Late complications after radiotherapy for prostate cancer.

BACKGROUND: The aim of the present study was to analyze in detail the time course of the incidence of radiation-induced late effects. For this purpose, unpublished data of patients treated by radiation therapy in Hamburg in the late 1980s were analyzed. Relatively large volumes were exposed to comparatively high doses, thus leading to a high rate of treatment-related side effects. PATIENTS AND METHODS: A total of 180 consecutive patients received radiotherapy for prostate cancer. The median age was 66 years (range 41-88 years). The median of the maximum dose was 77.5 Gy (range 56.3-95 Gy) and overall treatment time was 51 days (range 28-128 days). Endpoints analyzed were late complications of grade 3 or higher, overall and disease-free survival, local tumor control, and distant metastases. Data analysis was actuarial and the log-rank test was used to compare the various subgroups. RESULTS: After 2 years, 80.5 ± 3.2% of the patients were without any complications of grade 3 or higher, and after 5 years a constant level of 70.3 ± 4.0% was approached. When multiple lesions occurred per patient, the later events were disregarded. A total of 66 complications occurred in 42 patients. The percentage of patients being free from late complications, plotted as a function of time after start of radiation therapy, was adequately described by an exponential function and a constant fraction. Complications approached a constant level of 70.3% at a rate of 5.3% per month. This means that patients who will develop a complication do so at exponential kinetics and at a relatively high rate, whereas about 70% of the patients will never experience a late effect even over long observation periods. After subdividing the maximum dose into three equal dose groups of 55 patients each (< 73.3 Gy, 73.3-80 Gy, > 80 Gy), the constant fraction decreased from 85.7 to 72.8% and 52.2%, whereas the incidence rate was 4.3%, 7.7%, and 5.6% per month and, thus, almost independent of radiation dose. CONCLUSION: For a given group of patients, the rate of the incidence of late complications appears to be independent of radiation dose and (from analyzing data in the literature) independent of the grade of lesions, whereas the fraction of patients without late effects depends on both parameters.

Should tumors be clamped in radiobiological fractionation experiments?

In radiobiological fractionation experiments tumors are often clamped during irradiation. This is done not only to facilitate the interpretation of the data by eliminating the influence of reoxygenation but also to avoid uncontrolled changes in the hypoxic fraction that are caused by anesthesia or stress from physical restraint. In this study it is shown that clamping of tumors during irradiation also affects repair, repopulation, and redistribution. From these results it is concluded that clamping of tumors during fractionated irradiation treatment does not appear to be an adequate measure to elucidate the mechanisms that determine tumor response to radiotherapy. Even some of the fundamental concepts of tumor radiobiology might contain some uncertainties, since they are often based on data resulting from clamped tumors.

Combined modality treatment of the rhabdomyosarcoma R1H of the rat: tumor and normal tissue response after cisplatin and conventional or accelerated irradiation treatment.

PURPOSE: To test the importance of the sequence of cisplatin and irradiation, either conventional or accelerated fractionated. METHODS AND MATERIALS: 30 fractions of 2 Gy were given in 6 or 3 weeks preceded or followed by (time interval between cisplatin and radiotherapy: 3 days) a single IP dose of 5 mg/kg cisplatin in the rhabdomyosarcoma R1H of the rat. Survival curves were generated, and comparisons were made by the log-rank test. RESULTS: After 60 Gy in 6 weeks, no local tumor controls were observed. If cisplatin was injected 3 days before start of 60 Gy/6 weeks, 11 +/- 10% (mean +/- SE) of the tumors were controlled. Cisplatin after radiotherapy resulted in 50 +/- 14% local controls. The difference was significant (p = 0.01) for cisplatin after radiotherapy in comparison to radiotherapy alone where no local controls were observed. After accelerated fractionation, 57 +/- 19% of the animals were cured with or without cisplatin before radiotherapy. If the drug was injected after end of 60 Gy/3 weeks, 86 +/- 13% survived recurrence free. The difference to accelerated radiotherapy alone was not significant. Accelerated radiotherapy produced significantly higher control rates than conventional radiotherapy (p < 0.001). CONCLUSIONS: Accelerated radiotherapy resulted in higher local tumor control rates as compared to conventional fractionated irradiation. Cisplatin combined with radiotherapy showed significantly better results if given after but not before irradiation, either conventional or accelerated fractionated.

Radiotherapy of the rhabdomyosarcoma R1H of the rat: kinetics of cellular inactivation by fractionated irradiation.

