As you review these studies, think about the differences in the study designs and approach to answering their respective research questions. Think about the strengths and limitations of each design. Consider the benefits or disadvantages of each design. Consider how the research question might affect the type of study design we choose. Maybe consider if the experimental study could have been conducted using a cross-sectional approach. Similarly, consider the research question answered in the cross-sectional study and if that could lend itself to an experimental design. Think about the limitations one design might have compared to its alternative. You should also consider feasibility, cost, or other considerations you imagine an investigator might have. After completing your readings, answer the questions below. 1. Could the research question addressed in the study by Lane et al. have been conducted using a cross-sectional design? Why or why not? What would be some of the potential advantages/disadvantages of using this alternative design? 2. Could a relationship between jogging and bone mineral density in the study by Mussolino et al. have been evaluated using an experimental approach? Why or why not? What would be some of the potential advantages/disadvantages of using this alternative design? Could the research question addressed in the study by Lane et al. have been conducted using a cross-sectional design? Why or why not? What would be some of the potential advantages/disadvantages of using this alternative design? Yes, since we are told that the observational study involves analysis of data which was collected from a population at one specific point in time, it means that the research question addressed in the study by Lane et al. have been conducted using a cross-sectional design. From the article, the cross-sectional design was an alternative design which had some advantages as well as the disadvantages. The potential advantages were that; the data which was collected frequently using this design allows large cross-sectional studies to be done at little or no expense. This implies that the use of the design minimizes costs. Further, in this study, a specific group is always checked to verify whether an activity carried out is related to the effect being investigated. However, one of the main disadvantages is that the routine data may not be designed to answer a specific question. The other disadvantage is that the specific questions asked may act as a source of biases in issues where personal feelings are involved. Could a relationship between jogging and bone mineral density in the study by Mussolino et al. have been evaluated using an experimental approach? Why or why not? What would be some of the potential advantages/disadvantages of using this alternative design? Yes, the relationship between jogging and bone mineral density in the study by Mussolino et al. could also have been evaluated using an experimental approach. An experimental approach is a systematic approach to research where the researcher manipulates one or more variables, and controls and measures any change in other variables. Here, an experimental approach would be appropriate for the population from which the data was being collected, analyzed and then divided into five different groups so that at the end of data collection they could be compared and analyzed properly. This approach has some advantages as well as the disadvantages according to the article. The main advantage is that one has full control over the variables. This is an important aspect to the researchers since they will get accurate results by isolating and determining what they are looking for from the study. Another advantage of the experimental design is that it is a basic and efficient form of design thus makes it useful in many different types of situations. However, the design is disadvantageous. Firstly, the design is prone to human errors. This destroys the validity of the experiment completely. Secondly, the design can be manipulated by the researchers since they have control of the variables being tested. Lastly, an experimental design is time-consuming since it takes researchers an extensive amount of time to complete the comprehensive research. The n e w e ng l a n d j o u r na l of m e dic i n e original article Tanezumab for the Treatment of Pain from Osteoarthritis of the Knee Nancy E. Lane, M.D., Thomas J. Schnitzer, M.D., Ph.D., Charles A. Birbara, M.D., Masoud Mokhtarani, M.D., David L. Shelton, Ph.D., Mike D. Smith, Ph.D., and Mark T. Brown, M.D. A BS T R AC T Background Increased expression of nerve growth factor in injured or inflamed tissue is associated with increased pain. This proof-of-concept study was designed to investigate the safety and analgesic efficacy of tanezumab, a humanized monoclonal antibody that binds and inhibits nerve growth factor. Methods We randomly assigned 450 patients with osteoarthritis of the knee to receive tanezumab (administered at a dose of 10, 25, 50, 100, or 200 g per kilogram of body weight) or placebo on days 1 and 56. The primary efficacy measures were knee pain while walking and the patient's global assessment of response to therapy. We also assessed pain, stiffness, and physical function using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC); the rate of response using the criteria of the Outcome Measures for Rheumatology Committee and Osteoarthritis Research Society International Standing Committee for Clinical Trials Response Criteria Initiative (OMERACT-OARSI); and safety. Results When averaged over weeks 1 through 16, the mean reductions from baseline in knee pain while walking ranged from 45 to 62% with various doses of tanezumab, as compared with 22% with placebo (P<0.001). Tanezumab, as compared with placebo, was also associated with significantly greater improvements in the response to therapy as assessed with the use of the patients' global assessment measure (mean increases in score of 29 to 47% with various doses of tanezumab, as compared with 19% with placebo; P0.001). The rate of response according to the OMERACT-OARSI criteria ranged from 74 to 93% with tanezumab treatment, as compared