APSS-66-00 nSTRIDE APS versus Hyaluronic Acid for Knee Osteoarthritis
Research type
Research Study
Full title
A Two-Phase, Multicenter, Randomized Study Comparing Autologous Protein Solution with Hyaluronic Acid Intra Articular Injections in Patients with Knee Osteoarthritis
IRAS ID
224637
Contact name
Anish K Amin
Contact email
Sponsor organisation
Zimmer Biomet
Duration of Study in the UK
6 years, 2 months, 1 days
Research summary
Summary of Research
The purpose of this double-blinded, randomized controlled trial is to evaluate the safety and effectiveness of the nSTRIDE APS Kit, a device used to treat patients with Osteoarthritis (OA) of the knee. Pain, function, stiffness, and quality of life in patients injected with APS will be compared to the same measures in patients injected with Synvisc One (hyaluronic acid, a compound found naturally in the fluid of the knee). About 246 patients will take part in this study at several investigative centers across Europe. 123 patients will be given an APS injection and 123 patients will receive a Synvisc One injection.
Following the injection, subjects will have four visits over the course of 12 months. After the 12 month visit, individual treatment allocation will be unblinded. From this time-point on, only subjects will be blinded to the individual treatment allocation resulting in a single-blind design during the long-term follow-up period (12 – 60 months). Subjects from both groups will be able to request additional injections of their originally assigned treatment as frequently as needed, provided they did not experience any significant clinical concerns after previous treatment administrations and are benefiting from it, as determined by the investigator. During the long-term follow-up phase (12 – 60 months), subjects may also elect to cross over to the other treatment arm and receive additional injections of the other treatment as frequently as needed. Subjects may only cross over from their originally assigned treatment group to the other treatment group one time during the study.The primary endpoint will be the change in pain from baseline to 12 months following injection of nSTRIDE APS or HA, as measured by the WOMAC LK 3.1 pain subscale. In addition, the EQ-5D, the NRS, the Patient's Preference and Resource Costing questionnaires will be administered to assess several secondary endpoints.
Summary of Results
REPORT SUMMARYIntroduction
Osteoarthritis (OA) is a degenerative and disabling articulating joint disease that affects both younger, more active patients (e.g., patients with trauma or who have prolonged participation in highly demanding sports) and the elderly. The disease is progressive and debilitating, eventually resulting in pain that may be so severe that restive sleep is impossible, along with life-altering loss of function. Surgical intervention is clinically successful, and widely used, in treating severe degenerative OA; however, treatment modalities for less advanced OA are associated with varying, and often disappointing, rates of success. Current treatment options include nonsteroidal anti-inflammatory drugs (NSAIDs), and intra-articular corticosteroid or hyaluronic acid (HA) injections. Although these treatments can relieve pain temporarily for some OA patients, they do not address the biological mechanisms causing the disease.
Although OA is classified as a non-inflammatory disease, inflammation is implicated in many symptoms and in OA progression. Pro-inflammatory cytokines that are involved in OA development and maintenance include Tumor Necrosis Factor alpha (TNFα), as well as interleukin (IL)-1, -6, and -8. Evidence exists that these catabolic cytokines are abundantly present in the OA joint, whereas the level of anti-inflammatory cytokines is decreased. Autologous Protein Solution (APS), produced using the nSTRIDE APS Kit (further referred to as nSTRIDE APS), contains concentrated levels of anti-inflammatory cytokines, including IL-1ra, sIL-1RII, sTNF-RI, and sTNF-RII. It is designed to halt and potentially reverse the OA disease process by rebalancing intra-articular cytokine levels.
In this clinical investigation ‘A Two-Phase, Multicenter, Randomized Study Comparing Autologous Protein Solution (APS) with Hyaluronic Acid (HA) Intra Articular Injections in Patients with Knee Osteoarthritis (OA),’ the objective was to determine whether nSTRIDE APS is superior to HA (Synvisc-One) in improving the mean Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) LK 3.1 pain score (change from baseline to 12 Months post-injection) in patients with early to moderate symptomatic OA. Secondary objectives were to evaluate whether the nSTRIDE APS was superior to HA regarding the improvement in Numeric Rating Scale (NRS) pain, Outcome Measures in Rheumatology – Osteoarthritis Research Society International (OMERACT-OARSI) responder criteria and WOMAC Function. Lastly, the safety of nSTRIDE APS was compared to HA following intra-articular knee injections up to 12 Months.Long-term (until 48-month) follow-up included: patient preference, information on the choice of a possible second injection, the number of subjects receiving additional intervention, treatments success (improvement in WOMAC Pain score), injection success (subjects not receiving a second injection or additional intervention and showing improvement in WOMAC Pain score), initial treatment success (subjects receiving the same or no second injection, no additional intervention, and showing improvement in WOMAC Pain score), WOMAC function, EQ-5D, NRS Pain, and OMERACT-OARSI scores over time as well as long-term safety analysis on (S)AEs and deaths.