The kinetics of cellular inactivation by fractionated irradiation in the R1H rhabdomyosarcoma of the rat was studied in the dose range of 1.07 to 12.50 Gy per fraction. Regimens of 1, 3, 5, 7, and 10 fractions per week for several weeks were compared. The number of clonogenic tumor cells per tumor in the course of the different treatment schedules was determined using an in vitro colony assay. The results show that the proliferation of clonogenic tumor cells is decelerated in the course of a fractionated radiotherapy. The deceleration persists for several days after end of treatment, until accelerated repopulation is initiated. The fraction of tumor cells inactivated per week was only dependent on the total dose per week, that is the cellular response was the same whether the weekly dose was applied in 1,3,5,7, or 10 fractions. Thus, the fractionation regimens were considerably more effective than expected from calculations based on single-dose in situ survival curves.

Net growth delay: a novel parameter derived from tumor growth curves.

Growth delay does not only reflect the effect of treatment on the tumor parenchymal cells but also on the stroma. Due to tumor bed effect, the extent of growth delay determined from tumor growth curves is highly dependent on the end volume chosen. It was aimed to minimize the influence of the tumor bed effect on the growth delay calculated by choosing a smaller size and essentially an earlier time for regrowth. Net growth delay is a novel parameter derived from the tumor growth curves, allowing a better comparison of the results with colony assay and tumor control data.

Radiobiology of the rhabdomyosarcoma R1H of the rat: influence of the size of irradiation field on tumor response, tumor bed effect, and neovascularization kinetics.

R1H tumors were irradiated with a single dose of 15 Gy X rays using varying sizes of treatment fields. Damage to tumor cells and tumor stroma was determined separately by analysis of growth delay to ten times treatment volume (GD10vo) and net growth delay. GD10vo comprises irradiation effects on tumor parenchymal cells and on tumor stroma, whereas net growth delay only measures effects on tumor parenchymal cells. Stromal damage was observed to increase with increasing field size; the effect on the tumor parenchymal cells, however, was independent of the field size. An increase of GD10vo of 13 days per cm increase of field size diameter was observed. From this the velocity of neovascularization of the irradiated tumor bed was calculated to be 0.30 to 0.38 mm per day.

Radiotherapy of the rhabdomyosarcoma R1H of the rat: the influence of the number of fractions on tumor and skin response.

The influence of the number of fractions on tumor and skin response to fractionated irradiation was studied. R1H rhabdomyosarcomas of the rat (volume doubling time 3 days) were irradiated with 6, 18, 30, or 42 fractions in 6 weeks. Total doses of 45, 60, or 75 Gy were applied in each fractionation scheme, that is, the dose per fraction ranged from 1.07 to 12.5 Gy. Tumor response was assessed by tumor control probability and tumor net growth delay. A clearcut reduction of skin damage was observed with increasing number of fractions, whereas the tumor response was found to be the same whether the dose was given in 6, 18, 30, or 42 fractions. Thus, the fractionation regimens were more effective than expected from calculations based on single-dose in situ survival curves. This result can be explained by assuming that the clonogenic tumor cells become less hypoxic with increasing number of fractions. Since normal tissue damage decreases with increasing number of fractions, the therapeutic gain may be improved by applying a greater number of fractions.

Kinetics of cellular inactivation by fractionated and hyperfractionated irradiation in Lewis lung carcinoma.

The kinetics of cellular inactivation by fractionated irradiation in Lewis lung carcinoma was studied in the dose range of 2.3 to 6.5 Gy per fraction. Regimens of one and two fractions per day for 10 days were compared. The number of clonogenic tumor cells was determined using an in vitro soft-agar colony assay. Under the experimental conditions used, the fraction of tumor cells inactivated per day was only dependent on the total dose per day, i.e., the cellular response was the same whether the daily dose was given in one or two fractions. Thus, the two fraction per day regimen was more effective than expected from calculations based on acute in vivo radiation dose-survival curves. This result could be explained if the clonogenic tumor cells were less hypoxic during the two fractions per day regimen than the one fraction per day schedule.

Radiotherapy of the R1H-tumor: dose-rate effect on tumor response in brachytherapy with 106-ruthenium eye applicators.