with 44% with placebo (P<0.001). The rates of adverse events were 68% and 55% in the tanezumab and placebo groups, respectively. The most common adverse events among tanezumab-treated patients were headache (9% of the patients), upper respiratory tract infection (7%), and paresthesia (7%). From the University of California at Davis Medical School, Sacramento (N.E.L.); Northwestern University Feinberg School of Medicine, Chicago (T.J.S.); University of Massachusetts School of Medicine, Worcester (C.A.B.); Rinat Neuroscience, South San Francisco, CA (M.M., D.L.S.); and Pfizer, New London, CT (M.D.S., M.T.B.). Address reprint requests to Dr. Lane at the Department of Medicine, University of California at Davis Medical School, 4800 Second Ave., Suite 2600, Sacramento, CA 95817, or at nelane@ucdavis.edu. Drs. Lane and Schnitzer contributed equally to this article. This article (10.1056/NEJMoa0901510) was published on September 29, 2010, at NEJM .org. N Engl J Med 2010;363:1521-31. Copyright 2010 Massachusetts Medical Society. Conclusions In this proof-of-concept study, treatment with tanezumab was associated with a reduction in joint pain and improvement in function, with mild and moderate adverse events, among patients with moderate-to-severe osteoarthritis of the knee. (Funded by Rinat Neuroscience; ClinicalTrials.gov number, NCT00394563.) n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1521 The N n e w e ng l a n d j o u r na l erve growth factor is a neurotro phin that regulates the structure and function of responsive sensory neurons, including small-diameter nociceptive afferents. There has been increasing recognition of the potential role of nerve growth factor in pain modulation through nociceptor sensitization.1-7 In animals and humans, exogenous nerve growth factor increases pain either locally or systemically, depending on the dose and the route of administration.8,9 Increased expression of nerve growth factor is found in inflamed tissues from patients with conditions such as arthritis, pancreatitis, and prostatitis.10-12 Levels of nerve growth factor are also elevated in animal models of inflammatory pain, and pharmacologic inhibition of the activity of nerve growth factor in these models reduces or blocks signs of pain. Therefore, nerve growth factor appears to have a role in causing and augmenting pain in these models.1,13-15 The development of therapeutic interventions that are based on antagonism of nerve growth factor is of interest.16,17 The treatment options for patients with painful osteoarthritis of the knee are inadequate. Nonsteroidal antiinflammatory drugs and narcotic analgesics are commonly used18,19; however, these medications have well-described gastrointestinal and cardiorenal side effects,20,21 and the response to them is unsatisfactory in some patients.22,23 Potent analgesic medications with acceptable sideeffect profiles may help to avoid or delay surgical intervention.24 Tanezumab is a humanized IgG2 monoclonal antibody directed against nerve growth factor that blocks the interaction of nerve growth factor with its receptors, TrkA and p75.25 A small phase 1 clinical trial showed that a single intravenous injection of tanezumab substantially reduced pain in patients with osteoarthritis of the knee.26,27 We report the results of a proof-of-concept study of tanezumab in patients with advanced osteoarthritis of the knee who did not have a satisfactory response to nonopiate pain medications or who were considered to be candidates for invasive intervention. We compared the safety, side-effect profile, and efficacy of repeat doses of tanezu mab as compared with placebo. of m e dic i n e the basis of American College of Rheumatology criteria,28 with radiographic confirmation (Kellgren-Lawrence grade 2 or higher, on a scale of 0 to 4, with higher numbers indicating more severe signs of osteoarthritis). Patients were eligible only if they were unwilling to take nonopiate pain medications or had had an unsatisfactory response to them or if they were candidates for or seeking invasive interventions such as intraarticular injections or total knee replacement. All pain medications except the \"rescue\" medications, acetaminophen and tramadol, were discontinued at the screening visit. At the time of randomization, patients had to have pain while walking on a flat surface (the walking-pain measure of the Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]) that they rated between 50 and 90 on a visual-analogue scale that ranged from 0 to 100, with 100 indicating maximal pain. In addition, among patients who discontinued pain medication during the screening period, an increase in the walking-pain score of 10 or more was required between screening and randomization. The exclusion criteria were pregnancy, a history of or current symptoms of an autoimmune disorder, cancer within the previous 5 years except for cutaneous basal-cell or squamous-cell cancer resolved by excision, allergic reaction to monoclonal antibodies or IgG-fusion proteins, infection with hepatitis B or hepatitis C virus or the human immunodeficiency virus, drug abuse, fibromyalgia, clinically significant cardiac disease, diabetes mellitus requiring oral treatment or insulin, clinically significant neurologic disease, or a clinically significant psychiatric disorder. All participants provided written informed consent. Study Design and Oversight Patients were recruited between March 30, 2006, and May 3, 2007, at 46 study centers in the United States and were screened within 30 days before randomization. Eligible patients who were taking pain medication other than acetaminophen and tramadol underwent a washout period (of at least 5 half-lives of the medication). Patients rated their knee pain and recorded the score in an electronic diary every day for 3 days before randomization to establish their baseline pain score. Eligible patients were randomly assigned on day 1, with the Me thods use of an interactive voice-response system, to plaStudy Population cebo or to tanezumab at a dose of 10, 25, 50, 100, We enrolled patients, 40 to 75 years of age, who or 200 g per kilogram of body weight, such that had osteoarthritis of the knee as diagnosed on there were equal numbers in each study group. 