Reirradiation with stereotactic body radiotherapy: analysis of human spinal cord tolerance using the generalized linear-quadratic model.

AIM: Using the generalized linear-quadratic (gLQ) model, we reanalyzed published dosimetric data from patients with radiation myelopathy (RM) after reirradiation with spinal stereotactic body radiotherapy (SBRT). MATERIALS & METHODS: Based on a published study, the thecal sac dose of five RM patients and 14 no RM patients were reanalyzed using gLQ model. Maximum point doses (Pmax) in the thecal sac were obtained. The gLQ-based biological effective doses were calculated and normalized (nBEDgLQ) to a 2-Gy equivalent dose (nBEDgLQ = Gy2/2_gLQ). The initial conventional radiotherapy dose, converted to Gy2/2_gLQ, was added. RESULTS: Total (conventional radiotherapy + SBRT) mean Pmax nBEDgLQ was lower in no RM than RM patients: 59.2 Gy2/2_gLQ (range: 37.5-101.9) versus 94.8 Gy2/2_gLQ (range: 70.2-133.4) (p = 0.0016). The proportion of total Pmax nBEDgLQ accounted for by the SBRT Pmax nBEDgLQ was higher for RM patients. No RMs were seen below a total spinal cord nBEDgLQ of 70 Gy2/2_gLQ. CONCLUSION: The gLQ-derived spinal cord tolerance for total nBEDgLQ was 70 Gy2/2_gLQ.

A generalized linear-quadratic model incorporating reciprocal time pattern of radiation damage repair.

PURPOSE: It has been conventionally assumed that the repair rate for sublethal damage (SLD) remains constant during the entire radiation course. However, increasing evidence from animal studies suggest that this may not the case. Rather, it appears that the repair rate for radiation-induced SLD slows down with increasing time. Such a slowdown in repair would suggest that the exponential repair pattern would not necessarily accurately predict repair process. As a result, the purpose of this study was to investigate a new generalized linear-quadratic (LQ) model incorporating a repair pattern with reciprocal time. The new formulas were tested with published experimental data. METHODS: The LQ model has been widely used in radiation therapy, and the parameter G in the surviving fraction represents the repair process of sublethal damage with T(r) as the repair half-time. When a reciprocal pattern of repair process was adopted, a closed form of G was derived analytically for arbitrary radiation schemes. The published animal data adopted to test the reciprocal formulas. RESULTS: A generalized LQ model to describe the repair process in a reciprocal pattern was obtained. Subsequently, formulas for special cases were derived from this general form. The reciprocal model showed a better fit to the animal data than the exponential model, particularly for the ED50 data (reduced χ(2) (min) of 2.0 vs 4.3, p = 0.11 vs 0.006), with the following gLQ parameters: α/ß = 2.6-4.8 Gy, T(r) = 3.2-3.9 h for rat feet skin, and α/ß = 0.9 Gy, T(r) = 1.1 h for rat spinal cord. CONCLUSIONS: These results of repair process following a reciprocal time suggest that the generalized LQ model incorporating the reciprocal time of sublethal damage repair shows a better fit than the exponential repair model. These formulas can be used to analyze the experimental and clinical data, where a slowing-down repair process appears during the course of radiation therapy.

When tumor repopulation starts? The onset time of prostate cancer during radiation therapy.

PURPOSE: To analyze published clinical data and provide a preliminary estimate of tumor repopulation rate and its onset time during radiation therapy for prostate cancer. METHODS: Data on prostate cancer treated with external beam radiotherapy (EBRT) by Perez et al. (2004), Amdur et al. (1990) and Lai et al. (1991) were analyzed in this study. The stage-combined pelvic control rate from Perez et al. was calculated to be 0.95±0.01, 0.87±0.02, and 0.72±0.04 for patients treated ≤7 weeks, 7.1-9 weeks, and >9 weeks respectively. Based on the Linear-Quadratic model, extended to account for tumor repopulation, the least χ² method was used to fit the clinical data and derive the onset time (T(k)) and effective doubling time (T(d)) for prostate cancer. Similar analysis was performed for the other two datasets. RESULTS: Best fit was achieved with onset time T(k)=34±7 days and doubling time T(d)=12±2 days. These parameters were independent of the choice of the α/ß values currently published in the literature. Analyses of the other two datasets showed T(k)=42±7 days with T(d)=9 ± 3 days, and T(k)=34±6 days with T(d)=34±5 days, respectively. T(k) was found to be dependent on tumor stage. CONCLUSIONS: Consistent values for onset time T(k) were obtained from different datasets, while the range of doubling time T(d) was large. Tumor repopulation starts no later than 58 days (at 90% confidence level) in the course of EBRT for prostate cancer.

