Chronic Myeloid Leukemia: Choosing a First TKI Around a Damaged Vascular Tree
A single patient newly diagnosed with chronic-phase CML, whose vascular tree has already failed once. Every drug that would treat his leukemia fastest carries the signal his femoral artery has the least room to absorb.
D.R., a 58-year-old man, ran a small HVAC business until a claudication flare-up last spring forced him to hire a second technician for the jobs that meant climbing onto roofs. That same spring workup found an 80% superficial femoral artery stenosis, treated with angioplasty; six months later a stress test ordered for unrelated fatigue turned up a white count of 68,000 with a leukoerythroblastic smear, and a bone marrow biopsy confirmed chronic-phase CML with a BCR-ABL1 transcript ratio of 42% on the international scale. His ELTS score, using his age, spleen size, platelet count, and blast percentage, places him in the intermediate-risk band — not the lowest-risk group where drug choice matters least, and not the highest-risk group where the case for the most potent agent is closest to unambiguous.
The choice of first TKI in CML is usually framed as a tradeoff between imatinib's slower, shallower response curve and a second-generation agent's faster path to deep molecular response — the response depth that eventually makes treatment-free remission a realistic future conversation. For D.R. that framing collides with a second body of evidence: ENESTnd's own decade-long follow-up found nilotinib's excess of arterial occlusive events over imatinib does not level off with time, it keeps accruing, and his one angioplasty already means his vascular tree has demonstrated it can fail. The two hazards are not the same shape, which is what makes them hard to weigh against each other. An inadequate molecular response is a problem that announces itself on a scheduled transcript check and can be answered by changing drugs. A non-plateauing arterial signal is a problem that announces itself as an event, in a vascular bed that has already had one.
Oncology clinic, treatment selection visit
Start with what his own vessels have already told us: an 80% SFA stenosis significant enough to need angioplasty, in a 58-year-old with a treated but real cardiovascular risk profile. ENESTnd's own investigators have published that nilotinib's arterial occlusive event excess over imatinib does not plateau through ten years of follow-up — it is a hazard that keeps compounding the longer he stays on the drug, and CML therapy in chronic phase is measured in years to decades, not months.
If his vascular exam were clean and his risk factors were only on paper, I would not be raising this as forcefully — a patient with no demonstrated vascular disease absorbs that risk very differently than one who has already had a vessel fail.
I'm not disputing the arterial data — it's real and it's his to weigh. But ELTS intermediate risk with a 42% baseline transcript burden is not a patient I want to commit to imatinib's response curve by default. IRIS's own long-term data shows a meaningful fraction of imatinib-treated patients plateau above MR4.5 and never reach the depth that makes a treatment-free-remission conversation possible ten or fifteen years from now, when he's still relatively young and might genuinely want off therapy.
Framing this purely as accruing arterial risk versus a one-time procedural risk understates what foreclosing future TFR eligibility actually costs a patient his age — it isn't a surrogate outcome, it's the difference between a life sentence on a drug and a real chance at stopping it.
You're both arguing as if the only real options are the drug with the cleanest vascular record and the drug with the deepest response — but bosutinib's phase 3 data in BFORE showed response rates and depth close to nilotinib's, without nilotinib's own specific arterial-occlusive signal in that trial's follow-up. Its real toxicity is different — diarrhea early on, transaminase elevation — not absent, but not the same vascular liability. It doesn't have imatinib's or nilotinib's length of real-world track record, so I won't pretend the long-term vascular data is as mature. But it's a genuine third option, not a forced choice between the two positions already on the table.
Agreed: imatinib 400mg daily, with BCR-ABL1 transcript monitoring at the standard 3-, 6-, and 12-month milestones rather than the shortened intervals sometimes used to justify an early switch on a second-generation agent's kinetics.
Not agreed, and left as an explicit branch point rather than resolved today:
Continue imatinib indefinitely; the vascular argument that justified starting it doesn't weaken just because the response looks adequate.
Switch to bosutinib rather than nilotinib, accepting its less mature long-term vascular data as the more reasonable trade against his known disease.
The hematologist-oncologist's preference for nilotinib — the deeper cytogenetic response, if his vascular history weren't in the picture — doesn't disappear because today's decision went the other way. If a future angiogram someday shows a genuinely stable vascular tree with no further events, that preference has real standing to be revisited; nothing about today's reasoning closes the door permanently, it just doesn't open it now.