Hip Osteoarthritis (OA) is one of the most common diseases affecting the ageing population. No specific and safe treatment slowing the development of hip OA as reflected in reduced joint space narrowing has been developed so far. Implantation of hip prosthesis has been a great success, but is expensive and puts a great financial pressure on hospitals and national healthcare programs. It also poses risks of serious complications. Current non-surgical treatment options focus on pain management and physiotherapy.

Bisphosphonates have been shown to exert multiple beneficial effects in OA. Animal studies have demonstrated that bisphosphonates protect against subchondral bone degradation and preserve articular cartilage in mice and rats with OA.

In this study 70 participants will be treated with either Zoledronic Acid (a potent Bisphosphonate) or Placebo (Saltwater). Allocation will be random and both parties are blinded. The investigators hypothesis is that participants treated with Zoledronic acid will have less pain, and better function after 12 months. The results will be tested with functional and pain scores, and functional tests. The study team will also measure the effects on disease related quality of life, bone quality measurements and the effect on bone marrow lesions on MRI (magnetic resonance imaging) and histological examination

Background Hip Osteoarthritis (OA) is one of the most common diseases affecting the ageing population. No specific and safe treatment slowing the development of hip OA as reflected in reduced joint space narrowing has been developed so far. Implantation of hip prosthesis has been a great success, but is expensive and puts a great financial pressure on hospitals and national healthcare programs. It also poses risks of serious complications. Current non-surgical treatment options focus on pain management and physiotherapy.

Bisphosphonates have been shown to exert multiple beneficial effects in OA, not only through effects on subchondral bone, but also through chondroprotection and possibly by inhibiting synovitis. Animal studies have demonstrated that bisphosphonates protect against subchondral bone degradation and preserve articular cartilage in mice and rats with OA. Efforts to further improve subchondral bone using a combination of zoledronic acid (ZA) and parathyroid hormone (PTH), was, however, shown to be ineffective.

Rationale for study Effects of bisphosphonates in OA

Bisphosphonates may target Pain and OA development by several different mechanisms:

Anti-osteoclastic action:

Increased bone turnover in OA causes bone loss and osteoporosis of subchondral bone beneath the thickened subchondral plate, which alters flexibility, and increases the risk of microfracture, which may cause of bone marrow lesions (BMLs) in OA. Bisphosphonates may also increase the strength of subchondral bone by reducing turnover and improve bone material properties.

Osteoclasts mediate the extension of channels from marrow spaces into the non-calcified articular cartilage. This leads to loss of integrity at the osteochondral junction exposing subchondral nerves to proinflammatory and pain inducing factors from the synovial fluid. This action has also been shown to induce sensory nerve growth into the non-calcified articular cartilage.

Osteoclasts may also reduce pH at the osteochondral junction, thereby sensitizing and activating sensory nerves through actions on ion channels on their peripheral terminals.

Effects on synovitis Synovitis is an integral part of OA pathogenesis. Multinucleated giant cells have been reported in synovitis, these cells are targets for bisphosphonates as are osteoclasts, and may be responsible for the positive effects noted in synovitis. Indeed bisphosphonates have been shown to inhibit macrophage activation in the synovium and reduce synovitis Effects on Bone marrow lesions The formation of bone marrow edema (BME, Bone marrow lesions) is part of OA pathogenesis, and is associated with progressive disease and disability. Several studies have demonstrated that increased remodeling activity in the subchondral region is associated with increased cartilage lesions and the investigators own studies supported by a grant from HSØ (South-East Health Region of Norway), have demonstrated that the presence of increased remodeling activity and vascularization of bone marrow lesions also show association with cartilage degradation. Bisphosphonates may target both the turnover increase, but also angiogenesis as bisphosphonates, and ZA in particular; exert pronounced anti-angiogenic effects.

Direct effects on chondrocytes Chondroprotection with less fibrillation and ulceration has been demonstrated for ZA in an OA models in rabbits and dogs and it has been hypothesized that bisphosphonates may exert an anabolic effect on chondrocytes. In humans, several bisphosphonates have been shown to reduce cartilage degradation as reflected in the biomarker CTXII (C-terminal telopeptides of type 2 collagen).

Clinical trials testing bisphosphonates in OA Bisphosphonates have been tested prospectively as treatment modalities in previous human clinical trials with equivocal results. One trial tested the effects of a less potent bisphosphonate (risedronate) with 4 different doses (5 and 15 mg daily, 35 and 50 mg/week) in knee OA. Joint space width in the placebo group decreased non-significantly by 8 mm vs. 4 and 1mm in the 5 and 15 mg groups, respectively and pain scores did not show significant reductions. Furthermore, serendipitously, the placebo group in the Risedronate trial was very healthy, as it did not exhibit worsening of symptoms over the trial period. This may further have jeopardized the trial outcome. Markers of cartilage degradation, however, revealed significant reductions. In a Japanese study looking at the effects of another oral bisphosphonate, Alendronate, however, Nishii et al. demonstrated significant reductions in VAS and WOMAC pain scores for up to 24 months.