This final report describes the data from all visits up to and including the 48 Month follow-up visit. Note that the study was terminated after all subjects completed the 48 Month visit, due to the company’s business decision (see Section 3.13).
Results through 12-Months (Phase 1)
Study set-up
The target population for this study was patients aged at least 18 years, diagnosed with knee OA, with a baseline WOMAC Pain score of ≥ 9 and ≤ 19 and who failed at least 1 prior conservative OA therapy. After confirmation of all eligibility criteria, the subjects were randomized in a 1:1 ratio into two treatment groups: nSTRIDE APS or Synvisc-One. Balanced randomization with random block sizes was implemented. Subjects were followed up at 1, 3, 6 and 12 Months (first phase) until 48 months after the initial treatment (second phase) or until they have received another invasive treatment in the index knee.The first subject was enrolled on 25 July 2017 and subject recruitment was completed on 20 September 2018, when the last subject was randomized and received injection. The last 12 Month visit took place on 27 September 2019 and the last 24 Month visit took place on 13 November 2020, later than originally scheduled due to COVID-19. A total of 127 subjects were randomized in the Synvisc-One group and 125 subjects in the nSTRIDE APS group. At 12 Months, 111 subjects in the nSTRIDE APS group and 106 subjects in the Synvisc-One group had WOMAC data. Fourteen (14) subjects in the nSTRIDE APS group had missing 12 Month follow-up data due to: additional invasive treatment in index knee within 12 months of initial injection (n=8), informed consent withdrawal (n=1), lost to follow-up (n=3), exit for non-compliance (n=1) and one subject was randomized but not treated (n=1). In the Synvisc-One group, 21 subjects had no 12 Month follow-up data due to: study exit for intolerable adverse event (AE) leading to invasive treatment in index knee (n=2), additional invasive treatment in index knee within 12 months of initial injection (n=12), informed consent withdrawal (n=1), lost to follow-up (n=2), exit for non-compliance (n=1) and study exit for other reason (n=3).
Demographics
The mean age of all enrolled subjects in the nSTRIDE APS group was 58.9 ± 10.5 years and in the Synvisc-One group 57.9 ± 10.8 years with overall age of the study population ranging between 27 and 91 years. In both groups most treated subjects were female (nSTRIDE APS: 60%; Synvisc-One: 61.4%). Demographic distribution of gender, age, BMI, race, and baseline WOMAC Pain were similar between treatment groups, and no significant differences were observed between the groups in any of these parameters.Primary Endpoint
The primary objective of this study was to determine whether nSTRIDE APS is superior to Synvisc-One with respect to the improvement in mean WOMAC Pain score (change from baseline to 12 Months post single-injection). The primary analysis was performed on the intent-to-treat (ITT) population (n=252, nSTRIDE APS: n=125; Synvisc-One: n=127) and missing data was imputed. Scores of non-missing values for each case were added and the mean value was used for the missing values, unless there were too many missing items, then the response was considered invalid and was not included in the analysis.Although the primary analysis utilized the ITT population, the primary endpoint (i.e. the WOMAC Pain score) was also analyzed for improvement over time in raw and percent changes using the Per Protocol (PP) population. These are exploratory analyses per the Statistical Analysis Plan (SAP), but as their results provide noteworthy information about the primary endpoint, they are also mentioned in this section.
The mean WOMAC Pain scores show a statistically significant percent decrease from baseline to 12 months within each treatment group. In particular, at 12 Months there was a mean WOMAC Pain percent change of -41.1% [CI: -47.9, -34.3] (raw score decrease: -4.4, confidence interval [CI]: -5.2, -3.7) in the nSTRIDE APS group and -40.4% [CI: -47.2, -33.7] (raw score decrease: -4.1, CI: -4.9, -3.4) in the Synvisc-One group. The mean raw change likewise shows a statistically significant decrease within each treatment group. The difference between the group means (raw change) at 12 Months was 0.3 (CI: -1.4, 0.8), which was not significantly different.
