-nociceptive PMP should be treated primarily by non-modulating drugs, while pro-nociceptive ones should be given modulating drugs. EEG is an additional source of relevant data on brain pain processing. Being objective and stable along time, EEG based parameters are, thus, very attractive candidates to be useful biomarkers for prediction of analgesia efficacy. This study will focus on the patients with painful knee osteoarthritis. The aims of this study are: 1. To identify psychophysical and neurophysiological biomarkers that can serve as predictors of response to analgesic pain modulating and non-pain modulating drugs. 2. To establish a conceptual framework of individualized pain therapy based on inter-relations between patient's parameters of pain modulation and drugs' mode of action.

Study design: The study design includes two experimental meeting sessions (before and at the end / after the treatment) which include clinical and experimental assessments. After the first experimental session, the patients will be asked to rate twice a week their daily pain along two weeks, in order to confirm their OA pain level; the patients with the mean pain score of ≥4 will be supplied with the study medications. Along the 8 weeks-long treatment period, they will provide the rating of OA pain, subjective estimation of pain alleviation and reports of side effects Clinical assessment: Will be performed by the study physician. The data on OA severity by Kellgren and Lawrence system classification, range of motion and current OA pain (last 48 h) will be collected. In addition, all patients will fill the brief pain inventory questioner (BPI) to assess their pain characteristics. In addition, all patients will be tested for the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) for assessment of OA pain, stiffness and physical function. Experimental session: 1. At the beginning of pre-treatment experimental session, all patients will fill a following set of psychological pain-related questioners organized in one document: (1) pain catastrophizing scale (PCS), (2) HADS anxiety and depression, (3) short-form health survey (SF-12), (5) pain sensitivity questionnaire (PSQ). In addition, basic assessment of psychomotor attention and cognitive functioning will be performed using (6) Trial making tests A and B (TMT A and B) and (7) Digits symbol substitution test (DSST). All the data will be coded and no personal data will be exposed. 2. Resting-state EEG recording. Three minutes of resting-state EEG (eyes closed) will be recorded using the 64-channel EEG recording (Brain Products GmbH, Munich, Germany). 3. Psychophysical pain assessment. All tests will be performed remotely from the painful area - on arm or hand. The following tests will be performed: * Electrical and mechanical temporal summation (TS). For the assessment of electrical TS, a train of 10 electrical stimuli (Digitimer DS7A, Digitimer Ltd, WelWyn Garden City, England) of 2 msec width will be delivered to the non-dominant forearm with ISI of 1 sec. Subject will rate the perceived pain after the 1st and after the 5th stimuli using verbal numerical rating scale (NRS). Mechanical TS with be assessed using ten application of 256mN dull needle (from the DFNS set for quantitative sensory assessment) aimed to evoke slight pricking sensation. The TS responses will be calculated as a difference of the perceived pain between last vs. first pain score. * CPM. A combination of tonic heat stimulus (up to 50oC, duration up to 20 sec; TSA, Medoc, Israel) and of 3 pressure pain threshold (PPT) measurements performed on the dominant trapezius (Algomed, Medoc, Israel; 3-sec inter-stimulus interval (ISI) will be given. Stimulus heat intensity will be individually adjusted to induce pain at intensity of 50 (on 0 - 100 numerical rating scale, NRS). These stimuli will be delivered to the dominant forearm, serving a 'test-stimulus'. After a 10 min break, the non-dominant hand will be immersed into cold water bath for a period of 60-70 sec (4-10 deg). After 10 sec of the immersion, the 'test-'stimulus' will be delivered again while the hand is still in the water. Along all stimulation period, the subjects will rate their pain perception using numerical pain scale. A CPM response will be calculated as a difference in the pain perception to the 'test-stimulus' during the immersion as compared to the 'test-stimulus' given stand-alone. Treatment follow-up: Phone follow-up will be performed: weekly reports at weeks 1-2 and 5-6; twice a week for weeks 3-4 and 7-8. The patients will provide their OA pain score, rating of the pain-relieving drug effect (0-100 scale) and describe the treatment-related side effects for the period of last 48 hours). Statistical analysis The classical statistical analysis will be based on correlations between PMP and degree of drug efficacy, represented by percentage pain reduction. We then construct 3 independent model systems, one for each of the 3 PMP parameters (CPM, TS, and EEG based connectivity). Within each model we first test the two correlations, under the presumed pain modulating and non-modulating drugs, between PMP and drug efficacy. A machine learning-based cross-validation and permutation tests will be used in order to access generalizability and statistical significance of the of the findings.

No related papers found yet.

Papers published before 2008-02-04 may not appear yet. Historical indexing is in progress.

Israel