This is a very difficult case for whom a lot has been done and any possibility of a permanent solution is remote. The interventions should be basically palliative.
One should however not ignore a combination of imatinib, or any of the second generation tyrosine kinase inhibitors with ponatinib, asciminib or olverembatinib.
This can be given a thought as I have seen some such combinations make a difference.

I wish to follow up on the last comments and refocus the discussion. First of all, I would like to hear from anyone who believes they can still "cure" this individual. I do not mean "control" or "harness" or similar doublespeak. What then is the goal of these suggestions? It is never given with these. After the massive therapy given so far, the best scenario is hopefully counts controlled for a few weeks or months. This should be the goal here. I was reminded of a reference by a practical American colleague after a similar case discussed here a few months ago. If you are poker players, remember the song "The Gambler" by Kenny Rogers. "You have to know when to hold and know when to fold". This point has been reached. The only failure here would have been not trying and trying was done heroically. Guilt feelings should not drive futility.

This is an extremely challenging case.
First and foremost, Dr Lovisa Vennström and team have provided excellent care to this young woman over the years. Unfortunately, and thankfully rarely, CML can be an extremely difficult disease, especially with myeloid blast phase progression. This is the hardest to treat and there are no good or easy answers. Much depends on the patient’s goals of care in a non-curable situation. It sounds like the patient is willing to try “everything to have more treatment”.

Re: “Combine Olverembatinib and Asciminb? Start Azacytidine + Venetoclax and then DLI? Interferon in hope to find a GvL effect? Homoharringtonin? Other immunotherapy? “
1. In this non-curable multiply relapsed situation now with progression to blast phase it is difficult to make intensive treatment recommendations, but I don’t expect the combination of BCR::ABL1 inhibitors such as olverembatinib and asciminib will yield much of a response. If the patient has a good performance status, is seeking aggressive treatment and has proliferative disease then the most aggressive plan would be debulking with induction chemo therapy such as FLAG-ida +/- ven (ven could be considered rather than TKI).
a. Induction could then be followed by DLI to which I would try to add interferon. Many years ago at Fred Hutch DLI followed by interferon to optimize GVL in CML was used. The regimen of peg interferon after DLI was 90 mcg SC weekly for 6 weeks as tolerated.
i. If the patient has GVHD after initial DLI again then can consider therapy with AZA+ven. If the patient remains BCR::ABL1 dependent at all, then adding a TKI may be of some use. In my own practice I have used AZA+ven and added on ponatinib. One could substitute olverembatinib IF allowed by the access program. This combination leads to very prolonged cytopenias and I typically have had to limit venetoclax to 2 weeks per cycle (and sometimes less). In this scenario would start with AZA+ ven and then add olverembatinib if possible. Although the patient progressed on olverembatinib, she was on it for the shortest period of all TKIs- 4 weeks (it is a potent therapy and progression could have already been underway).
2. If a less aggressive approach is favored (may be possible with hypoproliferative disease), there are limited data on homoharringtonine in blast phase. PMCID: PMC5478443 reports a combination with imatinib in 3 blast phase pts and could consider combination with olverembatinib or other TKI here. The drug is not available in the US and I’m not sure if it is available abroad. In a less aggressive scenario would try AZA+ven followed by DLI and INF as above.
3. Only E255V was identified, suggesting that BCR::ABL1 independent mechanisms are driving relapse as the patient has been on therapies that ostensibly should treat that mutation. Do you have a recent extended myeloid mutation panel or cytogenetics? It may be helpful to know of any new additional targetable/actionable mutations.
4. There are CAR T trials and bispecific antibody trials, but none have robust data at this point, and I would not specifically recommend any. I did find a few studies in North America that did not explicitly exclude CML and (multiple) prior HCT, but this would need discussed with PI as most studies are enrolling R/R AML using ELN criteria and this may be a barrier. I can share with you if interested/feasible.
Please let me know if I can be of further help with this difficult case.
All the best
Vivian

Azacytidine + Venetoclax alongwith DLI and Olverembatinib should be the next option as TKI alone wouldn't be able to harness this aggressive disease

This is a very tragic situation. The physician has done absolutely everything they could for this patient, but the patient has relapsed again with blast phase CML following a second allograft. Although the patient is still very young, I believe the focus should be on optimising quality of life for the short time the patient has. I would discuss with the patient hydroxycarbamide and ven/aza as options. Given the extensive treatment the patient has had already, I think any benefit from ven/aza over hydroxycarbamide would be limited if there was any benefit at all, and it would result in more hospital visits and potentially more side effects. Like Professor Rosti, I think my preferred option would be hydroxycarbamide and involvement of the palliative care team to optimise symptom management and provide additional support.

I suggest olverembatinib 30mg QOD + AZA +/- Ven (14 days). I conducted a study in CML-MBP patients receiving this regimen. The response rate was 70%, even in those who failed olverembatinib monotherapy. Adding Ven will increase the response rate, but more BM suppression.