The Impact of the usage of nSTRIDE APS on function was further evaluated in support of the primary analysis and a qualitative assessment of the changes in WOMAC Function over time for nSTRIDE APS and Synvisc-One were discussed in order to ensure that the impact of the treatment on function was neutral or positive. The mean WOMAC Function percent change at post-treatment visit was -42.1% (95% CI: -48.8, -35.5; raw data score decrease: -14.3 ± 13.5) at 12 Months for the nSTRIDE APS group and -40.2% (95% CI: -48.0, -32.4; raw data score decrease: -15.7 ± 14.8) at 12 Months for the Synvisc-One group. A significant improvement of WOMAC Function at 12 Months compared to baseline was observed within each treatment group (APS: P<0.0001; Synvisc-One: P<0.0001), although there was no significant difference in WOMAC Function improvement between the nSTRIDE APS and Synvisc-One groups at the 12 Month follow-up (secondary analysis, P=0.4314). Sensitivity analyses were performed on the per protocol (PP) population, but similar results were obtained as with the ITT population, and the mean change from baseline at 12 Months was not significantly different between both groups.
These results indicate that both treatments were successful in producing a clinically relevant (i.e. >20%) and statistically significant reduction in pain and improvement of function at one year post-injection.
Secondary Endpoints
Secondary objectives of this study were to determine whether nSTRIDE APS was superior to Synvisc-One regarding the improvement in MCID WOMAC Function and pain, the OMERACT-OARSI responder criteria, WOMAC Pain and function for KL II subgroup, and NRS Pain. Rescue medication (acetaminophen) and restricted medication usage was also compared between groups. The secondary analyses were performed on the PP population (n=200; nSTRIDE APS: n=102; Synvisc-One: n=98). The secondary tests were performed if the primary null hypotheses had been rejected. However, as the null hypothesis was not rejected, these tests were performed as exploratory analyses instead.MCID WOMAC Responders: Function and Pain The values of MCID for WOMAC Function and MCID for WOMAC Pain were based on the data collected from a previous nSTRIDE APS study, PROGRESS II. The calculations for each MCID value are specified in the SAP. A subject was considered an “MCID Function Responder” if they showed an absolute improvement of >= 20 points in WOMAC Function from baseline to 12 Months. A subject was considered an “MCID Pain Responder” if they showed an absolute improvement of >= 7.0 points in WOMAC Pain from baseline to 12 Months. Results showed no significant difference between treatment groups in WOMAC MCID Pain and Function responders, with the majority considered non-responders.
OMERACT-OARSI Responders
The OMERACT-OARSI criteria are well-established and defined in the SAP. Additionally, non-responders also included subjects with two or more documented uses of rescue medication within 48 hours of the 6- or 12-Month visits for index knee OA, or two or more documented uses of restricted medication for index knee OA between the 6- and 12-Month visits. OMERACT-OARSI criteria, based on improvement of NRS, did show the majority being responders (nSTRIDE APS: 58.8%; Synvisc-One: 60.2%), but no significant difference between treatment groups was observed.KL II WOMAC Subscales: Pain and Function Per the baseline confirmation of the standing knee radiograph, 22.6% (28/124 subjects) of the nSTRIDE APS group and 23.6% (30/127 subjects) of the Synvisc-One group were considered Kellgren-Lawrence Grade II (KL II). Subgroup analyses for subjects with KL II OA at baseline also showed no difference between treatment groups for mean improvement in WOMAC Pain and Function at 12 Months.
NRS Pain and Improvement
NRS Pain, which is a validated 11-point Likert-type pain scale showed a mean change from baseline of -1.8 in both treatments (nSTRIDE APS: 1.8 ± 2.6; Synvisc-One: 1.8 ± 2.5) at the 12 Month follow-up visit. A significant improvement in NSR Pain at 6 Months compared to baseline was observed within each treatment group (nSTRIDE APS: P<0.0001; Synvisc-One: P<0.0001), although there was no significant difference in NRS Pain improvement between the nSTRIDE APS and Synvisc-One groups at the 12 Month follow-up (P=0.8832).Rescue Medication Usage for Index Knee OA The majority of subjects reported usage of acetaminophen (rescue medication) of “Never” or “A few days” for index knee OA at Months 1, 3, 6 and 12. The percentage of subjects reporting “never” using rescue medication for index knee OA is higher for the nSTRIDE APS group at Months 1, 3, 6 and 12. The minority of subjects in both groups reported usage of acetaminophen (rescue medication) “Almost half of the days”, “Most days”, or “Every day” at all time points. Considering all responses at each time point, there were no significant differences observed in the overall acetaminophen usage for index knee OA between groups. Usage of rescue medication or restricted medication was found not to significantly affect the primary endpoint in both groups.
