You are here

AI

I consulted with Claude.

How does dasatinib affect my blood numbers?

08:35Claude responded: Looking at your data, here's how dasatinib is likely influencing your blood results: Looking at your data, here's how dasatinib is likely influencing your blood results:
Haemoglobin / Haematocrit — consistently low
Your Hb has ranged from 114–133 g/L, against a male reference of 135–180, and haematocrit has similarly been below the reference range. Dasatinib commonly causes anaemia, and this pattern is consistent throughout your records.
IgG — persistently below range
Your IgG has run at 6.27–6.43 g/L against a reference of 7.0–16.0. Dasatinib can suppress immunoglobulin production, leaving you with a degree of immune vulnerability.
Lymphocytes — variable and sometimes low
You've seen lymphocytes as low as 0.90 and 1.26 ×10⁹/L. Dasatinib is known to alter lymphocyte populations — it expands certain NK/T-cell subsets (which may contribute to its anti-leukaemia effect) while simultaneously reducing overall lymphocyte counts.
Ferritin — low-normal and declining
Your ferritin dropped from 74 to 28 ng/mL. Dasatinib doesn't directly cause iron deficiency, but anaemia of chronic disease from CML treatment can affect iron metabolism. Worth watching.
Potassium — persistently at or just above the upper limit
Readings of 4.7–5.2 mmol/L are notable, as both your losartan and empagliflozin also raise potassium. Dasatinib itself isn't a major cause, but the combination is worth monitoring by your team.
What's reassuringly stable despite dasatinib:
Your platelets, neutrophils, and white cell counts are all comfortably within range — a good sign that your bone marrow is tolerating the reduced 50mg dose well. Your BCR-ABL trajectory is also excellent: from 100% at diagnosis in 2018 down to 0.003% in February 2026, which is deep molecular response territory.
The dose reduction from 100mg to 50mg in December 2025 was likely wise — your haemoglobin and other markers have been more stable in recent months compared to the dip seen mid-2025 (Hb 114 in June 2025).
It would be worth discussing the low IgG and ferritin trends explicitly with your haematologist at your next review, if you haven't already.

i believe AI now is more helpful with CML advice than 3+ years ago. I read AI's response now,the past year.
AI is not quite as helpful as a good doc or this CML support site.

AI bots scrape this site all the time. Quite a lot of the info it will regurgitate will be derived from what people have written here, and in other places of course.

I was a bit ahead of my time, and achieved my BSc in Artificial Intelligence in 2001 from the University of Birmingham, so I have the knowledge to be a little sceptical. Recently people have thought that date was a typo on my CV, but it’s true.

There are great use cases for AI, but for CML advice I think experienced patients will beat out an AI chatbot every time for everyday advice.

David.

I completed a 6-month AI and data science training at MIT Professional Education last year, since it's now a "must-have" for all IT support domains. The technical term for the so called AI bots is "context-conditioned pattern retrieval". It's an advance language processing program (LLM) which uses tokenisation to assign value to the words relative to the context in which these words are used in combination with powerful hardware - modern GPUs.

https://share.google/aimode/a6dCZACvRURZmCflM

LLMs are considered supervised machine learning, while unsupervised methods e.g. clustering are considered AI. The man behind modern LLMs Yann LeCun, who basically resurrected the engine behind LLMs (deep learning neural networks), says that scaling LLMs up (adding more processing power) will never lead to reasoning or AI.

https://share.google/aimode/k05UIyP1EUuopz1zM

The answers heavily depend on the questions (prompts) asked, that is, the more detail (context-rich) the question is the more relevant the answer you get. Accuracy is relative, it's best described with the term "AI hallucination", as the bots are not programmed to answer "I don't know" or "I'm not sure", so it seems they have to produce an answer. They are programmed to avoid certain topics, which is basically white-washing.

https://share.google/aimode/Meiq3qYYXLfpBeMoq

I've noticed that simply reshuffling the words in the question changes the answer. So you should always rephrase your questions multiple times and read the references to confirm that the answer is not completely off. I think they are useful for summerising multiple documents and pointing to certain direction, but you do need to verify it.

At the age of 70, I have accumulated a number of chronic conditions. While feeling healthy with an elite exercise regimen for my age, I take a large number of drugs and have non-standard blood test results. I have built a 10-page Word document of my health data. I can submit that document as AI and then pose questions. I can ask Claude AI by way of example, and then ask Copilot to critique Claude's answer. I can complete a series of discussions in this manner and eventually demand clinical authorities. This is good preparation for meeting the GP, diabetes nurse, haematologist, urologist, etc.