GROWTH PATTERNS OF DIFFUSE NON-HODGKIN’S LYMPHOMAS ESTIMATED FROM MITOTIC AND APOPTOTIC INDICES Donatella SPINA 1 , Lorenzo LEONCINI 1 , Tiziana MEGHA 1 , Marcella GALLORINI 1 , Piero T OSI 1 *, Stefano PILERI 2 , Rainer KRAFT 3 , Jean A. LAISSUE 3 and Hans COTTIER 3 1 Institute of Pathologic Anatomy and Histology, University of Siena, Siena, Italy 2 Institute of Hematology ‘‘L. and A. Sera `gnoli’’, Second Service of Pathologic Anatomy, Hemoloymphopathology Unit, University of Bologna, Bologna, Italy 3 Institute of Pathology, University of Berne, Berne, Switzerland Growth rates of neoplasms could be calculated only on the basis of mitotic and apoptotic indices (MI and AI, respec- tively), assessed on tissue sections, if the duration of mitosis and apoptosis (T m and T a , respectively) in vivo were known. For humans, this is practically never the case. W hat use then can be made of MI and AI to arrive at a relative, crude estimate of the state of growth?As a model system to study this problem, we chose diffusely growing stage I  II non- H odgkin’s lymphomas (dN H L, n  94). Cluster analysis re- vealed the existence of 3 highly distinct groups of dNHL (clusters I, II and III) in the MI vs. AI per case plot, with a roughly linear relation between both parameters. Most noso- logic entities defined by the REAL classification comprise cases that were represented in more than one cluster. W e adopted the simple formula GI (growth index)  XMI  AI, where X (  T a /T m ) remains to be evaluated. Based on the assumption that spontaneous regressions of dN H L are rare but do occur, we estimated that X  2 or, possibly, 3 are best fits for the pooled dN HLs studied. W ith the assumption of X  2, (i) 2MI  AI gave relatively lower valuesfor dN H L than proliferative indices such as %Ki-67  cells; (ii) values for 2MI/AI per cluster showed a pattern inverse to that for %bcl-2  cells; and (iii) a plot of 2MI  AI vs. 2MI/AI per case allowed the recognition, especially among N HLs with a low cell turnover, of cases where accumulation of presumably longer-lived cellsisan important factor in determining growth. Int. J. Cancer 73:178–183, 1997.  1997 Wiley-Liss, Inc. Neoplastic growth results from cell production minus cell loss (Steel, 1977). Although evaluation of proliferation alone has provided prognostic information independent of other histologic and clinical variables (Hall and Levison, 1990), important aspects of neoplastic growth will escape attention if we ignore the cell deletion factor. Even if the assessment of proliferative indices alone should ultimately prove sufficient for predicting response to therapy and clinical outcome of a given neoplasia, this should be documented first. In histologic sections, mitosis and apoptosis are the morphologi- cally recognizable phenomena that most closely reflect cell birth and individual cell death and, thus, cell production and cell deletion in numerical terms. In theory, mitotic rate (cell production rate) = MI/T m (Steel, 1977); MI corresponds to the mitotic index (Quinn and Wright, 1992), and T m (mitotic time) to the time period during which a mitosis can be identified as such in sections of tissue obtained ex vivo and immediately fixed. By analogy, the rate of cell deletion via apoptosis (cell deletion rate) = AI/T a ; AI is the apoptotic index of the respective cell population (Sarraf and Bowen, 1988; Del Vecchio et al., 1991), and T a (apoptotic time) defines the time during which the apoptotic process of a cell can be recognized in histologic sections by light microscopy. Thus: Actual growth rate (cell accretion rate) = MI/T m - AI/T a (1) While MI and AI can readily be registered in tissue sections, T m and T a of human neoplastic cell populations in vivo are most often unknown and may vary considerably among tumor types and from case to case. Without reliable data on T m and T a in each case, it is theoretically impossible to obtain solid information on absolute cell accretion rates of human neoplastic disorders in vivo [see equation (1)]. What use can then be made of MI and AI, assessed on histologic sections, to get more insight into neoplastic growth patterns in humans? In the present study, we tested the notion that crude but reasonable assumptions regarding the relative magnitude of T m and T a may improve our understanding of neoplastic growth patterns in vivo. Therefore, we substituted equation (1) by the simple formula GI = XMI - AI (2) where GI is a rough relative growth index, and X, i.e., T a /T m , remains to be estimated. In this context, the formula XMI/AI reflects the ratio between the cell birth and the cell death rates. As a model system, we chose diffuse, or diffusely growing, portions of non-Hodgkin’s lymphoma (dNHL) because they consist predomi- nantly of neoplastic cells and exhibit a quite homogeneous distribution of mitotic figures and apoptotic cells/bodies (Kraft et al., 1994). Further, dNHLs rarely undergo localized ischemic necrosis and offer other advantages (Del Vecchio et al., 1991). To reduce the chance of cell emigration and immigration, which may falsify MI and AI, especially in generalized dNHL, we restricted the selected cases to stage I + II (‘‘localized’’) disease according to the pathologic Ann Arbor classification (Carbone et al., 1971). Estimates for the value of X were based on the assumptions that most dNHLs are continuously growing neoplasias and that sponta- neous lymphoma regression (Krikorian et al., 1980) is a rare event. The fraction of Ki-67 + cells (Gerdes et al., 1993) also was assessed. In addition, we registered the percentages of cells immunoreactive for the bcl-2 protein, which counteracts apoptosis (Reed, 1994). MATERIAL AND METHODS Selection of cases and conventional histology Paraffin blocks containing pre-treatment lymph node biopsies from 94 patients with dNHL were collected from the files of our 3 institutes. Criteria for case selection included sufficient amounts of lymphoma tissue in the blocks, absence of extended tissue necrosis and appropriate clinical staging data (Moormeier et al., 1990). Biopsy specimens were immediately sliced and fixed in a neutral, buffered, 4% aqueous formaldehyde solution and processed for conventional histology (Leoncini et al., 1993). Diagnoses were made independently by 4 experienced pathologists. Final consen- sus was reached, with the knowledge of the results of immunohisto- Contract grant sponsor: Italian Ministry of University and Scientific Research; Contract grant sponsor: Beatrice Borer Foundation for Human Monoclonal Antibody Research. *Correspondence to: Institute of Pathologic Anatomy and Histology, University of Siena, Strada delle Scotte 6, I-53100 Siena, Italy. Fax: (0039) 577 263 235. Received 25 February 1997; Revised 28 May 1997 Int. J. Cancer: 73, 178–183 (1997)  1997 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer