ARCH Extension Data and What Happened After the Trial Ended

By
Published Updated Last reviewed
Clinical medical image for trials arch: ARCH Extension Data and What Happened After the Trial Ended

ARCH Extension Data and What Happened After the Trial Ended

At a glance

| Field | Detail | |---|---| | Trial name | ARCH (Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk) | | N | 4,093 postmenopausal women | | Intervention | Romosozumab 210 mg SC monthly × 12 months, then alendronate | | Comparator | Alendronate 70 mg oral weekly × 24 months | | Duration | 24 months (12-month anabolic phase + 12-month transition phase) | | Primary endpoint | New vertebral fractures at 24 months | | Key result | 48% relative risk reduction in vertebral fractures; 19% reduction in clinical fractures | | Cardiovascular signal | Serious cardiovascular adverse events 2.5% (romosozumab) vs 1.9% (alendronate), p = 0.07 |

Why "After the Trial Ended" Is the Right Question

Most clinical summaries of the ARCH trial stop at the 24-month headline: romosozumab superior to alendronate for vertebral and clinical fractures in postmenopausal women with severe osteoporosis. That framing is accurate but incomplete. Anabolic agents in osteoporosis present a distinctive challenge that antiresorptives do not: bone mass built during the anabolic window must be maintained by a subsequent agent, because anabolic effects reverse on discontinuation. So the durability question, whether the advantage lasts, is inseparable from the clinical value of the drug.

The ARCH trial enrolled women with postmenopausal osteoporosis and either a prior vertebral fracture or very low T-scores (femoral neck T-score <-2.5 with a non-vertebral fracture, or T-score <-3.0). This was not a moderate-risk population. Understanding what happened to these high-risk women after month 24 is therefore directly relevant to practice.

What the 24-Month ARCH Design Did and Did Not Show

The trial's 12-month active treatment phase was followed by a mandatory transition to open-label alendronate in both arms. This design was intentional. The investigators were testing the full sequence: anabolic priming with romosozumab then antiresorptive consolidation, versus antiresorptive therapy alone from the start. The published ARCH results therefore reflect a combined-phase outcome, and the 48% vertebral fracture reduction and 27% hip fracture reduction in the romosozumab-then-alendronate arm versus alendronate-alone arm represent the sequenced strategy, not romosozumab in isolation.

What the original trial could not answer:

  • Does the BMD advantage persist beyond month 24 as alendronate continues?
  • Does fracture risk reduction persist, or does it regress toward the alendronate-only trajectory over time?
  • Do the cardiovascular signals observed through 24 months stabilize or worsen with continued follow-up?
  • Is there a "legacy effect" analogous to what has been observed after denosumab or teriparatide discontinuation?

These questions require either a formal extension period or post-marketing observational data.

BMD Trajectories After Month 24: The Consolidation Evidence

Bone mineral density data from the ARCH trial showed that by month 12 (end of romosozumab treatment), lumbar spine BMD had increased approximately 13% from baseline in the romosozumab arm versus roughly 5% in the alendronate arm. At month 24, after both arms were receiving alendronate, lumbar spine BMD was approximately 15% above baseline in the romosozumab-then-alendronate arm versus 7.9% in the alendronate-alone arm. The gap had not closed. It had, if anything, widened slightly at the lumbar spine as alendronate continued to add incrementally to both groups but from different baselines.

The femoral neck data followed the same pattern: a roughly 6.9% versus 3.4% advantage at month 24 in favor of the sequential arm. Total hip BMD showed approximately 6.3% versus 2.9%.

This pattern is consistent with the theoretical expectation that alendronate, as an antiresorptive, locks in BMD gains made during the anabolic phase rather than dissipating them. It is also supported by the earlier FRAME trial sequence (romosozumab then denosumab), though FRAME used a placebo comparator rather than an active one. The ARCH design is more clinically relevant precisely because it tests the comparison clinicians face in practice.

The Durability of Fracture Protection: What Subsequent Analyses Revealed

The ARCH trial's fracture data at 24 months already embedded 12 months of alendronate follow-on in both arms. No formal long-term extension beyond 24 months was published for the ARCH population specifically. However, the FDA prescribing information for romosozumab (Evenity) approved in April 2019 incorporates ARCH data as the primary efficacy basis and acknowledges that durability beyond 12 months of romosozumab is contingent on sequential antiresorptive use.

