Mysterious changing Blood Type

[MAGNUM]

Two years ago an AML patient was a patient in our facility where he was subsequently typed as a O Positive. Actually he front typed as a O and reversed as an A. After performing a discrepancy workup that included testing at room temp, 37 and 4, the Anti-A corrected so that we could call the type as O Pos. Three subsequent crossmatches yielded the same results. Last night the patient returned after coming out of remission and presented for transfusion and now he is typing as a A+. The patient was recollected 3 other times with the same results, and another collection was sent to our BB reference lab for verification.

My question is WHAT HAPPENED? Malcolm any suggestions?:confused::confused:

[Malcolm Needs ☆]

Two years ago an AML patient was a patient in our facility where he was subsequently typed as a O Positive. Actually he front typed as a O and reversed as an A. After performing a discrepancy workup that included testing at room temp, 37 and 4, the Anti-A corrected so that we could call the type as O Pos. Three subsequent crossmatches yielded the same results. Last night the patient returned after coming out of remission and presented for transfusion and now he is typing as a A+. The patient was recollected 3 other times with the same results, and another collection was sent to our BB reference lab for verification.

My question is WHAT HAPPENED? Malcolm any suggestions?:confused::confused:

Well yes, actually!

ABO typing of individuals with leukaemia, particularly a leukaemia involving the myeloid cell line, often shows weakening of the antigens in relapse and, as in your case, can, on rare occasions, apparently "lose" their ABo antigens.

There is something on chromosome 9 (you'll have to forgive me for not knowing exactly what - I am neither a genetisist, nor a haematologist) which goes wrong in such circumstances that is very close to the ABO locus. This affects the production and/or enzyme kinetics of the A and B transferases that confer the "A-ness" and "B-ness" of red cell ABO antigens, and so such an individual can "lose" their ABO antigens.

I was not involved in this case myself, but I do know of a Haematology Consultant at one of the large London Teaching Hospitals who had AML, and the Transfusion Laboratory could tell whether she was in remission or in relapse by how "strong" her A antigen was when grouped.

I hope that helps.

Others may have a better explanation, or a better way of explaning what I have written above (or, indeed, dispute it). That is the joy of this site.

[David Saikin]

Did they get a bone marrow transplant? We had one of our surgeons develop leukemia. Her ABO status was in question (front type O, back type A). We transfused group O until she eventually got her ABO back to "normal" (ie, A).

[Malcolm Needs ☆]

Sorry David, are you asking MAGNUM or are you asking me??????

Sorry to be thick!!!!

[adiescast]

I believe the transplant question was directed at Magnum, as it would be my question also. Usually when we see that kind of typing discrepancy it is the result of a transplant (either BMT or HPC).

[MAGNUM]

No bone marrow transplant or transfusions othr than the O's at our place

[David Saikin]

Magnum . . .

Sorry David, are you asking MAGNUM or are you asking me??????

Sorry to be thick!!!!

[MAGNUM]

Thanks for everyone's replies, any new knowledge is helpful especially for an old fart.

[Thandar Aye-zheng]

Today I came across a case with AML patient in the lab. Patient was dx with AML when admitted in mid 2010 and typed as A pos. Pt has been multiply transfused with irradiated O pos RBC (and non-type specific platelets) throughout the year. 2 months ago patient forward typing was changed to O while the reversed typing remains A (I understand that patient was volume replaced due to repeated transfusion with O pos cells).

Just last week patient was re-admitted and both forward and reversed typing demonstrated O. Forward typing was AntiA-0, Anti B-0 at both Immediate Spin and 10 min RT incubation. Reverse typing was 1+s immediated spin, 2+s 10' incubation and 2+ at 37C, 10' incubation. Antibody screen was negative with Liss AHG 10'.

I also did a "mini cold panel" using patient serum O cord cell, O adult cells, A1 adult cells, A2 adult cells and auto control, at IS, 10'RT and 10' 4C. Everything turned up negative except A1 Adult cells. So I ruled out non specific cold agglutinins and cold auto antibodies.

I called the floor and asked whether patient had BM transplant at some point of his life and the answer was NO.

