Morphology and cytochemistry of acute lymphoblastic leukaemia.

ALL is characterized by small to medium sized leukaemic blasts with a rather low grade of cell-to-cell variability. The nucleocytoplasmic ratio is high with just a small cytoplasmic rim in many cases. The cytoplasm tends to be moderately basophilic. Usually, though not in each instance, it is agranular and free of vacuoles. The chromatin is more condensed than in AML and the nucleoli tend to be indistinct. The FAB classification of haematological malignancies separates ALL into three categories: ALL L1, L2, and L3. However, just the identification of the L3 variant is of major importance. The L3 cells are medium sized to large and are characterized by intensely basophilic and moderately abundant cytoplasm with prominent cytoplasmic vacuolation in the bone marrow but not necessarily in the peripheral blood. According to our experience there is a high but not universal correlation of the L3 phenotype as defined by morphology with the immunologically defined B-ALL with surface expression of immunoglobulins. Until recently, acute leukaemias proving negative for all cytochemical tests especially for the PAS reaction and for the focal type of acid phosphatase, were termed 'acute undifferentiated leukaemia' (AUL). However, this morphological/cytochemical diagnosis may be confused with the immunological diagnosis of unclassifiable leukaemia. Since almost any of these cases can be recognized as ALL or AML by immunology, the term AUL should be reserved for cases which can be classified neither by morphology/cytochemistry nor by immunology. The morphological and cytochemical distinction of ALL from poorly differentiated AML remains a problem, especially if the FAB criteria for distinguishing ALL from AML by cytochemistry (3% of the blasts positive for peroxidase) are applied rigidly. A small but significant percentage of poorly differentiated leukaemias have less than 3% of the blasts positive for peroxidase and the myeloid nature of the leukaemia can be identified by cytochemistry. The absence of blasts being positive for peroxidase is no reliable indicator for the lymphatic nature of a leukaemia, even if the PAS reaction is typical for ALL. The morphological diagnosis of ALL needs confirmation by immunology in each instance.