1522 n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Tanezumab for Osteoarthritis of the Knee A pharmacist at each study site received each patient's randomization number and prepared each patient's dosing solution. Other than the pharmacist at each site and one statistician at the contract research organization, all the staff members and patients involved in the study were unaware of the group assignments. The study medication was administered intravenously on days 1 and 56. Study visits were scheduled for days 14, 28, 70, 84, 112, 136, and 182, during which safety and efficacy assessments were performed and serum samples for routine laboratory tests and for pharmacokinetic analyses were obtained. In addition, patients were contacted by telephone on day 42 to ask about adverse events. Patients recorded their knee pain and their use of rescue medication daily in an electronic diary. The rescue medications that were permitted were acetaminophen at a dose of 3000 mg or less per day, tramadol at a dose of 400 mg or less per day, or both, during the washout period and days 1 through 28 (weeks 1 through 4) and acetaminophen at a dose of 3000 mg per day for the remainder of the study. Patients could enter an open-label extension of the trial (ClinicalTrials.gov number, NCT00399490) at week 16 (day 112) if they had received two doses of the study drug and had been followed for at least 8 weeks after the last dose. The study was designed and coordinated by Rinat Neuroscience, a subsidiary of Pfizer. An external data and safety monitoring board monitored safety, and day-to-day study operations, including data management, were overseen by PPDI (a contract research organization contracted by Pfizer). The data were analyzed by Pfizer. The first author wrote the first draft of the manuscript. All the authors were involved in the design of the study and interpretation of the data, contributed to the writing of the manuscript, made the decision to submit the manuscript for publication, attest that the study was performed in accordance with the protocol and the statistical analysis plan, and vouch for the accuracy and completeness of the reported results. Editorial support was provided by UBC Scientific Solutions and was funded by Pfizer. The study protocol was approved by the local ethics committee at each study center before patient enrollment began. The protocol, including the statistical analysis plan, is available with the full text of this article at NEJM.org. Efficacy Assessments The primary efficacy outcomes were the change from baseline in the pain the patient felt in the index knee while walking on a flat surface and in the patient's global assessment of response to therapy, averaged over weeks 1 through 16. Secondary efficacy outcomes included the change from baseline in overall knee pain and in scores on the WOMAC subscales for pain, stiffness, and physical function. Pain while walking and overall knee pain were recorded daily in an electronic diary, whereas the patient's global assessment of response to therapy and scores on the WOMAC subscales were recorded on study-visit days. Pain, the patient's global assessment, and scores on the WOMAC subscales were assessed with the use of a visual-analogue scale that ranged from 0 to 100. In the case of pain and WOMAC scores, a lower score indicated improvement (i.e., less pain, less stiffness, and less limitation of physical function), whereas in the case of the patient's global assessment, a higher score indicated improvement (i.e., a better response to therapy). Another secondary outcome was the response to therapy on the basis of the criteria of the Outcome Measures for Rheumatology Committee and Osteoarthritis Research Society International Standing Committee for Clinical Trials Response Criteria Initiative (OMERACT-OARSI).29 Patients were classified as having had a response if the WOMAC pain or physical-function score decreased by 50% or more and by 20 or more points on the visual-analogue scale or if two of the following three findings were recorded: a decrease in the WOMAC pain score by 20% or more and by 10 or more points on the visual-analogue scale, a decrease in the WOMAC physical-function score by 20% or more and by 10 or more points on the scale, or an increase in the score on the patient's global assessment by 20% or more and by 10 or more points on the scale. Rescue medication use, also a prespecified secondary outcome, was recorded daily in the patients' diaries. Safety Assessments The nature, onset, duration, severity, and outcome of all adverse events, as well as any relationship of an adverse event to the study drug were ascertained and documented at each visit. Safety assessments included physical and neurologic examinations (e.g., evaluation of mental status, n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1523 The n e w e ng l a n d j o u r na l of m e dic i n e medication and underwent at least one efficacy assessment (the modified intention-to-treat population) (Fig. 1). Most patients in the modified intention-to-treat population had a Kellgren-Lawrence grade of 3 (52%) or 4 (17%) and severe pain33 both at the time of screening (mean [SD] score on the visual-analogue scale, 5812) and at the time of randomization, after washout of previous medications (score on the visual-analogue scale, 7111). A total of 87% of the patients reported Statistical Analysis taking pain medication for knee pain before they We estimated that we would need to enroll 75 enrolled in the study. The baseline characteristics patients in each group for the study to have 80% of the patients were similar across study groups power to detect a difference between the tanezu (Table 1). mab groups and the placebo group of 15 points or more on the visual-analogue scale for the pri- Efficacy mary outcomes (the average change from base- As compared with placebo, tanezumab, at all the line through week 16 in knee pain while walking doses studied, was associated with an improveand in the patient's global assessment of re- ment in the primary efficacy measures. The sponse to therapy), with an effect size of 0.5 (in- mean reduction from baseline in the score on the dicating a moderate difference).31 With respect to visual-analogue scale for knee pain while walkthe intensity of pain, decreases of 10 or more ing, averaged over weeks 1 through 16, ranged points on a visual-analogue scale that ranges from 31.0 to 45.2 points with various doses of from 1 to 100 are considered to be minimally tanezumab, as compared with 15.5 points with important improvements, and decreases of 20 or placebo (a reduction of 45 to 62% with tanezummore points are considered to be moderately im- ab vs. 22% with placebo, P<0.001 for the comportant improvements.32 parison of all doses of tanezumab with placebo) Changes from baseline in all the measures that (Fig. 2A). We observed significant improvements were assessed with the use of a visual-analogue among patients receiving tanezumab as comscale were determined with a mixed-model, re- pared with those receiving placebo by the end of peated-measures analysis, with model terms for the first week, and significant improvements study site, study group, study week, and the in- continued to be seen throughout the remainder teraction between study medication and study of the treatment period. The mean increase from week, and with the baseline score on the visual- baseline in the score on the patient's global assessanalogue scale as a covariate, with no imputation ment of response to therapy, averaged over weeks for missing data. A repeated-measures analysis 1 through 16, ranged from 16.3 to 23.7 points was also used to assess the number of rescue with various doses of tanezumab, as compared medication pills taken. For the analysis of rates with 9.2 points with placebo (an increase of 29 to of response according to OMERACT-OARSI crite- 47% with tanezumab vs. 19% with placebo, ria (calculated on the basis of the average change P0.001 for the comparison of all doses of tanfrom baseline to week 16), we used the Cochran- ezumab with placebo) (Fig. 2B). By week 2, the Mantel-Haenszel test, stratified according to study scores on the patient's global assessment had imsite, to compare the proportions of patients in the proved in the group receiving 25 g of tanezumtanezumab groups who had a response with the ab per kilogram, as compared with placebo proportion of those in the placebo group who had (P=0.002); by week 4, the scores had improved in a response. the groups receiving 50 g, 100 g, and 200 g of tanezumab per kilogram, as compared with placebo (P=0.01, P<0.001, and P<0.001 for the three R e sult s comparisons, respectively); and during weeks 10 Baseline Characteristics of the Patients and 12, the scores improved in the group receivOf the 450 patients who underwent randomiza- ing tanezumab at a dose of 10 g per kilogram, tion, 440 received at least one dose of the study as compared with placebo (P=0.008). The im- strength, reflexes, sensation, and coordination), cognitive testing with the use of the Hopkins Verbal Learning Test-Revised,30 assessment of postural vital signs, and electrocardiography. Clinically significant abnormalities on a neurologic examination performed by the investigator or adverse events suggestive of peripheral neuropathy were further evaluated by an independent neurologist. 1524 n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Tanezumab for Osteoarthritis of the Knee 450 Patients underwent randomization 75 Were assigned to receive placebo 75 Were assigned to receive tanezumab, 10 g/kg 75 Were assigned to receive tanezumab, 25 g/kg 75 Were assigned to receive tanezumab, 50 g/kg 75 Were assigned to receive tanezumab, 100 g/kg 75 Were assigned to receive tanezumab, 200 g/kg 65 Discontinued study by wk 26 24 Entered OLE 28 Had no response 13 Had other reason 10 Completed study 56 Discontinued study by wk 26 36 Entered OLE 6 Had an adverse event 7 Had no response 7 Had other reason 19 Completed study 53 Discontinued study by wk 26 34 Entered OLE 1 Had an adverse event 12 Had no response 6 Had other reason 22 Completed study 62 Discontinued study by wk 26 38 Entered OLE 4 Had an adverse event 10 Had no response 10 Had other reason 13 Completed study 50 Discontinued study by wk 26 33 Entered OLE 3 Had an adverse event 4 Had no response 10 Had other reason 25 Completed study 60 Discontinued study by wk 26 38 Entered OLE 8 Had an adverse event 5 Had no response 9 Had other reason 15 Completed study 73 Were included in modified intentionto-treat analysis 74 Were included in safety analysis 74 Were included in modified intentionto-treat analysis 74 Were included in safety analysis 75 Were included in modified intentionto-treat analysis 74 Were included in safety analysis 72 Were included in modified intentionto-treat analysis 74 Were included in safety analysis 74 Were included in modified intentionto-treat analysis 74 Were included in safety analysis 72 Were included in modified intentionto-treat analysis 74 Were included in safety analysis Figure 1. Randomization and Follow-up. Eligible patients could enter the open-label extension (OLE) of the study at week 16. provements were maintained through week 16 in the groups receiving 25 g, 100 g, and 200 g of tanezumab per kilogram. The mean reductions from baseline in overall knee pain over the course of weeks 1 through 16 were similar in magnitude to those reported for knee pain while walking (reductions of 43 to 62% with tanezumab vs. 23% with placebo, P<0.001 for the comparison of all doses of tanezumab with placebo). Treatment with tanezumab, as compared with placebo, was also associated with reductions in the mean WOMAC scores for pain (reductions of 46 to 64% vs. 23%), stiffness (48 to 65% vs. 22%), and physical function (47 to 65% vs. 22%) over the same period (P<0.001 for all comparisons) (Table 2). The percentage of patients who had a response to therapy according to OMERACT-OARSI criteria, averaged over weeks 1 through 16, was significantly higher with tanezumab treatment than with placebo (74 to 93% vs. 44%, P<0.001 for the comparison of all doses of tanezumab with placebo) (Table 2). Rescue medications that were allowed per protocol were used less frequently by tanezumab-treated patients than by placebo-treated patients during