Cell kinetics and repopulation during multifraction irradiation of spheroids: Implications for clinical radiotherapy

Abstract

T here seems to be little question that development of a predictive assay that could reliably forecast the response of individual to treatment would be of tremendous value to practicing oncologists. However, exactly what type of information is needed in such an assay is less certain. For example, is the determinant of success the inherent sensitivity of the majority, or that of a minority of the tumor cells? Because treatment is typically recognized by the regrowth or continued growth of the tumor mass, what is the role of cell kinetics in that process? Perhaps of most pressing importance, can predictive assays lead to alternate, hopefully superior, forms of treatment? While measuring tumor cell sensitivity to both radiation and drugs has become fashionable in recent years, correlations between in vitro sensitivity analyses and eventual patient outcome have generally been found only between tumor types, but not between individual of a certain type. This suggests that factors in addition to inherent sensitivity must be considered. Analyzing patterns of failure can often be useful in this regard; in a provocative article analyzing outcomes for patients with head and neck cancer, Withers et al I reached the rather alarming conclusion that treatments lasting more than 4 weeks required a dose escalation of 0.6 Gy/extension day to retain the same probability of cure. More general reviews 2-4 similarly argue for a dose increase of 0.3 to 0.7 Gy/day of treatment for treatment extensions beyond 4 weeks. These results strongly' implicate time as a critical factor for successful treatments. However, identifying a clinical problem is considerably easier than resoMng it. For example, does the repopulation seen at later times during therapy indicate that cells are cycling more rapidly, that more cells are in cycle, that fewer clonogenic cells are lost, or combinations of the above? When does repopulation actuallyaccelerate? Perhaps of greater immediate importance, can reliable and unambiguous methods of measurement be developed for tumor cell kinetics, and when should they be used? Alternatively, can repopulation rate or potential be predicted? Clearly', several philosophical and practical limitations combine to inhibit addressing these questions in the most relevant system, the human cancer patient. At least one tumor model system is available, however, in which many relevant and highly detailed studies can be pursued: the multicell spheroid model. 5,6 It is of particular note that the spheroid system can be easily maintained in long-term culture, allowing large numbers of virtually identical tumors to be propagated and reproducibly tested, without host factors complicating the response. 7,8 Every cell of the system potentially contributes to the net response, and of particular importance, each cell can be recovered and assessed. The intent of this article is to examine the role of cell kinetics when spheroids are treated with multifraction protocols. In addition to multifraction radiation treatments, multiple drug exposures will be briefly explored. From the design of these studies, we have been able to address not only the role of pretreatment cell kinetics, but also factors controlling changes of those kinetic properties during, and in response to, the therapeutic manipulations.