PURPOSE: The influence of the dose-rate on tumor response in radiotherapy with 106-Ruthenium eye plaques has been investigated in an experimental tumor system. MATERIALS AND METHODS: The identical total dose was applied within three different overall treatment times (42 hr, 192 hr and 312 hr), corresponding to dose-rates of 6.0 Gy/hr, 1.3 Gy/hr and 0.8 Gy/hr. RESULTS: The therapeutic outcome of brachytherapy varied significantly between the three groups of animals treated with different dose-rates. At a dose-rate of 1.3 Gy/hr all tumors were locally controlled, but no local control was observed when a dose-rate of 6.0 Gy/hr was delivered. 0.8 Gy/hr was less effective than 1.3 Gy/hr but more effective than 6.0 Gy/hr. CONCLUSION: These results were unexpected but they might be explained by an incomplete reoxygenation if the overall treatment time is too short (42 hr, dose-rate 6.0 Gy/hr) and by proliferation of tumor cells under treatment if the overall treatment time is too long (312 hr, 0.8 Gy/hr).

Inoperable non-small cell lung cancer: a retrospective analysis of 427 patients treated with high-dose radiotherapy.

PURPOSE: The influence of patient and treatment characteristics on survival as well as normal tissue toxicity were retrospectively analyzed. METHODS AND MATERIALS: Four hundred twenty seven patients with unresectable non-small cell lung cancer received at least 60 Gy and two-thirds were treated with 70 Gy. RESULTS: Five-year survival rates and median survival time (95% confidence interval) were 2 +/- 2% (mean +/- s.e.) and 11.1 months (9.1-14.5) after 60-66 Gy (median 60 Gy); 8 +/- 2% and 14.9 months (13.3-16.5) after > or = 70 Gy (p = 0.0013). Stage I-II patients had significantly higher survival rates as compared to Stage III patients (p = 0.0015). Within the subgroup of Stage III patients those with Stage IIIA had significantly higher survival rates than Stage IIIB (p = 0.0167). Female patients achieved 5-year survival rates after 70 Gy of 15 +/- 7% as compared to only 7 +/- 2% of their male counterparts. Chemotherapy, histology, Karnofsky status, and age had no influence on survival after univariate and multivariate analysis. Nine percent and 11% of the patients suffered from moderate to severe pneumonitis and esophagitis. CONCLUSION: High-dose radiotherapy of unresectable non-small cell lung cancer with total doses > 60 Gy conventionally fractionated is feasible. With doses of > or = 70 Gy significantly higher survival rates were achieved as compared to 60-66 Gy. Normal tissue toxicity was acceptable. For Stage IIIB patients, however, treatment results are disappointingly low even after 70 Gy with no 5-year survivor.

Radiotherapy of the rhabdomyosarcoma R1H of the rat: hyperfractionation--126 fractions applied within 6 weeks.

The effect of a hyperfractionated irradiation treatment on the response of the rhabdomyosarcoma R1H of the rat was studied. Tumors were irradiated under ambient conditions with 126 fractions of X rays, applied in 3 fractions per day with a time interval of 8 +/- 1 hr between fractions on 7 days per week during 6 weeks. The total dose ranged from 54 to 90 Gy, that is the dose per fraction ranged from 0.43 to 0.71 Gy. Tumor response was assessed by tumor control probability and tumor net growth delay. The tumor response to the hyperfractionated treatment was found to be slightly more effective compared to the results obtained in a previous study where treatments with 6, 18, 30, and 42 fractions were applied. Since normal tissues are considerably spared with increased numbers of fractions, clinical studies with hyperfractionation seem to be very promising.

Radiotherapy of the rhabdomyosarcoma R1H of the rat: postoperative radiotherapy.

Rhabdomyosarcoma R1H of the rat was excised aiming for a complete macroscopic local excision. Adjuvant radiotherapy was performed from the third postoperative day on. Former tumor sites were locally irradiated with 200 kVp X rays 4 times per week over a period of 6 weeks. Total doses of 0, 15, 30, 45 and 60 Gy were applied. The tumor volume was measured and the time to regrow to initial volume was assessed. The results were compared to the effect of a standard radiotherapy alone with 30 fractions of 2 Gy applied within 6 weeks. All tumors recurred despite of the irradiation treatment. At high total doses (greater than 30 Gy), adjuvant radiotherapy was found to improve short term tumor response considerably, whereas no positive effect was seen at low doses. After a total dose of 60 Gy and long time intervals after start of treatment, radiotherapy alone and a combination of surgery and radiotherapy seem to be isoeffective in our tumor system.

Radiobiological determination of growth kinetics and radiosensitivity of pulmonary micrometastases of the R1H tumour.

