Proliferation kinetics of both leukemia and a variety of solid tumors have been assessed after in vivo infusions of the thymidine analogues, iododeoxyuridine (IUdR) and bromodeoxyuridine (BrdU). In acute myeloid leukemia (AML), these data indicate that the pretherapy cell cycle time (Tc) of myeloblasts is a prognostic indicator for remission duration since patients with slowly cycling myeloblasts had more durable remissions. The presence of in vivo differentiation detected from the day 7 biopsy after chemotherapy was also of favorable prognosis as these individuals had statistically significant improvement in their remission duration. The data in solid tumors are not mature enough for determining their clinical significance. Since cell kinetic information is readily available in a prompt fashion using these novel techniques, data can be used to plan therapeutic strategies for patients. This review discusses the state‐of‐the‐art techniques available for cell cycle kinetic studies and the clinical and prognostic utility of data that have been generated thus far. Cancer 1992; 69:1557‐1566.

Cell cycle kinetic analysis in vitro has conventionally been accomplished by labeling S-phase cells using two DNA specific labels given sequentially and separated from each other by a certain time interval. By counting the cells labeled by both versus those labeled by either one of the two labels, and using the formulas described by Wimber and Quastler, approximate values for durations of S-phase (Ts) and the total cell cycle (Tc) can be determined. More recently, instead of radioisotope labeled thymidine, two thymidine analogues have been used to label Sphase cells in vivo in a variety of human tumors based on the same principles. In the present report, new formulas are proposed for the calculation of Ts and Tc which are simpler to apply since only one type of labeled cells (those exiting Sphase as identified by containing only the first label) need to be differentiated from the remaining population for Tc calculations.Key terms: Thymidine analogues, cell cycle formulas, set theoryTo determine the rate of proliferation of cells by labeling S-phase cells sequentially with two varieties of radio-labeled thymidine, Wimber and Quastler (12) introduced their now classic equation, which has been extensively applied in cell cycle kinetic studies for approximately 30 years (5,11,14). We have also exploited this approach in studying large number of patients with acute myelocytic leukemia (AML). Our strategy required infusion of the first DNA-specific label bromodeoxyuridine (BrdUrd), in vivo and the second, tritiated thymidine label in vitro. By counting the number of leukemia cells which were singly labeled by either of the labels or doubly labeled, and employing the formulae described by Wimber and Quastler, we have been able to estimate the labeling index (LI), the duration of S-phase (Ts) and the total cell cycle time (Tc) of myeloblasts (8). Furthermore, we have determined the clinical relevance of these measurements by demonstrating a direct relationship between cell cycle times and duration of complete remission (7,9). With the rapid evolution of new techniques, better and more reliable probes are now available for detailed investigation of cell cycle kinetics. For example, instead of using the thymidine analogue bromodeoxyuridine in vivo and tritiated thymidine in vitro, we now routinely use iodo-deoxyuridine (IdUrd) and BrdUrd, both of which can be given in vivo. The actual counting of cells labeled by either one or both labels is an extremely tedious job, and because of the immunohistochemical techniques employed, it is not always possible to distinguish between a cell which contains BrdUrd alone or both BrdUrd and IdUrd. This problem is irrelevant in terms of results obtained since only IdUrd labeled cells need to be distinguished from all other labeled cells for the appropriate Ts calculations. However, attempts to improve upon the formula which would simplify the counting process lead us to the development of a reliable new mathematical equation which is being presented in this paper. This new equation can...

Expression of transforming growth factor-beta (TGF-beta), which inhibits the proliferation of hematopoietic progenitors, was investigated simultaneously with cell cycle characteristics in 63 bone marrow biopsies from 23 cases with acute promyelocytic leukemia (APL). Bromodeoxyuridine (BrdU) was administered to every patient (17 newly diagnosed) for determination of the labeling index (LI) and the durations of S-phase (Ts) and the cell cycle (Tc) of leukemic promyelocytes. APL cases had lower LI both in the bone marrow aspirate (6.1% v 11.4%, P = .008) and biopsy (21.1% v 28.0%, P = .001) and longer Tc (93.6 hours v 56.0 hours, P = .002) when compared with other French-American-British subtypes. TGF-beta expression (detected by a monoclonal anti-TGF-beta 2/beta 3 antibody) was dramatically high, especially in interstitial areas of the biopsies. S-phase cells were found as geographically restricted islands of proliferation (GRIPs) in 20 of 22 cases. Weekly biopsies showed an increment in TGF-beta on day 7 of therapy in 13 of 17 cases, while in vivo differentiation was noted in 9 of 15. We conclude that the presence of high TGF-beta expression may explain the biologic basis for the slowly cycling nature of leukemic promyelocytes in APL as well as the unique clustering of S- phase cells observed in GRIPs.

Although the prognostic and clinical significance of cell cycle kinetic studies in neoplastic diseases has been inconclusive at worst and controversial at best, new advances in techniques to study cell cycle characteristics have dramatically improved our ability to more accurately measure the parameters of labeling index (LI), S-phase time (Ts) and total cell cycle time (Tc), and subsequently to find any correlations which would enable us to use these as prognostic indicators. Data from 285 patients with acute myeloid leukemia (AML) who were given in-vivo infusion of the thymidine analogue bromodeoxyuridine (BrdU) prior to chemotherapy show differences in mean labeling indices derived from bone marrow aspirates versus biopsies-being 9.2% and 22.2% respectively. The Tc and Ts were obtained by double labeling the aspirates with tritiated thymidine in-vitro. In 102 uniformly treated patients with standard risk de novo AML, correlations between cell cycle parameters thus measured and clinical data were sought. Although no relation of cell cycle characteristics to remission-induction outcome was observed, patients with below median biopsy LI and above median Tc showed longer remission durations (p = 0.02 and 0.04). We conclude that patients with slowly cycling leukemias have longer and more durable remissions most probably as a result of retarded regrowth of leukemic myeloblasts between courses of consolidation therapy. KEY WORDS:Acute myeloid leukemia cell cycle remission induction remission duration bromodeoxyuridine iododeoxyuridine