Temporal analysis of tumor heterogeneity and volume for cervical cancer treatment outcome prediction: preliminary evaluation.

In this paper, we present a method of quantifying the heterogeneity of cervical cancer tumors for use in radiation treatment outcome prediction. Features based on the distribution of masked wavelet decomposition coefficients in the tumor region of interest (ROI) of temporal dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) studies were used along with the imaged tumor volume to assess the response of the tumors to treatment. The wavelet decomposition combined with ROI masking was used to extract local intensity variations in the tumor. The developed method was tested on a data set consisting of 23 patients with advanced cervical cancer who underwent radiation therapy; 18 of these patients had local control of the tumor, and five had local recurrence. Each patient participated in two DCE-MRI studies: one prior to treatment and another early into treatment (2-4 weeks). An outcome of local control or local recurrence of the tumor was assigned to each patient based on a posttherapy follow-up at least 2 years after the end of treatment. Three different supervised classifiers were trained on combinational subsets of the full wavelet and volume feature set. The best-performing linear discriminant analysis (LDA) and support vector machine (SVM) classifiers each had mean prediction accuracies of 95.7%, with the LDA classifier being more sensitive (100% vs. 80%) and the SVM classifier being more specific (100% vs. 94.4%) in those cases. The K-nearest neighbor classifier performed the best out of all three classifiers, having multiple feature sets that were used to achieve 100% prediction accuracy. The use of distribution measures of the masked wavelet coefficients as features resulted in much better predictive performance than those of previous approaches based on tumor intensity values and their distributions or tumor volume alone.

Comparison of real-time intraoperative ultrasound-based dosimetry with postoperative computed tomography-based dosimetry for prostate brachytherapy.

PURPOSE: To evaluate whether real-time intraoperative ultrasound (US)-based dosimetry can replace conventional postoperative computed tomography (CT)-based dosimetry in prostate brachytherapy. METHODS AND MATERIALS: Between December 2001 and November 2002, 82 patients underwent (103)Pd prostate brachytherapy. An interplant treatment planning system was used for real-time intraoperative transrectal US-guided treatment planning. The dose distribution was updated according to the estimated seed position to obtain the dose-volume histograms. Postoperative CT-based dosimetry was performed a few hours later using the Theraplan-Plus treatment planning system. The dosimetric parameters obtained from the two imaging modalities were compared. RESULTS: The results of this study revealed correlations between the US- and CT-based dosimetry. However, large variations were found in the implant-quality parameters of the two modalities, including the doses covering 100%, 90%, and 80% of the prostate volume and prostate volumes covered by 100%, 150%, and 200% of the prescription dose. The mean relative difference was 38% and 16% for doses covering 100% and 90% of the prostate volume and 10% and 21% for prostate volumes covered by 100% and 150% of the prescription dose, respectively. The CT-based volume covered by 200% of the prescription dose was about 30% greater than the US-based one. Compared with CT-based dosimetry, US-based dosimetry significantly underestimated the dose to normal organs, especially for the rectum. The average US-based maximal dose and volume covered by 100% of the prescription dose for the rectum was 72 Gy and 0.01 cm(3), respectively, much lower than the 159 Gy and 0.65 cm(3) obtained using CT-based dosimetry. CONCLUSION: Although dosimetry using intraoperative US-based planning provides preliminary real-time information, it does not accurately reflect the postoperative CT-based dosimetry. Until studies have determined whether US-based dosimetry or postoperative CT-based dosimetry can better predict patient outcomes, the American Brachytherapy Society recommendation of CT-based postimplant dosimetry should remain the standard of care.

Hypofractionation regimens for stereotactic radiotherapy for large brain tumors.

