The images point to a progression of the process seen on the previously discussed specimen. The amount of HD-like cells with the same phenotype (CD30+,CD79a+)is much higher, forming cohesive sheaths, an indication for full blown high grade transformation of the marginal cell lymphoma, as diagnosed in the previous specimen (see final comment by Prof. Müller-Hermelink on ID131893).  
I cannot offer an explanaition for the T-clonality.

From the pictures now, I think that this is indeed an AILT:  
B-cell rims, large B-blasts, FDC prolifates, lots of vessels, CD4>>CD8, Hodgkin-like cells.  
 
AILT can transform to DLBCL or contain a lot of EBV-infected Hodgkin-like cells. But how does the MZL fit with it...? Two concommittant lymphomas or transfromation? I don^t know, there are several possible speculations which lymphoma most probably led to the next, but all these speculation would probably not help to resolve the actal problems of the patient, since actually AILT seems to be the urgent problem. AILT would also explain the presence of both B- and T-cell clonality as well as the morphology.

Thank you. CD23+ networks and CD10+ population support an AILT idea. But definitelly nodular (HL NS like) architecture, CD79a+;CD30+;CD20/CD3-; giant cell IH phenotype??? For sure I will repeat clonality on this case. The "simple" transformation of MzL to high grade B lymphoma (DLBCL) would be more convenient?

I favour the concept of angioimmunoblastic lymphadenopathy - both morphology and immunophenotype can fit and also one can see both B- and T-cell clones, see J Exp Med. 2001 Oct 1;194(7):927-40. Survival and clonal expansion of mutating "forbidden" (immunoglobulin receptor-deficient) epstein-barr virus-infected b cells in angioimmunoblastic t  
cell lymphoma. Bräuninger A et al.

PCR T and B clonality studies are in progress. A lot of giant cells with B phenotype are stricking for AILT or even transformation to High grade B lymphoma. IN primary case these giant cells were single...

I agree that the images shown here are most consistent with angioimmunoblastic T-cell lymphoma based on the overlap of CD3+/CD10+ cells and FDC pattern. Would do Bcl-6 and CD57 just so to see if you can demonstrate it in the T-cells. Large B-cells would fit and would be expected to be EBV+. However, where did the marginal zone lymphoma go? Looking at the previous biopsy, I was convinced that there was a component of marginal zone lymphoma with prominent plasmacytic differentiation. This would be further supported by what was referred to as "unusual CD68-positivity", which is actually one of the frequent findings in both benign and malignant marginal zone cells. Alternatively, the CD68+ plasmacytoid cells that were CD20-/CD3- were plasmacytoid dendritic cells, and this immunophenotype would be perfect for them. Yet another possibility is that she had benign plasmacytosis, which may be associated with hypergammaglobulinemia in some angioimmunoblastic T-cell lymphomas with high IL-6. I am not sure that I understand this case.

Thanx. Bcl6 cells are numerous in nodules. I'm not quite sure are ther CD3+, but most probably. CD57 will appended later.

Please find Bcl2 (T population +/+++; single giant cells +) and CD57 (10% of T population).

HI everybody,  
I won`t say that I understand that case either, but albeit you mentioned that Immunophenotype and the Molecular results would perfect fit into AILT, I am not convinced about that diagnosis. I really don`t see the morphology of AILT (High endothelial venule, arborisation, atypical (clear) cells. Also the CD23 cells represent in my eyes colonized follicels rather than proliferating FDCs.) In MZL one can observe excessive CD10 positivity. Unfortunately this doesn`t explain the TCRß positivity. Have you done EBV? My suggestion is transformation of MZL....

I agree with Marcus Kremers comment, I find it difficult to find morphologic convincing features to support AILT. Another feature, unusual for AILT is the prominent sinusoidal spread of large CD30+ cells, as well as the intravascular component. These features are rather consistent with high grade transformation of a low grade lymphoma. Of course this argumentation will not explain the T-clonality.

I have seen limited number of cases, but it seemed from CD10 vs. CD20 that it should be angioimmunoblastic T-cell lymphoma. Regarding FDCs, I think it is OK how they look, while there is some variation between the cases I have seen (see also 1: Am J Clin Pathol. 2002 Dec;118(6):848-54. and 2: Br J Haematol. 2003 Jun;121(5):681-91.). I cannot tell from the CD57 image if CD10 cells are also CD57+. That would help, too.

I still think that the case now best fits to AILT:  
1. small B-cell rims (subcapsular, perisinusoidal)  
2. CD23+ proliferations ouside form B-cell areas  
3. CD10+ cells outside from follicles and B-cell areas  
4. presence of high amounts of CD57+ cells  
5. presence of CD79a+/CD30+/CD15- giant HRSC-like cells  
6. T-cell clonality  
7. CD4 excess over CD8.  
I do not know any other entity that may have all the mentioned features. One question is still open, where did the MZL got lost?

The first biopsy in my eyes clearly shows a marginal cell lymphoma with additional large RS-like B-cells and nice geographic pattern and dot-like CD68 positivity, as well as marked secretory diferentiation. To me, the new biopsy represents progression to a large B-cell lymphoma with anaplastic features, partial loss of B-cell antigens and CD30 expression. I would not diagnose classical HL here, nor do I find enough evidence, morphological or phenotypical, for a diagnosis of AILT, despite the T-cell clonality. THe CD10 positivity as well as the FDC proliferation still would fit with the transformed MZL for me. The absence of EBV in the large cells is a further argument against a B-cell proliferation in AILT. I speculate that the increase in T-cells is a reaction towards the large transformed cells. Whether the T-cell clone detected by PCR is truly malignant remains to be shown - a composite neoplasm is certainly a possibility. Anyway, this is a really challenging case not easy to be solved conclusively.

since i was also asked by nina to study this indeed challenging case i would like to fully support the interpretation of falko fend (and i have studied the case, made my mind and only then read all the comments).  
this is in all likelyhood transformation of a mzl in a (anaplastic) dlbcl.  
criteria for ail, other t-nhl or hl are not fullfilled; the meaning of the t-cell clonality remains to be seen.

Thank you a lot. After verbal comunication TCRB/TCRG clonality was confirmed once more. IGH/IGK polyclonal. I will append data a little bit later. So I stay on biclone hypothesis. AILT or T NHL NOS question still remains, but at least some features of AILT are retained.

Please find T clonality and B polyclonality results in ppt file.

Sorry I doubt that we will ever be able to solve this case in an unanimous way. I still share the opinion of Marcus, Falko and Stephan of MZL in progression to large B-cell lymphoma.  
 
But in fact we do not know the correct drawer for this case.

Nina Hurwitz asked me to look again at this case. I believe that , if we consuider the morphology at the present biopsy we should not forget the previous biopsy , which ,in my view , was rather staight-forward a marginal zone lymphoma with massive plasmacytoid differentiation and kappa light chain restriction.  
In the present biopsy I don't see diagnostic features of AILT, in particular , there are no capsular invasion , there are no epithelioid venule proliferation and other features mentioned by Markus Kremer- rather I see still kappa light chain restriction with HD-like cells that exhibit a more T-cell dominated background ,partially within colonized follicles partially independent.I believe that it represents , as previously discussed , a HD-like transformation of this lymphoma, especially if ( as stated earlier) EBV is negative.I don't undestand the clonality studies: T cell receptor -Beta chain clonality may be found in B cell lymphomas . The polyclonal results in FR1- JH may be technically caused. At least the primary biopsy should show a clonal B cell receptor reaarangement that should be compared to the new biopsy.