The aim of the study was to determine the radiosensitivity and the growth kinetics of pulmonary micrometastases of the R1H tumour of the rat. Lung metastases were induced by intravenous injection of viable tumour cells. At different time intervals (3-32 days) after injection, lungs were locally irradiated with 200 kVp X-rays, using 1.5 Gy/fraction. Total doses of 6-33 Gy were administered within 11 days. Endpoints used were survival time, local control rate, and number of metastases in the lungs at autopsy. The data were evaluated using the multi-target model. Beginning in the fifth week after tumour cell inoculation the animals started to exhibit a pronounced dyspnoea and were sacrificed. Sections revealed an extensive metastatic infiltration of the lungs. With increasing total dose a prolongation of survival time as well as an increase in cure rate was observed. The number of metastases found in the lungs decreased with increasing total dose. It is concluded that metastatic growth does not start earlier than 3 days after tumour cell inoculation and accelerates continuously. The doubling time of the tumour cells in the micrometastases decreases continuously and from 5.2 to 1.2 days between day 3 and 40, whereas larger metastases containing more than 10(6) cells show gompertzian growth kinetics. The cell doubling time in this stage of metastatic growth is longer than 5 days. During the first 4 weeks of metastatic growth the radiosensitivity of metastatic R1H cells in the lungs is the same as in vitro.

Origins of radiotherapy and radiobiology: separation of the influence of dose per fraction and overall treatment time on normal tissue damage by Reisner and Miescher in the 1930s.

The effect of fractionation on the response of normal tissues to irradiation was already investigated in the 1930s. Reisner (Reisner, A. Hauterythem und Röntgenstrahlung. Erg. Med. Strahlenforsch, 6: 1-60, 1933) measured the time course of skin erythema on thighs of humans by applying different doses per fraction while keeping constant total dose and overall treatment time. The results showed that acute skin damage was reduced with small doses per fraction. Two years later Miescher (Miescher, G. Tierexperimentelle Untersuchungen über den Einfluss der Fraktionierung auf den Späteffekt. Acta Radiol, 16: 25-38, 1935) published his results on late radiation effects in rabbit skin. He also reported that the main influencing factor for tissue tolerance was dose per fraction. In addition, he found indications that there was no impact of overall treatment time on the development of late reactions. Strandqvist in his famous monograph on the time factor in treatment of skin cancer (Strandqvist, M. Studien über die kumulative Wirkung der Röntgenstrahlen bei Fraktionierung. Erfahrungen aus dem Radiumhemmet an 280 Haut -und Lippenkarzinomen. Acta Radiol. (Suppl.) 55: 1-300, 1944), however, postulated that total dose and overall treatment time were the main determinants of local control as well as of normal tissue damage, apparently omitting to consider the findings of Reisner and Miescher in his own analysis. It is our impression that mainly due to the large influence of Strandqvist's work on radiobiological thinking the early findings on the normal tissue sparing effect of small fraction size have been forgotten and had to be rediscovered about 40 years later.

Potential pitfalls in the use of p-values and in interpretation of significance levels.

In multiparameter analysis of clinical data the likelihood of obtaining significant results, just by chance, increases considerably with the overall number of tests performed. This can be compensated for by adjusting the p-values. Two tables are given from which adjusted p-values may be read off, providing a simple procedure to test the reliability of the results from clinical studies. Multivariate analyses often impress with extremely low p-values, but frequently these results turn out to be non-significant when the influence of multiple testing is considered. As a consequence the actually relevant results might be overlooked, because of the large number of spurious results accumulating in the literature. The use of inappropriate statistics hampers progress in clinical research. It is concluded that more care in the use of p-values in analysis and interpretation of clinical data is required.

Radiotherapy of the rhabdomyosarcoma R1H of the rat: influence of the time interval between two daily fractions during hyperfractionated radiotherapy.

The response of the rhabdomyosarcoma R1H of the rat to hyperfractionated irradiation with different time intervals between the two daily fractions has been investigated. All tumours were exposed to irradiation 5 days per week over 6 weeks. A standard treatment of 30 fractions, i.e. one fraction per day, of 1.83-2.75 Gy (200 kVp X-rays) was compared with a hyperfractionated schedule of 60 fractions, i.e. two fractions per day, of 0.92-1.38 Gy with time intervals of either 1, 2, 3, 5 or 6 h between the two daily fractions. Tumour response has been assessed by (a) net growth delay and (b) local tumour control. Compared with standard treatment (one fraction per day) significant reduction (p < 0.005) of net growth delay was observed for the tumours treated with two daily fractions separated by 2 h. However, at a time interval of 5 and 6 h between the two daily fractions net growth delay increased considerably (p < 0.0001 and p < 0.01) as compared with the standard treatment and an increased rate of local tumour control was observed. The major findings is that tumour response is not as would be predicted by repair if the interval between the two daily fractions exceeds 2 h. The competing or overriding mechanisms cannot be identified ultimately, but the data, though based on small animal numbers and collected over an extended time period, reflect a significant effect.