ANOVA WOMAC Pain and Covariates
Repeated measures ANOVA showed a significant decrease in WOMAC Pain at all timepoints when compared to baseline in both treatment groups (P<0.0001). At 6 Months follow-up, Synvisc-One showed a significantly higher decrease in WOMAC Pain when compared to nSTRIDE APS (-6.74 versus -5.60, respectively, P<0.05), but not at other timepoints.Exploratory Endpoints
Exploratory objectives of this study were to determine whether nSTRIDE APS was superior to Synvisc-One with regard to NRS Pain, change in WOMAC subscales (percent, mean/ MCID at all time points and excluding 12 Months), clinical success via OMERACT-OARSI, overall total WOMAC index score, and changes in quality of life via EQ-5D-3L and EQ-VAS. Post-injection 12 Month X-ray KL grades were also compared to baseline. A post hoc analysis was done to determine high pain responders. All exploratory analyses were done on the PP population.NRS Pain (1, 3, and 6 Months)
Both treatment groups showed a significant decrease in NRS Pain at all timepoints when compared to baseline (P<0.0001) but was not significantly different between treatment groups.WOMAC Total and WOMAC Pain, Function and Stiffness Subscales The mean WOMAC Pain changes over time was significantly improved when compared to baseline values at each follow-up visit for both treatment groups (P<0.0001). Similar results were obtained for WOMAC Function and WOMAC Stiffness changes over time, showing a significant improvement of WOMAC Function and Stiffness when compared to baseline at all follow-up visits (P<0.0001). WOMAC Total showed a significant improvement at all follow-up visits when compared to baseline in both treatment groups (P<0.0001). No significant difference between treatment groups for WOMAC Pain Δ and WOMAC Function Δ (percent change) at 12 Months was observed (P=0.8901 and P=0.7086, respectively). The mean WOMAC Pain and Function were not significantly different between research sites
WOMAC MCID Responder
Following the MCID responder criteria, the majority of subjects (>50%) were considered non-responders at all time points in both treatment groups, with exception of MCID Pain Responders at 6 Months. There was a significant difference between treatment groups at 1 Month for the WOMAC MCID Function Responder rate (P<0.05). No other significant differences between treatment groups in both WOMAC Pain and Function MCID were observed at all-time points.Quality of Life
The mean EQ-5D Global Assessment VAS changes over time showed significant differences between baseline and 3, 6 and 12 Months for the nSTRIDE APS group (P<0.05) showing a decrease in VAS pain over time. For the Synvisc-One group, significant differences compared to baseline were observed at 6 and 12 Months (P<0.05). The EQ-5D Dimensions measured at 1, 3 and 6 Months showed no differences between groups. Most subjects reported no problems, or some problems. For Self-Care, Usual Activities, Pain/Discomfort and Anxiety/Depression, in a few subjects’ extreme problems were reported at 1, 3 and 6 Months.Post-Hoc Pain / Medication Usage Responder Analysis A post hoc high pain responder analysis was performed in order to identify subjects who achieved a high degree of improvement in pain, as measured by the WOMAC pain subscale. High pain responders were defined by the following criteria: improvement in pain ≥ 90%, no restricted medication usage for Index Knee OA in the 12 Month interval, and no frequent (“most days” or “every day”) usage of pain medication for index knee OA. A significantly higher number of high pain responders were identified in the nSTRIDE APS group when compared to the Synvisc-One group (11.7% versus 3.3%, P=0.0145).
Imaging
Imaging analysis was done on X-rays that were evaluated for KL grade and absence of severe OA by the central laboratory for confirmation of eligibility. Post-injection X-rays taken at 12 Months received KL grades that were compared to baseline. Most subjects showed no change in KL grade at 12 Months (nSTRIDE APS: 91.2%; Synvisc-One: 89.6%). In both treatment groups, two subjects (2.9% and 3.0%) showed improvement in KL grade by one grade.Optional second Injection
At the 12 Month follow-up visit, subjects from both treatment groups were able to request additional injections from their originally assigned treatment as frequently as needed. At this visit, they were also allowed to cross-over to the other treatment arm. Further exploratory analyses were done on treatment success and if subjects requested a second injection. In both treatment groups the majority requested a second injection at the 12 Month follow-up visit (nSTRIDE APS: 64.7%; Synvisc-One: 60.2%). WOMAC Pain changes from baseline to 12 Months and from 6 Months to 12 Months were significantly associated with the choice of second injection type (APS, Synvisc-One, or Neither). This pattern was consistent for the two treatment groups. Of subjects that underwent a second injection, the majority chose their original injection (nSTRIDE APS: 68.2%, 45/66; Synvisc-One: 64.4%, 38/59). There was a significant association between the choice of second Injection type on WOMAC Function change from baseline to 12 Months, but the change from 6 Months to 12 Months was not significantly associated with the choice of second injection type.Safety Endpoints
During the first phase of the study (up to 12 Months), all AEs were collected regardless of relationship to the device. AEs were designated in MedDRA categories by a qualified reviewer that was blinded with respect to treatment assignment. Safety analyses were performed on the ITT population.At least one AE occurred in 65.6% of the subjects in the nSTRIDE APS group (82/125) and in 61.4% of subjects in the Synvisc-One group (78/127) from the time of treatment until completion of the 12 Month visit. This was not significantly different between groups (P=0.5150). Results further showed that the occurrence of device-related events between treatment and completion of the 12 Month visit was 10.4% in the nSTRIDE APS group and 5.5% in the Synvisc-One group, which was also not significantly different between the treatment groups.