Post-hoc analyses of the ARCH data have examined whether the fracture advantage was concentrated in the first 12 months or distributed across both years. The evidence suggests that while vertebral fracture protection was apparent by month 12 (the anabolic phase), clinical fracture protection, which includes hip and non-vertebral events, accumulated through month 24. This implies that the antiresorptive consolidation phase contributed meaningfully to the clinical fracture outcomes and not merely the BMD numbers.

Observational registry data from real-world osteoporosis cohorts, though not ARCH participants specifically, indicate that patients who complete 12 months of romosozumab and then transition to alendronate or denosumab maintain BMD gains for at least 24 to 36 additional months. Failure to transition, a clinically common scenario driven by cost, access, or patient preference, results in BMD loss that begins within 6 to 12 months of discontinuation, mirroring the pattern seen with teriparatide discontinuation in untreated patients.

Regression to Mean and the Ceiling Problem

A structural limitation of any 24-month antiresorptive consolidation analysis is that both arms of ARCH were receiving the same drug (alendronate) in the second year. Alendronate's fracture efficacy in postmenopausal osteoporosis has a relatively flat dose-response relationship at the established 70 mg weekly dose, and it tends to reduce bone turnover markers to similar troughs regardless of baseline BMD. This means that over a longer horizon, say five to ten years of continued alendronate, the BMD trajectories of the two original arms might converge, though the fracture histories would remain different.

This is not a trivial point. If fracture risk is partly a function of cumulative bone microarchitectural history and not just current BMD, then the structural repairs made during the romosozumab year may confer durable protection even as absolute BMD values converge. Micro-CT data from romosozumab studies in animal models and from iliac crest biopsy data in the FRAME trial support the idea that romosozumab improves trabecular number and connectivity, not just bone volume, and these microarchitectural improvements are not easily reversed by antiresorptive withdrawal. Whether this translates to a lasting fracture benefit in ARCH participants beyond what BMD tracking shows is an open question that would require longer follow-up data from this high-risk cohort.

Cardiovascular Safety: What Post-Trial Surveillance Added

The cardiovascular signal in ARCH is the most consequential clinical issue to emerge from this dataset, and it was not resolved at trial closure. The original ARCH publication reported serious cardiovascular adverse events (defined as cardiac ischemic events, cerebrovascular events, or cardiovascular death) in 2.5% of the romosozumab arm versus 1.9% of the alendronate arm. The p-value was 0.07, below conventional significance thresholds, but the absolute excess of approximately 0.6 percentage points in a single year was enough to prompt a boxed warning.

The FDA label carries a boxed warning stating that romosozumab may increase the risk of myocardial infarction, stroke, and cardiovascular death, and it contraindicates use in patients who have had an MI or stroke within the preceding year. This is a direct consequence of ARCH, not of other romosozumab trials. The FRAME trial, which compared romosozumab to placebo, did not show a cardiovascular imbalance, suggesting the signal in ARCH may reflect either a protective cardiovascular effect of alendronate (which was the comparator) rather than a harmful effect of romosozumab, or a chance imbalance. A meta-analysis published in the Journal of Bone and Mineral Research examined pooled data from romosozumab trials and found no statistically significant increase in major adverse cardiovascular events (MACE) across the program, adding nuance to the ARCH-specific observation.

Post-marketing pharmacovigilance from Japan, where romosozumab received approval in 2019, and from U.S. commercial use through 2022 has not identified a consistent cardiovascular signal in registry data, though observational studies in this area are methodologically limited by channeling bias: high-risk cardiovascular patients are selectively excluded from romosozumab prescribing precisely because of the boxed warning. The true post-marketing cardiovascular risk therefore remains difficult to characterize independently of the trial data.

Society guidelines have incorporated these findings carefully. The 2022 American Society for Bone and Mineral Research (ASBMR) clinical practice guide on sequential osteoporosis therapy recommends romosozumab as first-line anabolic therapy in patients at very high fracture risk, with explicit guidance to avoid use in patients with recent cardiovascular events, reflecting the ARCH signal without overstating it.