I understand that Antigens can be weaken and forward typing can change in oncology patients. What I dont understand is if the patient is intrinsically capable of producing ABH antibody while his antigens are weaken. Or Am I typing some "junk" as Anti A1? ?

[Thandar Aye-zheng]

I also did 4'C incubation for 3 screen cells and turned out negative. Therefore, I was able to rule out "cold" antibodies such as Lewis, M, N, P1. I am scrtching my head whether it could be Anti A1 or not.

[Malcolm Needs ☆]

Sounds very much like anti-A1 to me.

[Yanxia]

Sounds very much like anti-A1 to me.

me too.

[galvania]

But could the anti-A be coming from the group O platelets? I was not sure when she last had platelets, or what sort of platelets. Also, is the patient receiving large doses of IvIg? Anti-A could be coming from that too. How is the patient's DAT?

[Malcolm Needs ☆]

It's possible, but highly unlikely Anna. When you think about the amount of plasma that would be transfused with the platelets, one or more of the donors would have to have one heck of a high titre anti-A for it to show up after dilution in the patient's circulation (because it would be in the extravascular spaces, as well as in the circulation).

[Liz]

Sounds very much like anti-A1 to me.

If it is an anti-A1 are you sggesting that initially the patient is (was) A2?

[Malcolm Needs ☆]

Probably, but not necessarily. It could just be that the transferase enzyme has lost so much of its potency that the red cells now have far less A antigenicity than before. Although there is a qualitative difference between A1 and A2, as well as a quantitative difference, I have seen an "A1" produce an anti-A1 in such a situation.

There are also reports (one by Carolyn Giles) of individuals "losing" antigenicity [in her case Kp(] and the patient producing an apparent anti-Kpb.

[Liz]

But the immune system is already tolerant to the A1 as it would be in utero if it see it's (non-identical) twin's Blood group. What I am trying to say is that the RBCs don't have to keep expressing the A1 for the immune system to remain tolerant to it, or do they? As i see by your post Malcolm it is possible, so my question should is: what is the mechanism for this loss of "tolerance" ? How does the immune system "unlearn" what it once learnt?

[Malcolm Needs ☆]

Ah, there you would have to ask an immunologist who knows what they are talking about; I just know it happens, not how it happens!

Sorry.

[Malcolm Needs ☆]

I think that the key may be that the antibodies are auto-antibodies that appear to be alloantibodies. In Carolyn's case, the anti-Kpb disappeared as the Kp( antigen came back.

You see this kind of thing quite often in pregnancy with Lewis antigens disappearing and the lady making Lewis antibodies that then disappear after the birth, and the Lewis antibodies return to full strength.

[Thandar Aye-zheng]

Yes Malcolm, I had an Anti Lea positive woman in pre op work up for her C section while her prenatal screening was negative. So may be the same thing is happening to this patient. Yesterday, I also found out that patient never had remission since he started in our hospital. Patient was transferred to ICU yesterday. I am also suspecting platelet refractoriness in this patient (post tx count is lower than pre tx count by 1000 increment everyday, while he is receiving 1 unit per day)

Galvania- Yes patient has been receiving 1 platelet per day or every other day for a couple of week. We have been transfusing non type specific irr cmv neg. We tried our best to give A platelet since the patient typed A (even thou a couple of O might be tx). So chances of getting Anti A1 is very unlikely.

Liz – patient has been typing A pos since mid 2010 without any discrepancy in typing, until last month while the forward typing was changed to O. and 2 week ago I started seeing positive A1 cell reaction. So I don’t think the patient is subgroup of A. I am suggesting patient is developing anti A1 even thou he was initially typed as A.

I think patient blood type has changed to O from A. Even thou I have eliminated as much “cold” antibodies that may interfere with my reverse typing, my supervisor kept insisting that it was non specific cold and not to change the historical blood type. Being a fresh out of school blood banker, I do not have confident to argue with her.

[Thandar Aye-zheng]

I am sorry to find out patient is expired as soon as i got to work today

[Liz]

I am sorry to find out patient is expired as soon as i got to work today

I am sorry to hear that.