At the time of injection, a total of 40 AEs were recorded, 20% of AEs were reported in the nSTRIDE APS group and 11.8% in the Synvisc-One group. At 12 Months a total of 51 AEs were recorded, AEs were reported in 24% of the cases in the nSTRIDE APS group, and in 16.5% of the Synvisc-One group. Device-relatedness and severity of the AEs were further not significantly different between the treatment groups at injection, 3, 6, and 12 Months. Most AEs were reported in the “Musculoskeletal and connective tissue disorders” organ system (nSTRIDE APS: 19.2%; Synvisc-One: 11.8%) and included arthralgia, joint stiffness, joint swelling, and musculoskeletal discomfort.
A total of 56 device related AEs (related, probably, or possibly) were reported within 12 months of the initial injection. Of the 56 device-related AEs, 8 were noted as ongoing at study exit. The majority of device-related AEs had resolved within 30 days in both groups (nSTRIDE APS: 23/28, 82.1%; Synvisc-One: 13/16, 81.3%). Serious AEs (SAE) occurred in 6.4% (8/125) of the subjects in the nSTRIDE APS group and in 5.5% (7/127) in subjects of the Synvisc-One group.
Two SAEs were possibly related to the device: infection of the index knee in a Synvisc-One subject (06-008) requiring total synovectomy of index knee and study exit; and excessive pain in the index knee by an nSTRIDE APS subject requiring wash-out of index knee and who consequently withdrew their consent to continue study participation. All SAEs except one in the nSTRIDE APS group were resolved (excessive swelling of the index knee leading to study exit). In the Synvisc-One group, two SAEs were ongoing, one of which lead to study exit (index knee infection). One device deficiency was reported (nSTRIDE APS kit, syringe luer taper broke off whilst connected to APS device) but did not lead to any patient injury.
Results 12 through 48-Months (Phase 2)
Subject follow-up
The first subject completed the 48-Month follow-up on 29 June 2021 and the last 48-Month visit was completed on 30 January 2023. After the 12-Month follow-up visit, a total of 89 subjects dropped out. Consequently, at 48 Months, data was available from 129 subjects (64 that had initially been treated with nSTRIDE APS and 65 that had initially received Synvisc-One). Of the 114 subjects that had initially received nSTRIDE APS, at 48 months, data were available from seven (7) subjects that chose not to have a follow-up injection, 47 subjects that chose the same follow-up intra-articular injection(s) (nSTRIDE APS), and 10 subjects that chose a cross-over treatment (Synvisc-One) during (one of) their follow-up injection(s). Of the 104 subjects that had initially been treated with intra-articular Synvisc-One, at 48 months, data were available from 11 subjects that chose not to have a follow-up injection, 30 subjects that chose the same follow-up intra-articular treatment(s) (Synvisc-One), and 14 subjects that chose a cross-over treatment (nSTRIDE APS) during (one of) their follow-up injection(s). Note that ‘cross-over’ here includes subjects that crossed-over during one of their follow-up injections, not necessarily already during the second intra-articular injection.In the nSTRIDE APS group, 64/114 (56.1%) of the subjects completed the study, which is rather comparable with study completion of the Synvisc-One group (65/104 subjects, 62.5%). Study completion percentages were the highest in subjects that received the same follow-up injection(s) (nSTRIDE APS [72.3%] and Synvisc-One [73.2%]). Considerably less subjects that crossed-over from nSTRIDE-APS to Synvisc-One completed the study (10/34; 29.4%) than subjects that crossed-over from Synvisc-One to nSTRIDE-APS (24/44; 54.5%). This might possibly indicate that subjects were less satisfied with the follow-up Synvisc-One injection(s) than the follow-up nSTRIDE APS injection(s). If the study was early exited, reasons were rather similar between treatments and whether follow-up injections were given or not: additional invasive treatment for OA in index knee between, withdrawal of consent, non-compliant subjects, subjects lost to follow-up and other reasons such as the COVID pandemic, missed visits and loss of contact with the subject.