PURPOSE: To investigate equivalent regimens for hypofractionated stereotactic radiotherapy (HSRT) for brain tumor treatment and to provide dose-escalation guidance to maximize the tumor control within the normal brain tolerance. METHODS AND MATERIALS: The linear-quadratic model, including the effect of nonuniform dose distributions, was used to evaluate the HSRT regimens. The alpha/beta ratio was estimated using the Gammaknife stereotactic radiosurgery (GKSRS) and whole-brain radiotherapy experience for large brain tumors. The HSRT regimens were derived using two methods: (1) an equivalent tumor control approach, which matches the whole-brain radiotherapy experience for many fractions and merges it with the GKSRS data for few fractions; and (2) a normal-tissue tolerance approach, which takes advantages of the dose conformity and fractionation of HSRT to approach the maximal dose tolerance of the normal brain. RESULTS: A plausible alpha/beta ratio of 12 Gy for brain tumor and a volume parameter n of 0.23 for normal brain were derived from the GKSRS and whole-brain radiotherapy data. The HSRT prescription regimens for the isoeffect of tumor irradiation were calculated. The normal-brain equivalent uniform dose decreased as the number of fractions increased, because of the advantage of fractionation. The regimens for potential dose escalation of HSRT within the limits of normal-brain tolerance were derived. CONCLUSIONS: The designed hypofractionated regimens could be used as a preliminary guide for HSRT dose prescription for large brain tumors to mimic the GKSRS experience and for dose escalation trials. Clinical studies are necessary to further tune the model parameters and validate these regimens.

Fractionated grid therapy in treating cervical cancers: conventional fractionation or hypofractionation?

PURPOSE: To evaluate the conventionally fractionated and hypofractionated grid therapy in debulking cervical cancers using the linear quadratic (LQ) model. METHODS AND MATERIALS: A Monte Carlo technique was used to calculate the dose distribution of a commercially available grid in a 6-MV photon beam. The LQ model was used to evaluate the therapeutic outcome of both the conventionally fractionated (2 Gy/fraction) and hypofractionated (15 Gy/fraction) grid therapy regimens to debulk cervical cancers with different LQ parameters. The equivalent open-field dose (EOD) to the cancer cells and therapeutic ratio (TR) were defined by comparing grid therapy with the open debulking field. The clinical outcomes from 114 patients were used to verify our theoretical model. RESULTS: The cervical cancer and normal tissue cell survival statistics for grid therapy in two regimens were calculated. The EODs and TRs were derived. The EOD was only a fraction of the prescribed dose. The TR was dependent on the prescribed dose and the LQ parameters of both the tumor and normal tissue cells. The grid therapy favors the acutely responding tumors inside radiosensitive normal tissues. Theoretical model predictions were consistent with the clinical outcomes. CONCLUSIONS: Grid therapy provided a pronounced therapeutic advantage in both the hypofractionated and conventionally fractionated regimens compared with that seen with single fraction, open debulking field regimens, but the true therapeutic advantage exists only in the hypofractionated grid therapy. The clinical outcomes and our study indicated that a course of open-field radiotherapy is necessary to control tumor growth fully after a grid therapy.

In vitro determination of radiation sensitivity parameters for DU-145 prostate cancer cells.

PURPOSE: The prolonged delivery times associated with intensity modulated radiation therapy (IMRT) may reduce treatment effectiveness of radiation therapy for cancers with short repair half-times. In this study, in vitro radiation experiments with DU-145 prostate cancer cells were designed to quantify the half-time of sublethal damage repair. METHOD AND MATERIALS: A series of single-fraction and split-dose clonogenic survival experiments were performed and analyzed using the linear-quadratic (LQ) survival model with mono-/two-component exponential and reciprocal-time repair kinetic models. RESULTS: Our data indicate that DU-145 cells are very radiosensitive (alpha = 0.44 Gy(-1), standard CI: 0.41-0.49 Gy(-1)) and are relatively insensitive to dose fractionation (alpha/beta = 16 Gy, standard CI: 12-34 Gy). The estimated repair half-time is 23 min (standard CI: 10-97 min) with some evidence that a small portion of the sublethal damage is repaired more slowly. CONCLUSION: The reported radiosensitivity parameters (alpha and alpha/beta) are larger than those derived from other in vitro experiments and clinical data. In contrast, the half-time for sublethal damage repair ( approximately 23 min) is close to the one derived from clinical data ( approximately 16 min). For such short repair half-times, the effectiveness of IMRT treatments may be substantially improved by decreasing the fraction delivery time.

