Relationship between tumour cell in vitro radiosensitivity and clinical outcome after curative radiotherapy for squamous cell carcinoma of the head and neck
Introduction
It is well known that the radiation response of human tumours varies widely from one patient to another, which may partly be accounted for by variation in tumour cell radiosensitivity. A theoretical study, analysing the effect of inter-patient variation in tumour radiosensitivity on the accuracy of predictive tests, showed that assays based on in vitro tumour cell radiosensitivity were more likely to correlate with clinical outcome after radiotherapy than other assays [38]. Similarly, several experimental studies have suggested that tumour cell in vitro radiosensitivity, estimated by the fraction of surviving cells after a radiation dose of 2 Gy (SF2), is likely to be a useful parameter for predicting clinical tumour response to radiation therapy [8], [9], [16], [18], [19], [30], [31]. However, in clinical studies culturing primary tumour specimens [12], [42], [17] and human cell lines [3], [29], [37], the usefulness of tumour cell radiosensitivity as a predictor of tumour response to radiotherapy is far from an unequivocal matter.
Detection of real differences in cellular radiosensitivity among tumours depends on the reproducibility of the assay. Generally, a considerable inter-patient variability in SF2, and a relatively small intra-patient variation will increase the predictive accuracy of an in vitro test [27]. An analysis by Bentzen [4], taking an assumed inter-patient variability in tumour cell radiosensitivity into account, allowed the relatively shallow dose-response curves observed clinically for tumours to be broken down into a series of very steep curves. Each dose-response curve would apply to a homogeneous subgroup of patients, stratified according to radiosensitivity. From the different dose-response curves, it could be expected that some of the subpopulations would have a greater therapeutic window than other subpopulations [4]. Thus, if these subgroups could be identified before the start of treatment, using a reliable predictive assay, a substantial therapeutic benefit may be realized. A modified Courtenay–Mills soft agar clonogenic assay was established for estimating cellular radiosensitivity in specimens from squamous cell carcinoma of the head and neck [34], [36]. It was demonstrated that this soft agar clonogenic assay is also a substrate for remarkable growth of fibroblasts arising from the tumour stroma. Therefore, a colony-filter technique and a method for immunocytochemical staining were developed for typing the colony founder cell. The current assay produces in vitro data on both tumour cell radiosensitivity (tumour cell SF2), fibroblast radiosensitivity (fibroblast SF2), and an overall estimate irrespective of cell type (overall SF2) from the same tumour and individual [33], [34], [36]. A considerable inter-patient variability in all SF2 measures has been obtained, and the assay has shown to be reproducible [34]. Overall SF2 and tumour cell SF2 were not related to tumour and patient-related factors, except for increasing tumour cell in vitro radiosensitivity being significantly correlated with decreasing histopathological differentiation [34]. The aim of the present study was to test if tumour cell SF2 and overall SF2 measured in squamous cell carcinomas of the head and neck were able to predict clinical outcome of individual patients treated with curative radiotherapy. The current work is part of an analysis on predictive parameters in radiotherapy of head and neck cancer. This analysis included estimation of tumour oxygenation [25], tumour cell kinetics [22], normal tissue radiosensitivity [23] and tumour cell radiosensitivity.
Section snippets
Patients and tumours
One-hundred and five patients with squamous cell carcinoma of the head and neck entered the study. Among these, successful in vitro experiments were obtained in biopsies from 71 patients. Sixty-three of these patients had primary tumours whereas eight patients had recurrent tumours, as previously described [34].
Thirty-eight of the patients with primary tumours were treated with curative radiotherapy alone. Seven were females. Median age was 60 years (range 38-81). All patients were eligible for
Parameters of cellular in vitro radiosensitivity
Plating efficiency and SF2 values were obtained for all 38 patients, and the individual patient's data are listed in Table 1. Tumour radiosensitivity was characterized by the tumour cell SF2 and by overall SF2, which were both distributed over large ranges as illustrated in Fig. 1. Median SF2 values (with ranges) are given in Table 2. Using weighted linear regression, the tumour cell SF2 and overall SF2 showed to be two independent measures of tumour radiosensitivity in the patients for which
Discussion
The present study involved a rather small sample of patients followed for more than 1 year after being treated by primary radiotherapy with curative intent according to the DAHANCA guideline. Even though the number of patients was limited, a range of clinical parameters was evaluated with respect to treatment outcome and association with in vitro parameters. Advanced stage was correlated with poor loco-regional control, indicating that the patient sample is representative for a population of
Acknowledgements
The authors wish to acknowledge Ms. A. Baden for qualified assistance in the laboratory, as well as Dr. M.R. Horsman and Dr. O.S. Nielsen for critically reading the manuscript. This study was supported from the Danish Cancer Society, the Clinical Research Unit at the Oncology Center, Aarhus University Hospital, and the Faculty of Health Sciences, Aarhus University, Denmark.
References (43)
- et al.
Intratumoral heterogeneity of malignant gliomas measured in vitro
Int. J. Radiat. Oncol. Biol. Phys.
(1993) - et al.
Radiosensitivity testing in gynecological tumours and malignant gliomas
Radiobiological considerations in the design of clinical trials
Radiother. Oncol.
(1994)Towards evidence based radiation oncology: improving the design, analysis, and reporting of clinical outcome studies in radiotherapy
Radiother. Oncol.
(1998)- et al.
The p53 gene as a modifier of intrinsic radiosensitivity: implications for radiotherapy
Radiother Oncol.
(1996) - et al.
Comparison between in vitro radiosensitivity and in vivo radioresponse of murine tumour cell lines I: parameters of in vitro radiosensitivity and endogenous cellular glutathione levels
Int. J. Radiat. Oncol. Biol. Phys.
(1990) - et al.
Comparison between in vitro radiosensitivity and in vivo radioresponse in murine tumour cell lines II: in vivo radioresponse following fractionated treatment and in vitro/in vivo correlations
Int. J. Radiat. Oncol. Biol. Phys.
(1990) - et al.
Intratumoral heterogeneity as a confounding factor in clonogenic assays for tumour radioresponsiveness
Radiother. Oncol.
(1996) - et al.
Cellular radiosensitivity of primary head and neck squamous cell carcinomas and local tumour control
Int. J. Radiat. Oncol. Biol. Phys.
(1990) - et al.
Radiosensitivity testing of primary cervical carcinoma: evaluation of intra- and inter-tumour heterogeneity
Radiother. Oncol.
(1990)