Number and choice of second injections
In total, 184 second injections were given. Thereafter, the total number of follow-up injections that were applied decreased (until the 8th injection). No 9th injection was performed in any subject. There was no significant difference in the number of received injections between the subjects with an initial treatment of nSTRIDE-APS or Synvisc-One, or whether the same or cross-over follow-up injections were given to these subjects.Whether subjects left the study early term, chose not to receive a second injection, or chose to be treated with a second injection, was not dependent on the original treatment they were randomized to. In the nSTRIDE APS group, 6.3% of the subjects left the study early term (<12 months after initial injection) for additional invasive treatment for OA in the index knee, 20.5% preferred not to receive a second injection, while the majority (73.2%) preferred a second injection. In the Synvisc-One group this distribution was 9.7%, 29.0% and 61.3%, respectively. If subjects chose to receive a second injection, the majority chose to receive the same treatment as the originally assigned treatment (75.3% in nSTRIDE APS group, 67.1% in Synvisc-One group). Subjects with an age below or equal to the median age were 2.58 times more likely to choose to cross-over when compared to subjects with an age above median.
Patient preference at 24 and 48 Months
At 24 Months, subjects who crossed over, and thereby experienced both treatments, showed a significant different preference (first treatment, second treatment, or no preference) depending on what treatment they were initially assigned to (P=0.0417). Most patients preferred nSTRIDE APS treatment at 24 months (44% (18 out of 41)).At 48 months, no significant difference in preference for the treatment injection was observed anymore for those subjects who had crossed over. In the group which initially received nSTRIDE APS treatment, 35.7% had no preference, 35.7% preferred the first injection and 28.6% the second injection (Synvisc-One). In the Synvisc-One group, 28.6% had no preference, 19% preferred the first injection and 52.4% preferred the second injection (nSTRIDE APS).
Additional intervention at the index knee The number of subjects who received additional intervention for OA in their index knee until 48-Month follow-up did not significantly differ between treatment groups. Regarding the nSTRIDE APS subjects that crossed-over, 47.8% received additional intervention, while this was 32.1% of the Synvisc-One cross-over subjects. Of the subjects that received no second injection, 14.3% of the nSTRIDE APS subjects and 5.1% of the Synvisc APS subjects received additional intervention between 12-48 months. Lastly, 15.8% of the nSTRIDE APS subjects as well as Synvisc APS subjects that received a similar second injection received additional intervention.
(Initial) treatment or injection success Treatment success (defined as >20% improvement in WOMAC Pain scores from baseline) at 24 Months was significantly higher in the subjects that crossed over from Synvisc-One towards nSTRIDE APS than in subjects that crossed over from nSTRIDE APS to Synvisc-One. There was, however, no significant difference in treatment success between treatment groups when the same treatment was used as second injection. At 48 Months, there was no significant difference in percentage of subjects achieving treatment success, whether they received the same or a crossover second injection.
Injection success (subjects achieving treatment success from the initial injection; subjects who did not receive a second injection or had no additional intervention and showed >20% improvement in WOMAC Pain score from baseline) was not different between treatment groups at every follow-up time point.
Initial treatment success (subjects achieving success from originally assigned treatment; subjects who received the same or no second injection, no additional intervention, and showed >20% improvement in the WOMAC Pain score from baseline) did not differ between treatment groups at 12, 18 and 24 Months. At 48 Months, the percentage of subjects achieving success from the originally assigned treatment was, however, significantly higher in the nSTRIDE APS-treated subjects compared with the Synvisc-One-treated subjects (80.6% vs 62.7%, respectively).