Effect of edema, relative biological effectiveness, and dose heterogeneity on prostate brachytherapy.

Many factors influence response in low-dose-rate (LDR) brachytherapy of prostate cancer. Among them, edema, relative biological effectiveness (RBE), and dose heterogeneity have not been fully modeled previously. In this work, the generalized linear-quadratic (LQ) model, extended to account for the effects of edema, RBE, and dose heterogeneity, was used to assess these factors and their combination effect. Published clinical data have shown that prostate edema after seed implant has a magnitude (ratio of post- to preimplant volume) of 1.3-2.0 and resolves exponentially with a half-life of 4-25 days over the duration of the implant dose delivery. Based on these parameters and a representative dose-volume histogram (DVH), we investigated the influence of edema on the implant dose distribution. The LQ parameters (alpha=0.15 Gy(-1) and alpha/beta=3.1 Gy) determined in earlier studies were used to calculate the equivalent uniform dose in 2 Gy fractions (EUD2) with respect to three effects: edema, RBE, and dose heterogeneity for 125I and 103Pd implants. The EUD2 analysis shows a negative effect of edema and dose heterogeneity on tumor cell killing because the prostate edema degrades the dose coverage to tumor target. For the representative DVH, the V100 (volume covered by 100% of prescription dose) decreases from 93% to 91% and 86%, and the D90 (dose covering 90% of target volume) decrease from 107% to 102% and 94% of prescription dose for 125I and 103Pd implants, respectively. Conversely, the RBE effect of LDR brachytherapy [versus external-beam radiotherapy (EBRT) and high-dose-rate (HDR) brachytherapy] enhances dose effect on tumor cell kill. In order to balance the negative effects of edema and dose heterogeneity, the RBE of prostate brachytherapy was determined to be approximately 1.2-1.4 for 125I and 1.3-1.6 for 103Pd implants. These RBE values are consistent with the RBE data published in the literature. These results may explain why in earlier modeling studies, when the effects of edema, dose heterogeneity, and RBE were all ignored simultaneously, prostate LDR brachytherapy was reported to show an overall similar dose effect as EBRT and HDR brachytherapy, which are independent of edema and RBE effects and have a better dose coverage.

Serial therapy-induced changes in tumor shape in cervical cancer and their impact on assessing tumor volume and treatment response.

OBJECTIVE: The purpose of this study was to evaluate the patterns and distribution of tumor shape and its temporal change during radiation therapy (RT) in cervical cancer and the effect of tumor configuration changes on the correlation between region of interest (ROI)-based and diameter-based MRI tumor measurement. MATERIALS AND METHODS: Serial MRI examinations (T1-weighted and T2-weighted images) were performed in 60 patients (age range, 29-75 years; mean, 53.3 years) with advanced cervical cancer (stages IB2-IVB/recurrent) who were treated with RT at four time points: start of RT, during RT (at 2-2.5 and at 4-5 weeks of RT), and post-RT. Tumor configuration was classified qualitatively into oval, lobulated, and complex based on MR film review. Two methods of tumor volume measurement were compared: ellipsoid computation of three orthogonal diameters (diameter based) and ROI volumetry by delineating the entire tumor volume on the MR workstation (ROI based). Temporal changes of tumor shape and the respective tumor volumes measured by the two methods were analyzed using linear regression analysis. RESULTS: Most tumors (70%) had a non-oval (lobulated and complex) shape before RT and became increasingly irregular during and after RT: 84% at 2-2.5 weeks of RT (p = 0.037), 86% (p = 0.025) at 4-5 weeks, and 96% post-RT (p = 0.010), compared with 70% pre-RT. Diameter-based and ROI-based measurement correlated well before RT (r = 0.89) but not during RT (r = 0.68 at 2-2.5 weeks, r = 0.67 at 4-5 weeks of RT). CONCLUSION: Most cervical cancers are not oval in shape pretherapy, and they become increasingly irregular during and after therapy because of nonconcentric tumor shrinkage. ROI-based volumetry, which can optimally measure irregular volumes, may provide better response assessment during treatment than diameter-based measurement.

Cranial nerve involvement in nasopharyngeal carcinoma: response to radiotherapy and its clinical impact.