weeks 1 through 16 (odds ratio with tanezumab, 0.50; 95% confidence interval [CI], 0.24 to 1.02; P=0.05) and was significantly lower during weeks 1 through 4 (odds ratio, 0.49; 95% CI, 0.24 to 0.99; P=0.04). Safety Among patients in the tanezumab groups, the three most common adverse events were headache, upper respiratory tract infection, and paresthesia (Table 3). The incidence of treatmentrelated adverse events was higher among patients treated with 100 g or 200 g of tanezumab per kilogram than among patients who received lower doses (28% and 35% in the groups receiving 100 g and 200 g per kilogram, respectively, vs. 11 to 18% in the groups receiving other doses). Peripheral sensory symptoms, including paresthesia, were reported in 14% of the patients receiving tanezumab and in 4% of those receiving placebo (Table 3); the severity of these adverse n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1525 The n e w e ng l a n d j o u r na l of m e dic i n e Table 1. Baseline Characteristics of the Study Patients.* Characteristic Age yr Placebo (N=74) Tanezumab, 10 g/kg (N=74) Tanezumab, 25 g/kg (N=74) Tanezumab, 50 g/kg (N=74) Tanezumab, 100 g/kg (N=74) Tanezumab, 200 g/kg (N=74) 58.17.7 58.38.3 59.98.1 60.47.7 57.18.2 58.47.6 Female sex no. (%) 42 (57) 49 (66) 50 (68) 37 (50) 44 (59) 40 (54) White race no. (%) 66 (89) 62 (84) 67 (91) 66 (89) 67 (91) 64 (86) 2 18/73 (25) 21/73 (29) 23/74 (31) 29/74 (39) 22/74 (30) 19/73 (26) 3 or 4 Kellgren-Lawrence grade no./total no. (%) 55/73 (75) 52/73 (71) 51/74 (69) 45/74 (61) 52/74 (70) 54/73 (74) Knee pain while walking 71.610.0 70.610.9 71.710.5 68.110.2 71.111.0 72.411.5 Patient's global assessment of response 48.820.8 55.720.3 51.020.6 51.616.9 49.919.9 54.422.4 WOMAC score Pain 69.011.9 65.813.9 69.212.5 62.112.3 68.313.2 68.412.0 Stiffness 74.413.5 69.713.1 75.012.4 66.717.5 71.217.9 73.313.1 Physical function 69.012.5 63.813.6 69.214.6 62.612.3 67.414.8 67.814.0 *\tPlus-minus values are means SD. \tRace was self-reported. \tA Kellgren-Lawrence score of 2 (minimal signs of osteoarthritis) indicates definite osteophytes without reduction of the joint space; a score of 3 (moderate signs of osteoarthritis) indicates diminished joint space; and a score of 4 (severe signs of osteoarthritis) indicates greatly reduced joint space. Data are from the intention-to-treat population, and missing data are excluded. Knee pain while walking was assessed with the use of a visual-analogue scale that ranged from 0 to 100, with higher scores indicating more pain. Data are from the modified intention-to-treat population. \tPatients' global assessment of response to therapy was assessed with the use of a visual-analogue scale that ranged from 0 to 100, with higher scores indicating a better response to therapy. Data are from the modified intention-to-treat population. \tScores on the pain, stiffness, and physical-function subscales of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) were assessed with the use of a visual-analogue scale that ranged from 0 to 100, with higher scores indicating more pain, more stiffness, and more limitation of physical function, respectively. Data are from the modified intention-to-treat population. events was mild in 56 of the 70 patients with these symptoms in the tanezumab groups and in all 3 patients with these symptoms in the placebo group and was moderate in the remaining 14 patients in the tanezumab groups. The results of neurologic examinations of these patients were predominantly normal; in patients with clinically significant changes, abnormalities were minor and consisted mainly of changes in sensation in the extremities and diminished deep-tendon reflexes. A total of 15 patients had abnormal peripheral sensation that was unresolved at the final visit. Of these 15 patients, 6 had clinically significant findings on neurologic examination: decreased ankle reflexes in 1 patient who was receiving 100 g of tanezumab per kilogram; decreased temperature and a sensation of sharp pain, vibration, or both, in the toes or feet in 4 patients (1 each in the groups receiving 25 and 100 g of tanezumab per kilogram and 2 in the 1526 group receiving 200 g per kilogram); and bilateral decreased sensation in pain, fine touch, and temperature in a radial-nerve distribution in 1 patient who was receiving 100 g of tanezumab per kilogram; all other aspects of the neurologic examination were normal. Additional follow-up data were available for 7 patients with abnormal peripheral sensation who subsequently participated in the open-label extension study, and in each case, the adverse event resolved before the patient's entry into the extension study. Similarly, follow-up data were available for 1 of the 6 patients with neurologic findings, and in the case of this patient, the neurologic findings had normalized by the time of the baseline visit in the open-label extension study. The mean time to the onset of abnormal sensory symptoms was 33 days (median, 14) after the first dose of tanezumab, and the mean duration of symptoms was 18 days (median, 4) across tan- n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Tanezumab for Osteoarthritis of the Knee A Patient's Assessment of Knee Pain while Walking Dose 1 Mean Change from Baseline over Weeks 1-16 Dose 2 0 Placebo Change from Baseline 10 20 15.52.6 Tanezumab, 10 g/kg 32.12.5 Tanezumab, 25 g/kg 36.02.5 Tanezumab, 50 g/kg 31.02.6 Tanezumab, 100 g/kg 42.52.5 Tanezumab, 200 g/kg 45.22.6 30 40 50 60 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Week B Patient's Global Assessment of Response Dose 1 Mean Change from Baseline over Weeks 1-16 Dose 2 Placebo Change from Baseline 30 9.21.8 Tanezumab, 10 g/kg 16.31.7 Tanezumab, 25 g/kg 23.61.6 Tanezumab, 50 g/kg 17.51.7 Tanezumab, 100 g/kg 23.71.6 Tanezumab, 200 g/kg 21.01.7 20 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Week Figure 2. Change from Baseline in Patients' Assessment of Knee Pain while Walking and in Patients' Global Assessment of Response to Therapy. The patient's assessment of knee pain while walking and the patient's global assessment of response to therapy were obtained at baseline and at the indicated times with the use of a visual-analogue scale that ranged from 0 to 100. In the case of knee pain, a decrease in the score indicates improvement (i.e., less pain); in the case of the patient's global assessment, an increase in the score indicates improvement (i.e., a better response to therapy). Changes are reported as least-squares means SE. P<0.001 for the comparisons of all doses of tanezumab with placebo in the assessment of knee pain and global assessment of response, except for the comparison of 10 g of tanezumab per kilogram of body weight with placebo in the patient's global assessment, for which P=0.001. n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1527 The n e w e ng l a n d j o u r na l of m e dic i n e Table 2. Secondary Efficacy Outcomes.