Five Years of Alendronate: Is It Time for a Drug Holiday
A single patient, five years into alendronate with no fracture, whose own femur X-ray for an unrelated complaint reopens the drug-holiday question. Nobody disputes she's a candidate — the argument is over how long the holiday should run.
Ines M., a 66-year-old woman, has taken alendronate every week for five years without missing more than a handful of doses.
She came in today for something else entirely: a hip and thigh X-ray after a minor car accident left her with soreness her orthopedist wanted to rule out for anything worse. The film was reassuring about the accident. It was the radiologist's secondary note that changed the visit — subtle cortical thickening along the lateral femoral shaft, not a fracture line, but the finding that gets a second look in a patient on long-term bisphosphonate therapy because of its association with atypical femur fracture. She has never fractured a bone from osteoporosis, and she had genuinely forgotten that the drug-holiday conversation was supposed to happen around now. Her numbers make the case for stopping and the case for uncertainty at the same time. A femoral neck T-score of −2.1 with a FRAX 10-year hip risk of 1.8 percent sits below the threshold at which continuing therapy is generally held to earn its cumulative risks — and the cortical thickening on her film is the visible form of exactly the cumulative risk that threshold exists to weigh. Nobody in the room disputes that she is a reasonable holiday candidate. What is actually in contention is duration. The ASBMR Task Force's guidance suggests a three-to-five-year holiday in this profile, on the reasoning that atypical femur fracture and jaw osteonecrosis risk climb with cumulative exposure while skeletal retention preserves some fracture protection after stopping. But FLEX, the trial that guidance descends from, randomized continued versus discontinued alendronate across a defined ten-year window; it did not test an open-ended holiday and did not measure how long one should run. What her numbers cannot supply is a duration, because the trial the three-to-five-year figure descends from never measured one.
How long a holiday actually lasts
This is a textbook holiday candidate by the ASBMR Task Force's own working recommendation — five years of oral bisphosphonate, moderate rather than high risk, no active fracture. Their 2016 guidance suggests a three-to-five-year holiday in exactly this profile, precisely because atypical femur fracture and jaw osteonecrosis risk climb with cumulative exposure while fracture protection persists for a real stretch after stopping.
I agree she's a candidate, and I'm not arguing against a holiday. What I'd push on is treating "three to five years" as a number to set and revisit later rather than a range to actively monitor within. Bisphosphonate skeletal retention varies meaningfully between patients, and a fixed-interval default can quietly run past the point her own bone density starts declining again.
The incidental cortical finding is exactly the kind of thing that argues for closer surveillance during the holiday, not less — not because it's a fracture, but because it's a reminder that her individual risk picture doesn't stop being individual once the holiday starts.
Worth being honest about where the "three to five years" figure actually comes from. FLEX, the trial underlying the guidance, followed continued versus discontinued alendronate for a defined ten-year window — it wasn't designed to test optimal holiday duration against fracture outcomes directly, and extrapolating a specific interval from it is a real limitation the Task Force's own guidance acknowledges, not a hidden flaw someone found later.
That doesn't mean the recommendation is wrong — it's still the best available reference point — it means the honest plan is a monitored holiday with a DXA and clinical reassessment at two years rather than treating five as a number either of you should feel obligated to hold to precisely.
Agreed: alendronate discontinued, a monitored holiday begun rather than an open-ended one, with DXA and clinical reassessment set at two years instead of waiting the full five the guidance range allows.
Not fully settled: what specific finding would trigger restarting bisphosphonate therapy before that two-year mark — a repeat femur film for the cortical finding, a fracture, or a defined amount of BMD decline were each raised as candidate triggers, without the group settling on one over the others today.