The lack of efficacy of oral bisphosphonates in OA have been primarily ascribed to their poor absorption resulting in lower Cmax after administration (24). Therefore, more recent trials have focused on iv administration, where Cmax will be much higher. Varenna et al randomized 64 patients with OA and demonstrable BML's on MRI to receive either 4 iv. doses (100 mg per infusion of the bisphosphonate Neridronate or placebo. They demonstrated significant reductions in pain, reduced use of analgesics and improved QoL after 50 days. Similar improvements have been reported 5 weeks after intraarticular administration of the bisphosphonate clodronate. Data from the Osteoarthritis initiative demonstrated that patients treated with the iv bisphosphonate zoledronic acid for osteoporosis revealed significant reduction in OA pain scores for up to 3 years. In a later study, Laslett et al demonstrated significant effects on bone marrow edema size and pain in OA over a 1 year period. The doses given in the intravenous trials have been much higher than in the trials using oral administration.

Rationale for dosage and placebo It seems that the dose required for effect on OA should be higher than that which is used for treatment of osteoporosis shown by Varenna et al. Therefore, in this study the investigators choose to treat with two doses of ZA.

As there is no curable treatment for OA, best standard of care is alleviation through paracetamol, NSAIDs or in some cases opioids. Many patients already use one or more of these when they are referred to surgical evaluation. One of the endpoints of this study is to register use of these medications before and after treatment. Using paracetamol instead of placebo could potentially lead to overuse as it can be bought over the counter. Contraindication to NSAID's is common in the elderly population and could exclude many participants.

Simulating the active phase reaction commonly seen with ZA infusion with an active placebo is difficult. However, administration of prednisolone and NSAID's should minimize symptoms.

The investigators experience Based on the published experience with zoledronic acid in OA, the investigators administered two doses of IV. Zoledronic acid (5 mg) with a 3 month interval to patients with osteoporosis and OA of the hip and knee and demonstrable BML on MRI. The dosing regimen was based on previous studies reporting effects of zoledronic acid on bone marrow edema in transient osteoporosis. The patients achieved significant pain reduction as assessed by VAS within a period of 3 and 6 months. VAS decreased 85% from 6,8 (SD 1,8) at baseline to 1,0(0,9) at month 3 and 0,9(0,9) at month 6 (ANOVA p > 0,001). Among these were 3 patients scheduled for surgery. 80% of the patients have been followed for more than 1 year and still exhibit the same degree of pain reduction and increased mobility.

Risk versus benefits The risks of implant surgery are numerous, including infection, thrombosis, dislocation, fracture and in rare instances death. Therefore, a safe and effective conservative management can save the patient from unnecessary complications. The survivorship of hip implants has improved considerably over the last decades, but younger patients are at risk of needing one or more revisions during their lifetime. Revision surgeries are more prone to complications.

The goal of this treatment is to delay or prevent the need for implant surgery. To participate, patients will delay a possible operation for 12 months. Those that are included in the placebo group will not receive any active treatment and cannot expect any better function other than placebo effect. Patients can withdraw from the study at any time.

Although ZA is not approved for treatment of OA, it is regarded as safe for use in adults. The side effects of ZA are rarely serious. Known side effects such as atypical femur fracture and osteonecrosis of the jawbone is extremely rare (1/10.000-1/100.000). Acute phase reactions are common (30%), but usually mild to moderate and lasts 1-3 days. The symptoms can be alleviated with paracetamol, NSAID's or prednisolone.

Microindentation includes injection of local anesthetics and a needle puncture on the tibial bone. No serious adverse events have been recorded.

Magnetic resonance imaging poses no risks given that the participant has no contraindications to imaging such as metal or electronic implants sensitive to magnetic forces.

Inclusion Criteria:

• Age 50-80
• Continuous worsening of hip pain with an onset of minimum 3 months
• Hip Pain intensity >40mm on a visual analogue scale (VAS) ranging from 0 (no pain) to 100mm (maximal pain) during weight bearing for the last 7 days
• HHS < 65 (Harris Hip Score)
• Hip OA fulfilling the ACR (American College of Rheumatology) criteria [33]
• Hip MRI scan showing large (>1 cm diameter) Bone Marrow lesions (BMLs)
• Willing and able to consent and comply with the study protocol

Exclusion Criteria:

• eGFR (estimated glomerular filtration rate) < 35 ml/min or hypocalcemia
• Exposure to any treatment affecting bone other than Ca+D ( bisphosphonates, Denosumab teriparatide or continuous treatment of prednisolone)
• Diseases affecting bone and joints (i.e inflammatory joint diseases, osteonecrosis with segmental collapse and deformation, primary bone cancer or known skeletal metastases)
• Severe vitamin D deficiency (S-25(OH)D < 25 nmol/l) has to be supplemented with Ca+D before zoledronic acid infusion
• Ipsilateral knee prosthesis
• Contralateral hip prosthesis
• Women of child bearing potential (WOCBP). Female participants must be in a postmenopausal state or permanent sterile.
• Hypersensitivity or previous allergic reaction to ZA or other bisphosphonates.
• Hypersensitivity or previous allergic reaction to Calcigran Forte (a calcium and vitamine D supplement) or Prednisolone

No plan has been discussed within the study group yet

The treatment and placebo is administered through an infusion. Both will be administered in an identical infusion bag and are colorless. The mixing is performed by the same person who randomizes, less than 24 hours before administration. The bags are then put in a refrigerator and picked up later by a study nurse for administration. There is no "personal contact".