WOMAC Pain and Function
No statistically significant difference in WOMAC Pain and Function score was found between 12 and 48 Months when subjects did not receive a second injection, received the same second injection, or when they crossed over to the other treatment. Both treatments induced an improvement in WOMAC Pain and Function score until at least 48-Month follow-up.After the study, an ad hoc analysis was performed by the Sponsor to determine the cumulative effect of the treatments during the study. For this analysis, the average duration of effect per injection and the average change in WOMAC function score per injection were taken into account in all patients that did not cross-over from treatment. The first and second injection of nSTRIDE APS both separately resulted in improved WOMAC pain scores for a cumulative ~40 months. The third until sixth nSTRIDE APS injection resulted in a maintenance of WOMAC pain scores for an additional ~40 months, where the pain did not worsen from the original improvement of the first two injections, nor did it further improve. The first injection of Synvisc-One resulted in improved WOMAC pain scores for ~25 months. The second until sixth injection resulted in a maintenance of WOMAC pain scores for an additional ~60 months. Taken together, this ad hoc analysis suggests that nSTRIDE APS injections may result in a longer period of reduced pain as compared to Synvisc-One injections.
OMERACT-OARSI (non-)responders
At 48 Months after the first injection, no significant differences were found regarding the distribution of responders and non-responders between treatments based on the OMERACT-OARSI scores. Regarding the subjects that received a cross-over injection, 16.7% of the nSTRIDE APS cross-over subjects was classified as responder, while 46.7% of the Synvisc-One cross-over subjects was classified as a responder. The responder percentage of the subjects that did not receive a second injection was considerably higher: 85.7% of the nSTRIDE APS subjects and 90.0% of the Synvisc-One subjects. Regarding the subjects that received the same second injection, 70.4% of the nSTRIDE APS subjects was classified as responder, while 70.0% of the Synvisc-One subjects was classified as a responder.Quality of Life
At 12 until 48 Months after the first injection, no differences between treatment groups were found regarding EQ-5D scores, regardless of the choice of a second injection.NRS Pain scores
At 12 until 48 Months after the first injection, no significant differences were found in NRS scores between treatment groups, regardless the choice of a second injection.Health Economic Outcomes
No significant differences were observed in utility score and costs between nSTRIDE APS and Synvisc-One, regardless of the choice of a second injection.Safety analysis
After the second intra-articular injection (from 12 Months until 48 Months), 43 AEs of interest were reported for all subjects in this study. Twenty-seven (27) of the 43 AEs were not device related (62.8%). The sixteen (16) other AEs were considered device-related (n=5), probable device-related (n=3), possible device-related (n=6) and unlikely (n=2) device-related. The severity of the AEs ranged from mild (n=4), moderate (n=3), to severe (n=9). Most reported related AEs were excessive pain (n=6) and excessive swelling (n=5) of the index knee. One (1) infection of the index knee was reported. Most related AEs resolved, but four (4) were ongoing at study exit. Additionally, fifteen (15) SAEs were reported from the moment of the second injection onwards. Ten (10) of those 15 SAEs were not related (66.7%), two (2) unlikely6 related, two (2) possible related, and one (1) SAE was probable related to the device. Excessive pain (n=1), excessive swelling (n=1) and infection (n=1) of the index knee were reported as possible or probable related SAEs. Additionally, there were two unlikely6 related SAEs of the musculoskeletal system, that developed five (5) – six (6) months after the injection and were located in the shoulder joint.Seven (7) device-related AEs occurred after an intra-articular nSTRIDE APS injection, of which four (4) were considered an SAE. Most possible device-related (n=2), probable device-related (n=1) or device-related (n=1) AEs that developed after an intra-articular nSTRIDE APS injection happened in the index knee and consisted of excessive swelling (n=2), excessive pain (n=1) and infection (n=1). They were classified as moderate (n=1) to severe (n=3). Two (2) of the severe (one probable- and one possible-related) AEs in the index knee were considered SAEs and consisted of excessive swelling and infection of the index knee (both those subjects exited the study because of an “intolerable AE, or AE leading to invasive treatment in the index knee”). Three (3) out of four (4) of the AEs in the index knee (excessive swelling [n=2] and infection [n=1]) were ongoing at the study exit, despite provided medical and/or surgical treatment. Only the excessive pain (n=1) in the index knee resolved without treatment. Furthermore, one (1) mild AE occurred during blood draw and was related to the nSTRIDE APS procedure but resolved without treatment. Lastly, two (2) other AEs that developed after nSTRIDE APS injection were considered severe SAEs of musculoskeletal origin but were unlikely6 related to the device as they occurred five (5) and six (6) months after the nSTRIDE APS injection and were located in the shoulder joint.
In addition, all nine (9) possible (n=4), probable (n=2) or device-related (n=3) AEs that developed after an intra-articular Synvisc-One injection occurred in the index knee and consisted of excessive pain (n=5), excessive swelling (n=3), and “other” (excessive pain due to a possible allergic reaction to the injection; n=1). These AEs were classified as mild (n=3), moderate (n=2) or severe (n=4). All those AEs were resolved with/without medial and/or surgical treatment. One (1) possible device-related severe AE of excessive pain in the index knee was considered an SAE.