What Sequential Therapy Research After ARCH Taught Us

The ARCH trial contributed to a broader shift in thinking about osteoporosis treatment sequencing. Prior to ARCH and FRAME, the dominant clinical model was antiresorptive therapy first, with anabolic agents reserved for failures or very high-risk cases. ARCH, by directly comparing a sequence that starts with anabolic therapy against standard antiresorptive therapy, provided the first head-to-head active-controlled evidence that anabolic-first sequencing produces better outcomes than antiresorptive-first sequencing in the right patient.

Follow-up research has since examined what happens when the sequence is reversed: antiresorptive first, then anabolic. The DATA-Switch trial and subsequent analyses have shown that prior bisphosphonate use blunts the anabolic response to both teriparatide and romosozumab, particularly at the hip. This means ARCH's finding is not simply about the superiority of one drug over another in isolation; it is about the order of therapy and the biological state of bone at the time the anabolic agent is introduced. Post-ARCH prescribing guidance increasingly emphasizes identifying very high-risk patients early and initiating anabolic therapy before bisphosphonates, precisely because the window for maximum anabolic benefit may close once the osteoclast-osteoblast coupling has been suppressed.

Key Limitations Acknowledged by the ARCH Investigators

The authors of the original ARCH publication were explicit about several limitations that remain relevant to interpretation of follow-up data:

  1. The trial enrolled only postmenopausal women, so efficacy and safety in men or premenopausal women with secondary osteoporosis is not directly supported.
  2. Prior bisphosphonate use was an exclusion criterion. The real-world population most likely to receive romosozumab is, paradoxically, often bisphosphonate-experienced, because these are patients who have already experienced fractures on standard therapy.
  3. The active comparator design, while clinically meaningful, makes it impossible to know whether the cardiovascular imbalance reflects harm from romosozumab, protection from alendronate, or both.
  4. Follow-up beyond 24 months was not pre-specified, limiting conclusions about long-term durability or late safety signals.

Frequently asked questions

›Did the ARCH trial have a formal long-term extension period?
›Does the BMD advantage from romosozumab persist after switching to alendronate?
›What is the cardiovascular boxed warning on romosozumab based on?
›Did the FRAME trial show the same cardiovascular signal?
›What happens to bone if romosozumab is stopped without transitioning to another agent?
›Does prior bisphosphonate use affect the efficacy of romosozumab?
›Who was enrolled in ARCH and does that limit generalizability?
›How do current guidelines incorporate ARCH findings for sequencing decisions?
›Was hip fracture reduction statistically significant in ARCH?
›What is the clinical significance of the <1% absolute cardiovascular excess in ARCH?

References

  1. Saag KG, Petersen J, Brandi ML, et al. Romosozumab or alendronate for fracture prevention in women with osteoporosis. N Engl J Med. 2017;377(15):1417-1427. https://pubmed.ncbi.nlm.nih.gov/28892457/

  2. FDA prescribing information: Evenity (romosozumab-aqqg) injection. U.S. Food and Drug Administration. April 2019. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/761062s000lbl.pdf

  3. Eastell R, Rosen CJ, Black DM, et al. Pharmacological management of osteoporosis in postmenopausal women: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2019;104(5):1595-1622. https://pubmed.ncbi.nlm.nih.gov/30907953/

  4. Kendler DL, Marin F, Zerbini CAF, et al. Effects of teriparatide and risedronate on new fractures in post-menopausal women with severe osteoporosis (VERO): a multicentre, double-blind, double-dummy, randomised controlled trial. Lancet. 2018;391(10117):230-240. https://pubmed.ncbi.nlm.nih.gov/29129436/

  5. Shoback D, Rosen CJ, Black DM, Cheung AM, Murad MH, Eastell R. Pharmacological management of osteoporosis in postmenopausal women: an Endocrine Society guideline update. J Clin Endocrinol Metab. 2020;105(3):587-594. https://pubmed.ncbi.nlm.nih.gov/32068863/

  6. Lewiecki EM, Dinavahi RV, Lazaretti-Castro M, et al. One year of romosozumab followed by two years of denosumab maintains fracture risk reductions: results of the FRAME extension study. J Bone Miner Res. 2019;34(3):419-428. https://pubmed.ncbi.nlm.nih.gov/30508300/