Very-High-Risk Osteoporosis: Starting With an Anabolic or a Bisphosphonate
A single patient, newly diagnosed with severe osteoporosis after two vertebral fractures turn up on the same film. The disagreement isn't about which drug eventually gets used — both are on the table either way — it's about which one goes first.
Marguerite S., a 68-year-old woman, retired two years ago after four decades running her own tailoring shop, and still spends most afternoons at the sewing machine she moved into her spare bedroom rather than sell off. She came in for a persistent, dull ache low in her back that she'd assumed was just from bending over fabric all those years, and the lumbar-spine imaging ordered to work it up found something nobody was expecting: two vertebral compression fractures, at L1 and T12, one recent and one old enough that she has no memory of a fall or injury that could explain it. That second fracture is the one that matters most to what follows, because a compression fracture nobody noticed happening is a fracture that occurred under loads a normal vertebra tolerates. Her own DXA, ordered the same week, returned a lumbar spine T-score of −3.4 and a femoral neck of −2.9 — and read against the two fractures already on her films, that pairing is not simply "severe osteoporosis" but a named risk stratum: AACE and ESCEO define very-high-risk by a T-score at or below −3.0 or a prior vertebral fracture, and she satisfies both independently. She has never been treated for osteoporosis, never had a DXA before this visit, and takes nothing apart from an occasional antihistamine.
That treatment-naive status is the second thing her numbers establish, and it is doing more work here than it looks like it should. The reason drug order matters in this population is that prior bisphosphonate exposure blunts the bone-formation response when an anabolic is started afterward — Leder and colleagues' DATA-Switch work is where that effect was demonstrated directly — and no bisphosphonate has ever reached her skeleton. The blunting has not been set in motion in her. VERO put teriparatide head-to-head against risedronate specifically in patients with prevalent vertebral fractures, which is the population her films place her in rather than one she resembles at a distance, and found a larger reduction in new vertebral fractures on the anabolic; ARCH found the same directional pattern for romosozumab against alendronate. The older default — bisphosphonate first, anabolic held in reserve for failure — was built for patients being treated before anything broke. She is being treated after two things already did, and the sequencing question is genuinely open for her in a way it would not be for a woman who had already spent three years on alendronate before anyone thought to ask it.
Deciding what comes first, not just what's used
Start teriparatide, not alendronate. The order isn't a formality here — VERO put teriparatide head-to-head against risedronate in patients with prevalent vertebral fractures and found a meaningfully larger reduction in new vertebral fractures on the anabolic. ARCH found the same pattern for romosozumab against alendronate.
The mechanistic piece matters just as much as the outcome data: Leder and colleagues' DATA-Switch work showed that starting a bisphosphonate first measurably blunts the bone-formation response when an anabolic is added afterward. Going anabolic-first isn't just faster to a good outcome, it protects the anabolic's own effect from being blunted before it gets used.
I'm not contesting the trial data. What I'd flag is that teriparatide means a daily injection she has to self-administer for up to two years, a drug she's likely never budgeted for, and a level of monitoring most of my patients in her position don't sustain past the first few months.
That said — "most patients" isn't really the right comparison once you look at what actually defines her risk category. She isn't a typical new osteoporosis diagnosis being started on first-line therapy; she's already fractured twice before treatment began.
That's the piece worth making explicit: AACE and ESCEO's combined 2024–2026 consensus guidance defines very-high-risk specifically as a T-score at or below −3.0 with a prior vertebral fracture, and recommends anabolic-first sequencing for exactly that tier — not as a general preference across all osteoporosis patients, but as a stratified recommendation she meets by name.
The access and adherence concerns are real, and they're exactly why she needs a plan for the transition to a bisphosphonate once the anabolic course ends, not a reason to skip the anabolic phase for a patient whose own numbers are what triggered this recommendation in the first place.
Agreed: teriparatide for up to two years, with a bisphosphonate planned to follow rather than precede it, and a documented very-high-risk designation in the chart so the sequencing rationale doesn't get lost at a future visit with a different prescriber.
Not fully settled: how aggressively to plan for the injection-training and cost burden the primary care physician raised — a pharmacy-assistance referral was placed, but whether it will actually close the gap won't be known until the first refill cycle.