During this clinical study, one (1) death was reported. The subject was randomized to Synvisc-One treatment, did not cross-over and deceased one year after the last (fourth) Synvisc-One injection. The severe SAE started 7 days before the subject’s death with hospitalization due to kidney failure that led to multi-organ failure. This SAE was unrelated to the procedure or device.
Safety-related events per treatment during the entire study (0-48m) All safety-related events (AEs, SAEs, deaths) that occurred during the entire study (0-48 months) were analyzed as well. No considerable differences in severity and number of (S)AEs were observed between the two treatments. Most of the events were not related to the treatment.
Conclusions
The primary objective of this clinical investigation was to determine whether nSTRIDE APS is superior to Synvisc-One in improving mean WOMAC Pain (change from baseline to 12 Months post-injection). Short-term (0-12 Months) results indicated that both treatment groups showed a statistically significant and clinically relevant improvement at all timepoints for WOMAC Pain when compared to baseline, although no difference between treatments were observed. Both treatments showed a significant improvement in WOMAC Pain of >40% from baseline to the 12 Months follow-up (nSTRIDE APS: -41.1%; Synvisc-One: -40.4%). Secondary and exploratory objectives suggested that treatment with either nSTRIDE APS or Synvisc-One were effective in reducing pain and improving function up to 12 Months, although no greater improvement of nSTRIDE APS over Synvisc-One at 12 Months post-single injection was demonstrated. Most parameters measured (i.e. WOMAC Pain, Function or Total, NRS Pain, EQ-5D) suggested a significant improvement in function and reduction of pain after a single injection that lasted up to 12 Months. No major changes in joint morphology were observed through injection of either nSTRIDE APS or Synvisc-One. The safety profiles of both treatments were similar and device-related events included mild arthralgia, joint stiffness, joint swelling and musculoskeletal discomfort, especially after the injection. The average time to AE resolution for device-related events was 25.5 days. No unexpected safety events were reported in both groups, indicating a similar safety profile of both treatments.Long-term (12 through 48 Months) follow-up results indicate a continued improved function and decreased pain when compared to baseline, independent on the choice of a second injection. Both treatments induced an improvement in WOMAC Pain and Function score until at least 48-Month follow-up. Notably, the percentage of subjects achieving initial treatment success (subjects who received the same or no second injection, no additional intervention, and >20% improvement in WOMAC Pain score) was significantly higher in the nSTRIDE APS-treated subjects compared with the Synvisc-One-treated subjects (80.6 vs 62.7%). Additionally, both treatments showed comparable treatment success (>20% improvement in WOMAC Pain from baseline) and injection success (subjects who did not receive a second injection or had no additional intervention and showed >20% in WOMAC Pain scores) whether subjects crossed over or were treated with the same second injection. According to results on the number of subjects who received additional intervention for OA in their index knee and the OMERACT-OARSI, EQ-5D, NRS, and WOMAC Pain or Function scores, pain and function similarly improved with both treatments. In line with this, the majority of subjects in both treatment groups chose for a similar second injection as the one they were originally assigned to.
In the long-term safety analysis of this study (12 through 48 Months), 43 AEs and 15 SAEs were reported for all subjects. Most reported related AEs were excessive pain (n=6) and excessive swelling (n=5) of the index knee. Excessive pain (n=1), excessive swelling (n=1) and infection (n=1) of the index knee were reported as possible or probable related SAEs. Both (repeated) treatments resulted in relatively few device-related AEs and SAEs from 12 until 48 Months. Seven (7) out of the sixteen (16) device-related AEs occurred after an nSTRIDE APS injection, and nine (9) after a Synvisc-One injection. Moreover, four (4) out of the five (5) device-related SAEs happened after an nSTRIDE APS injection and one (1) after a Synvisc-One injection. Thus, overall, there were no considerable differences in number and/or severity of reported AEs and SAEs between the two intra-articular treatments. Lastly, during this clinical study, one (1) death was reported in a subject that received Synvisc-One, but this SAE of urological origin (kidney failure leading to multi-organ failure) was unrelated to the procedure or device.
In conclusion, treatment with an intra-articular nSTRIDE APS or Synvisc-One injection was considered safe at both short- and long-term follow-up and both treatments were effective in reducing pain and improving function at least up to 48 Months post-injection.
REC name
West of Scotland REC 4
REC reference
17/WS/0088
Date of REC Opinion
20 Jun 2017
REC opinion
Further Information Favourable Opinion