Difficult case in a young individual.
I agree completely with the previous comments from Tim. Dasatinib is probably the best choice here and regardless of the potential for pleural effusion, this can be monitored with close observation and managed if it ever develops. Although bosutinib may be an option, it too can have issues with liver enzyme abnormalities early and this is the reason in my mind to make it a second choice.
It is unfortunate about the nilotinib. Even at 300 bid which is the dose for newly diagnosed patients and not salvage such as this case, the liver toxicity is not acceptable. A switch was in order after the imatinib failure, but there is no reason to expect any of the second generation drugs are better than the others in the scenario, unless the very unlikely case where a kinase mutation is identified.
Again as pointed out, reporting this patient's results on the international scale would make it easier for outsiders to offer an opinion.
I think as well that identifying a related or unrelated donor is important. If dasatinib fails for whatever reason, response or toxicity, in a young individual like this, I believe that an allograft is the next line of therapy.

I find it difficult to interpret the BCR-ABL levels because they are not on international scale. Was there a big drop from diagnosis to the 3 month time point. We have recently found that this drop is highly predictive of the risk of treatment failure and progression.

Based on the cytogenetic result at 4 months - 25% Ph+ suggests that response is suboptimal but maybe not failure. I think a switch to nilotinib was appropriate at that stage. Unfortunately you have run into problems with hepatotoxicity which is quite significant. Your options are:

1. (FAVOURED OPTION) Switch to dasatinib as soon as the liver enzymes have settled. You rarely get cross-intolerance in this situation. That is assuming that the hepatoxicity is nilotinib related which seems likely. You still have a reasonable chance of salvaging this patient and achieving a deep durable molecular response. The risk of a pleural effusion with dasatinib in the younger patients (less than 40) is quite low - less than 5% in one series.
2. Gradually re-introduce nilotinib once the enzymes return to near normal but you may have further interruptions which you cannot afford at this critical stage.
3. Bosutinib would also be a possibility, but early GI toxicity can be challenging and frequently leads to interruptions.
4. Another option would be switching to ponatinib but I think the risk profile in terms of serious vascular events would not justify this approach, unless this patient had also failed to respond to dasatinib.

A mutation screen is clearly important but not very likely to be positive, so I wouldn't wait for this result before switching to dasatinib - I would switch as soon as the liver toxicity allows.

In this situation we would normally check the BCR-ABL level monthly until it is below 10% and then 3 monthly to confirm a steady fall over time.

Clearly this patient is in the high risk category and it would be sensible to go ahead and find the best donor while you are trying to achieve better control because this may rapidly become the best option.

This is an interesting and challenging case.

35 yr old male, presented 6 months ago with 1-2 months of sweats, spleen to level of umbilicus, WBC 160,000 with typical CML blood smear and 4% basophils, no blasts, bcr-abl quant diagnostic. No baseline marrow, started on imatinib 400 mg daily, immediate clinical and hematologic remission (modest pancytopenia), but after 4 months bcr-abl had plateaued at only about 1,000 bcr-abl transcripts per 100,000 control transcripts. Marrow then done, 10% cellular, 5 of 20 Philadelphia chromosomes, no other cytogenetic anomaly.

Switched to nilotinib 300 mg BID, QT interval fine, but now after 6 weeks’ treatment, serum AST and ALT 1.5 and 2.0 times upper normal limit, drug held, and 4 days later AST and ALT now 2.5 and 3.5 times upper normal limit. And the bcr-abl quantitation was 9,200 at 3 weeks and still 5,800 now. Our reference lab could not make a determination on abl mutational status; that has to be resent.

Any thoughts or help would be much appreciated. Thank you!