* Placebo (N=73) Tanezumab, 10 g/kg (N=74) Tanezumab, 25 g/kg (N=75) Tanezumab, 50 g/kg (N=72) Pain subscale -16.22.4 -30.12.3 -36.02.2 Stiffness subscale -16.32.4 -33.52.3 -37.72.2 Physical-function subscale -15.22.3 -30.12.3 Response to therapy according to OMERACT-OARSI criteria by week 16 (% of patients) 43.8 74.3 Outcome Tanezumab, 100 g/kg (N=74) Tanezumab, 200 g/kg (N=72) -29.02.4 -39.62.2 -43.52.3 -34.52.4 -42.72.2 -47.82.4 -34.92.2 -30.82.4 -40.52.2 -43.82.3 84.0 75.0 93.2 93.1 Change in WOMAC score from baseline through week 16 *\tPlus-minus values are means SE. P<0.001 for all comparisons of the five doses of tanezumab with placebo. These analyses were performed on data from the modified intention-to-treat population. \tOne patient who was randomly assigned to receive 25 g of tanezumab per kilogram of body weight instead received 50 g per kilogram. \tAccording to the criteria of the Outcome Measures for Rheumatology Committee and Osteoarthritis Research Society International Standing Committee for Clinical Trials Response Criteria Initiative (OMERACT-OARSI), patients were classified as having had a response if the WOMAC pain or physical-function score decreased by 50% or more and by 20 or more points on the visual-analogue scale or if two of the following three findings were observed: a decrease in the WOMAC pain score by 20% or more and by 10 or more points on the scale, a decrease in the WOMAC physical-function score by 20% or more and by 10 or more points on the scale, or an increase in the score on the patient's global assessment by 20% or more and by 10 or more points on the scale. ezumab groups. Some differences in the onset and duration of these symptoms were noted: allodynia, dysesthesia, and hyperesthesia tended to develop primarily after the first dose of tanezumab had been administered and were relatively short-lived, whereas the onset and duration of paresthesia were more variable. Serious adverse events were reported in 6 patients (2%) receiving tanezumab (appendicitis, bacterial arthritis, cellulitis, spinal stenosis, breast cancer, and syncope) and in 1 patient (1%) receiving placebo (noncardiac chest pain). A total of 6% of tanezumab-treated patients withdrew from the study because of adverse events; no placebotreated patients withdrew because of adverse events (Fig. 1). We observed no clinically important changes in electrocardiographic findings, postural vital signs, or mental status or cognition in any of the study participants, and we did not detect the presence of antitanezumab antibodies in any of the patients assigned to a tanezumab group. One site in the current study also participated in a subsequent phase 3 trial and was closed by the sponsor owing to substantial noncompliance with Good Clinical Practice guidelines and with the protocol in that phase 3 study. Therefore, all analyses from the current study were repeated, with the 23 patients from that site excluded; only very small changes in the efficacy and safety re1528 sults and slight increases in P values were seen when the patients from that site were not included in the analyses (see the Supplementary Appendix, available at NEJM.org). Discussion Two injections of tanezumab a monoclonal antibody that inhibits nerve growth factor 8 weeks apart at doses ranging from 10 to 200 g per kilogram resulted in clinically significant reductions in knee pain, stiffness, and limitations of physical function in patients with moderate-tosevere knee osteoarthritis. Entries in daily pain diaries indicated that differences between tanez umab therapy and placebo were apparent within days after the first injection, and the efficacy persisted throughout the 4-month treatment period. Although the study was not powered to assess dose response, and no formal dose-response analysis was performed, the reductions in pain appeared to be greater among patients taking higher doses of tanezumab (100 or 200 g per kilogram) than among those taking lower doses, with no clear benefit of the 200-g dose over the 100-g dose. Clinically meaningful pain relief is often described as a reduction in pain intensity of approximately 30% from the baseline level,34,35 and in this study, reductions ranged from 45 to 62% with tanezumab. Furthermore, reductions in n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Tanezumab for Osteoarthritis of the Knee Table 3. Frequency of Adverse Events. Placebo (N=74) Variable Tanezumab, 10 g/kg (N=74) Tanezumab, 25 g/kg (N=74) Tanezumab, 50 g/kg (N=74) Tanezumab, 100 g/kg (N=74) Tanezumab, 200 g/kg (N=74) number of patients (percent) Any adverse event 41 (55) 51 (69) 49 (66) 44 (59) 51 (69) 58 (78) Treatment-related adverse event 6 (8) 11 (15) 13 (18) 8 (11) 21 (28) 26 (35) Severe adverse event* 2 (3) 6 (8) 3 (4) 3 (4) 3 (4) 3 (4) Treatment-related severe adverse event 0 0 1 (1) 0 1 (1) 0 Serious adverse event 1 (1) 2 (3) 0 2 (3) 0 2 (3) 2 (3) 8 (11) 5 (7) 8 (11) 6 (8) 6 (8) Adverse event occurring in 5% of tanezumabtreated patients Headache Upper respiratory tract infection 4 (5) 2 (3) 6 (8) 5 (7) 7 (9) 7 (9) Arthralgia 0 1 (1) 2 (3) 5 (7) 4 (5) 7 (9) Pain in extremity 0 3 (4) 1 (1) 2 (3) 6 (8) 9 (12) Peripheral edema 2 (3) 0 2 (3) 5 (7) 6 (8) 8 (11) Allodynia 0 0 0 0 1 (1) 1 (1) Burning sensation 1 (1) 0 0 0 1 (1) 0 Adverse event involving abnormal peripheral sensation Dysesthesia 0 0 0 0 1 (1) 1 (1) Hyperesthesia 0 0 0 3 (4) 4 (5) 4 (5) Hypoesthesia 0 1 (1) 6 (8) 2 (3) 5 (7) 5 (7) Neuralgia 0 0 0 0 1 (1) 0 Neuritis 0 0 0 0 1 (1) 0 Pallanesthesia 0 0 1 (1) 0 0 1 (1) Paresthesia 2 (3) 4 (5) 4 (5) 1 (1) 8 (11) 8 (11) Sensory disturbance 0 0 0 1 (1) 1 (1) 2 (3) Sensory loss 0 0 1 (1) 0 0 1 (1) *\tThe severity of an adverse event refers to the maximum intensity of the event. An event was considered to be severe (as compared with mild or moderate) if it interfered substantially with the patient's usual functioning. \tAn adverse event was classified as serious if it was fatal or life-threatening, required or prolonged inpatient hospitalization, was disabling, resulted in a congenital anomaly or birth defect, or required medical or surgical intervention to prevent permanent impairment or damage. No serious adverse event was considered to be treatment-related. \tThe determination of a rate of 5% or higher was made on the basis of all tanezumab groups combined. pain with tanezumab therapy resulted in pain scores that were equal to or lower than those reported by patients at the time of screening, when they were taking their previously prescribed pain medications. The majority of adverse events that were reported by patients taking tanezumab, including abnormal peripheral sensations, were mild to moderate in severity. The occurrence of adverse events appeared to be dose-dependent. Since nerve growth factor is thought to act on small-diameter sensory afferents, the occurrence of paresthesia and other signs associated with large-fiber sensory function is interesting. The more frequent occurrence of these events within a short time after administration of the first dose suggests that there may be transient changes in sensitivity or \"tone\" of different afferent fiber populations, leading to altered sensations. Owing to their largely transient nature, it is unlikely that these adverse events are indicative of neurodegenerative changes. The transient nature of these events is consistent with n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1529 The n e w e ng l a n d j o u r na l findings in long-term studies of the safety of highdose tanezumab in nonhuman primates.36 Elucidation of the mechanisms underlying these effects and any potential long-term consequences require further investigation. The assessed measures of mental status and cognition were unchanged, suggesting that the effects of tanezumab were limited to the peripheral nervous system. The limitations of this study include the lack of a comparison group receiving a different active treatment, a study population that was too small for a statistical comparison of efficacy according to dose, and short-term exposure to tanezumab (since knee pain from osteoarthritis usually requires long-term treatment). However, our study shows efficacy in patients with more severe osteoarthritis than those in other trials. For example, two studies37,38 of glucosamine and chondroitin in patients with osteoarthritis of the knee excluded patients who had the highest degree of severity on radiography (a Kellgren-Lawrence grade of 4), whereas 17% of our study population had this degree of severity. Baseline scores for knee pain on the visual-analogue scale in these two other studies were 54 and 57, respectively, whereas our population had a mean score for knee pain on the day of randomization of 7111. Since the completion of this study and through May 24, 2010, progressively worsening osteoarthritis associated with radiographic evidence of bone necrosis developed in 16 subjects participating in 1 of 13 phase 3 studies of tanezumab for of m e dic i n e osteoarthritis of the hip and knee; all 16 subjects required total joint replacements. The affected joints were the knee, hip, or shoulder (predominantly unilateral involvement), with more than half the cases occurring in a joint other than the index joint under evaluation in the study. These 16 events led the Food and Drug Administration (FDA) on June 22, 2010, to put the osteoarthritis clinical program for tanezumab on clinical hold until more information can be obtained to determine the true incidence and the causality of these events. More recently, the FDA requested the suspension of two additional trials of tanezumab, one involving patients with low back pain and the other involving patients with diabetic neuropathy. Our proof-of-concept study showed that tanezumab had a favorable efficacy profile for the treatment of moderate-to-severe knee pain associated with osteoarthritis. Longer trials involving larger samples are needed to better understand safety and tolerability issues and explore the clinical potential of tanezumab as an alternative to current pharmacologic treatments. Supported by Rinat Neuroscience, now a subsidiary of Pfizer. Editorial support was provided by Papia Das and Elizabeth Young of UBC Scientific Solutions and was funded by Pfizer. Dr. Lane was also supported by a grant (K24-AR04884) from the National Institutes of Health. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. We thank Leslie Tive, Susan Simpson, and Carol Zhao for assistance in the preparation of the manuscript. References 1. Woolf CJ, Safieh-Garabedian B, Ma QP, Crilly P, Winter J. Nerve growth factor contributes to the generation of inflammatory sensory hypersensitivity. Neuroscience 1994;62:327-31. 2. McMahon SB. NGF as a mediator of inflammatory pain. Philos Trans R Soc Lond B Biol Sci 1996;351:431-40. 3. Skaper SD. The biology of neurotrophins, signalling pathways, and functional peptide mimetics of neurotrophins and their receptors. CNS Neurol Disord Drug Targets 2008;7:46-62. 4. Petruska JC, Mendell LM. The many functions of nerve growth factor: multiple actions on nociceptors. Neurosci Lett 2004; 361:168-71. 5. Barker PA, Murphy RA. The nerve growth factor receptor: a multicomponent system that mediates the actions of the neurotrophin family of proteins. Mol Cell Biochem 1992;110:1-15. 6. Frade JM, Barde YA. Nerve growth 1530 factor: two receptors, multiple functions. Bioessays 1998;20:137-45. 7. Nicol GD, Vasko MR. Unraveling the story of NGF-mediated sensitization of nociceptive sensory neurons: ON or OFF the Trks? Mol Interv 2007;7:26-41. 