Impact of the interfraction interval on clonogenic cell survival in split-dose irradiation of R1H rhabdomyosarcoma of the rat in vitro.

PURPOSE: In a previous study, the response of the R1H rhabdomyosarcoma of the rat to conventional irradiation (1.83-2.75 Gy fractions once-daily) and hyperfractionated radiotherapy (0.92-1.38 Gy fractions with different time intervals between the two daily fractions) was investigated [Kleineidam, M., Pieconka, A., Beck-Bornholdt, H.-P. Radiotherapy of the rhabdomyosarcoma R1H of the rat: influence of the time interval between two daily fractions during hyperfractionated radiotherapy. Radiother. Oncol. 30: 128-132, 1994]. Compared to once-daily irradiation, interfraction intervals of 2 h led to reduced tumour response, due to recovery from sublethal damage, and intervals of 5-6 h resulted in increased tumour response, possibly due to cell cycle effects. The purpose of the present study was to complement these tumour data by measuring clonogenic cell survival of R1H cells in vitro after split-dose irradiation. METHODS AND MATERIALS: Experiments were performed with 2 x 2.5 Gy and 2 x 3.5 Gy either at 21 degrees C (preventing cell cycle progression) or at 37 degrees C (allowing for cell cycle effects). RESULTS: For 3.5 Gy fractions, a cell survival curve equivalent to the in vivo results was obtained with the lowest surviving fraction observed at time intervals of 6-7 h, but only when cells were incubated at 37 degrees C during the interval. This phenomenon was absent in the 21 degrees C experiments. CONCLUSIONS: Our data provide further evidence to support the hypothesis that cell cycle effects are responsible for such observations. We conclude that the length of the interfraction interval has a considerable potential effect on tumour response to altered fractionation regimens.

Combination of fractionated irradiation with nicotinamide and carbogen in R1H-tumours of the rat and its pulmonary metastases.

BACKGROUND AND PURPOSE: The combination of radiotherapy with nicotinamide and carbogen is a promising method to improve treatment outcome. The aim of the present study was to investigate its effect on rat lungs, the R1H-tumour and its pulmonary metastases. MATERIALS AND METHODS: Subcutaneous tumours and artificially induced pulmonary metastases of the rhabdomyosarcoma R1H of the rat were treated either with fractionated irradiation alone or in combination with nicotinamide and carbogen. Local metastatic control, net growth delay and lethal lung damage were used as endpoints. RESULTS: Radiosensitivity of the lungs of WAG/Rij rats increased by a factor of 1.13, while no positive effect of the combined treatment either on subcutaneous R1H-tumours or on pulmonary metastases could be detected. This resulted in a therapeutic loss of 18% (P = 0.003). CONCLUSION: The results indicate that deleterious effects of the combination of radiotherapy with nicotinamide and carbogen cannot be excluded under all circumstances.

Tumor volume: a basic and specific response predictor in radiotherapy.

BACKGROUND AND PURPOSE: Predictive assays of the response of tumor and normal tissues in individual patients offer the possibility of individualized prognosis and treatment decisions. For this purpose a variety of assays are currently being explored. The impact of tumor volume on radiotherapy outcome has long been recognized and in this paper its predictive potential is investigated. METHODS: Re-evaluation of clinical data from the literature. RESULTS: Tumor volume significantly influences radiotherapy outcome and in many sites it is likely a superior prognostic indicator to tumor stage, which reflects tumor size only partially and is mainly correlated to operability. Tumors even of identical stage may vary by factors of more than 100 in volume and neglect of this heterogeneity clearly reduces the power of a study considerably. The precision requirements for the measurement of tumor volume are small; +/-50% is sufficient for reasonable results. CONCLUSION: The data evaluated here suggest that tumor volume is the most precise and most relevant predictor of radiotherapy outcome. Its determination is achievable with sufficient accuracy in most radiotherapy departments. Individual tumor volume should always be reported in clinical studies and considered in data analyses.

Hyperfractionation: where do we stand?

Hyperfractionation is generally expected to allow an escalation of total dose, thereby increasing tumour control rate, without increasing the risk of late complications. The purpose of this review is to assess the empirical evidence for this therapeutic gain from hyperfractionated radiotherapy. Although extensive clinical data have been accumulated until now, especially on treatment of head and neck cancer, the line of evidence is not consistent. The present analysis indicates that the dose per fraction generally used in standard radiotherapy is already a good choice.