8. Hao J, Ebendal T, Xu X, WiesenfeldHallin Z, Eriksdotter Jnhagen M. Intra cerebroventricular infusion of nerve growth factor induces pain-like response in rats. Neurosci Lett 2000;286:208-12. 9. Ruiz G, Ceballos D, Baos JE. Behavioral and histological effects of endoneurial administration of nerve growth factor: possible implications in neuropathic pain. Brain Res 2004;1011:1-6. 10. Halliday DA, Zettler C, Rush RA, Scicchitano R, McNeil JD. Elevated nerve growth factor levels in the synovial fluid of patients with inflammatory joint disease. Neurochem Res 1998;23:919-22. 11. Friess H, Zhu ZW, di Mola FF, et al. Nerve growth factor and its high-affinity receptor in chronic pancreatitis. Ann Surg 1999;230:615-24. 12. Miller LJ, Fischer KA, Goralnick SJ, et al. Nerve growth factor and chronic prostatitis/chronic pelvic pain syndrome. Urology 2002;59:603-8. 13. McMahon SB, Bennett DL, Priestley JV, Shelton DL. The biological effects of endogenous nerve growth factor on adult sensory neurons revealed by a trkA-IgG fusion molecule. Nat Med 1995;1:774-80. 14. Ramer MS, Bisby MA. Adrenergic innervation of rat sensory ganglia following proximal or distal painful sciatic neuropathy: distinct mechanisms revealed by antiNGF treatment. Eur J Neurosci 1999;11: 837-46. 15. Zahn PK, Subieta A, Park SS, Brennan TJ. Effect of blockade of nerve growth factor and tumor necrosis factor on pain behaviors after plantar incision. J Pain 2004;5:157-63. 16. Hefti FF, Rosenthal A, Walicke PA, et al. n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. Tanezumab for Osteoarthritis of the Knee Novel class of pain drugs based on antagonism of NGF. Trends Pharmacol Sci 2006; 27:85-91. 17. Dray A, Read SJ. Arthritis and pain: future targets to control osteoarthritis pain. Arthritis Res Ther 2007;9:212. 18. Dillon CF, Rasch EK, Gu Q, Hirsch R. Prevalence of knee osteoarthritis in the United States: arthritis data from the Third National Health and Nutrition Examination Survey 1991-94. J Rheumatol 2006;33:2271-9. 19. Zhang W, Moskowitz RW, Nuki G, et al. OARSI recommendations for the management of hip and knee osteoarthritis. II. OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage 2008; 16:137-62. 20. Johnsen SP, Larsson H, Tarone RE, et al. Risk of hospitalization for myocardial infarction among users of rofecoxib, celecoxib, and other NSAIDs: a populationbased case-control study. Arch Intern Med 2005;165:978-84. 21. Whelton A. Renal and related cardiovascular effects of conventional and COX-2specific NSAIDs and non-NSAID analgesics. Am J Ther 2000;7:63-74. 22. Bjordal JM, Ljunggren AE, Klovning A, Slrdal L. Non-steroidal anti-inflammatory drugs, including cyclo-oxygenase-2 inhibitors, in osteoarthritic knee pain: metaanalysis of randomised placebo controlled trials. BMJ 2004;329:1317. 23. Bjordal JM, Klovning A, Ljunggren AE, Slrdal L. Short-term efficacy of pharmacotherapeutic interventions in osteoarthritic knee pain: a meta-analysis of randomised placebo-controlled trials. Eur J Pain 2007;11:125-38. 24. Hawker GA. Who, when, and why total joint replacement surgery? The patient's perspective. Curr Opin Rheumatol 2006; 18:526-30. 25. Abdiche YN, Malashock DS, Pons J. Probing the binding mechanism and affinity of tanezumab, a recombinant humanized anti-NGF monoclonal antibody, using a repertoire of biosensors. Protein Sci 2008;17:1326-35. 26. Hefti F, Mokhtarani M, Gray M, Zhao C, Chan C. RN624 (anti-NGF) reduces pain and improves function in subjects with moderate to severe pain from osteoarthritis of the knee. J Pain 2006;7:Suppl:S45. abstract. 27. Lane N, Webster L, Lu S, Gray M, Hefti F, Walicke P. RN624 (anti-NGF) improves pain and function in subjects with moderate knee osteoarthritis: a phase I study. Arthritis Rheum 2005;52:Suppl: S461. abstract. 28. Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039-49. 29. Pham T, van der Heijde D, Altman RD, et al. OMERACT-OARSI initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage 2004;12:389-99. 30. Shapiro AM, Benedict RH, Schretlen D, Brandt J. Construct and concurrent validity of the Hopkins Verbal Learning Testrevised. Clin Neuropsychol 1999;13:34858. 31. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: L. Erlbaum, 1988. 32. Dworkin RH, Turk DC, Wyrwich KW, et al. Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. J Pain 2008;9:105-21. 33. Collins SL, Moore RA, McQuay HJ. The visual analogue pain intensity scale: what is moderate pain in millimetres? Pain 1997;72:95-7. 34. Rowbotham MC. What is a \"clinically meaningful\" reduction in pain? Pain 2001; 94:131-2. 35. Salaffi F, Stancati A, Silvestri CA, Ciapetti A, Grassi W. Minimal clinically important changes in chronic musculoskeletal pain intensity measured on a numerical rating scale. Eur J Pain 2004;8:283-91. 36. Shelton DL, Zorbas M, Bales R, Rogers B, Pons J. Preclinical safety evaluation of PF04383119 after single and multipledose administration in monkeys. Anesthesiology 2008;109:A597 (Web only). abstract. (Accessed July 30, 2010, at http://www .asaabstracts.com/strands/asaabstracts/ searchArticle.htm;jsessionid=618E87D0E 0ABC77F4B2A108910544EA7?index=1& highlight=false&highlightcolor=1&bold= true&italic=false.) 37. Clegg DO, Reda DJ, Harris CL, et al. Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. N Engl J Med 2006;354:795808. 38. Kahan A, Uebelhart D, De Vathaire F, Delmas PD, Reginster JY. Long-term effects of chondroitins 4 and 6 sulfate on knee osteoarthritis: the study on osteoarthritis progression prevention, a two-year, randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2009;60: 524-33. Copyright 2010 Massachusetts Medical Society. apply for jobs electronically at the nejm careercenter Physicians registered at the NEJM CareerCenter can apply for jobs electronically using their own cover letters and CVs. You can keep track of your job-application history with a personal account that is created when you register with the CareerCenter and apply for jobs seen online at our Web site. Visit NEJMjobs.org for more information. n engl j med 363;16 nejm.org october 14, 2010 The New England Journal of Medicine Downloaded from nejm.org on March 18, 2017. For personal use only. No other uses without permission. Copyright 2010 Massachusetts Medical Society. All rights reserved. 1531