OBJECTIVES: To evaluate the cranial nerve (CN) palsy associated with nasopharyngeal carcinoma (NPC), we studied factors that influenced the neurologic outcome of radiotherapy (RT), and the patterns and time course of neurologic recovery of CN palsy. METHODS: Between July 1987 and July 1989, 93 patients who presented with CN palsy at the time of diagnosis of NPC were studied. All patients underwent external-beam RT with either cobalt-60 or 6-MV photon beams to a dose of 69 to 84 Gy at 2 Gy per fraction. The time course and pattern of neurologic recovery (complete, partial, or none) from CN palsy were evaluated. Age, sex, stage, histology, incidence and distribution of types of CNs involved, duration of CN palsy, and time course of tumor response during RT were correlated with the patterns and the time course of neurologic CN recovery by univariate and multivariate analyses. RESULTS: The cases of CN palsy most commonly involved CN V (38%), CN VI (26%), and CN XII (11%), which accounted for the majority of the cases (75%). The time course of CN recovery was variable and protracted. Most patients showed significant improvement upon completion of RT (51%, 19%, and 30% complete, partial, and no recovery, respectively) and further improvement 6 months after RT (58%, 17%, and 25%, respectively). Cranial nerves V, VI, and XII accounted for 75% of cases with no recovery. Recovery was best for CNs II, IX, and XI and the sympathetic nerve (100%, 87%, 100%, and 100%, respectively) and worst for CNs IV, VII, and XII (67%, 60%, and 40%, respectively, with no recovery). Neurologic CN recovery correlated significantly with the pretherapy duration (<3 months versus > or =3 months) of CN palsy (88% versus 62%; p = .002, multivariate analysis), the time course of clinical tumor regression, and neurologic symptom improvement during RT. Age, sex, T stage, N stage, histology, anterior versus posterior CN palsies, and base of skull involvement were not significant. CONCLUSIONS: According to our limited data, most patients with CN palsy respond well to RT. That the time course of neurologic recovery is variable and can be protracted indicates a need for continuous and close neurologic surveillance. The poorer neurologic outcome associated with a longer duration of CN symptoms may be related to a more severe longterm CN compression that results in irreversible damage. Timely diagnosis of NPC and fast institution of therapy are therefore critical to improving the neurologic outcome.

Impact of tumor repopulation on radiotherapy planning.

PURPOSE: Biologic/functional imaging (e.g., fluorodeoxyglucose/3'-deoxy-3'-fluorothymidine-positron emission tomography) is promising to provide information on tumor cell repopulation. Such information is important in the design of biologically conformal radiotherapy for cancer. The questions remaining unclear are whether it is necessary to escalate the dose to the regions with rapid cell repopulation in the tumor target and, if so, by how much. The purpose of this work was to address these questions using radiobiologic modeling. METHODS AND MATERIALS: The generalized linear-quadratic model, extended to account for the effect of clonogenic cell repopulation, was used to calculate the cell-killing efficiency of radiotherapy. The standard Poisson tumor control probability (TCP) model was used to bridge cell killing to treatment outcome. Prostate cancer was chosen as the example for this study. In situ measurements of prostate cancer patients have shown that the potential doubling time of tumor cells has a large variation, ranging from 15 to 170 days. On the basis of the linear-quadratic and TCP parameters (alpha = 0.14 Gy(-1), alpha/beta = 3.1 Gy, and the number of clonogens K = 10(6)-10(7) cells) determined in earlier studies, we evaluated the influence of tumor cell repopulation during protracted treatment courses on treatment outcome. The dose escalations, which can be used to combat aggressive cell repopulation in regions with different doubling times (15-170 days) and sizes (5, 10, 15, and 40 cm(3) of a 40-cm(3) tumor), were calculated for commonly practiced radiotherapy modalities. The influence of linear-quadratic parameters on this calculation was also considered. RESULTS: The impact of tumor cell repopulation on TCP and the corresponding dose escalation required to account for this impact were investigated for both external beam radiotherapy and permanent implantation. The results indicated that for regions with aggressive tumor cell growth, dose escalation is necessary to compensate for the repopulation effect. For example, for tumors with an effective doubling time changing from 42 days to 15 days, the prescription dose of external beam radiotherapy needs to be increased from 75.6 to 81 Gy to maintain a target TCP of 80% for intermediate-risk prostate cancer. For (125)I implants, dose escalation from 152 to 160 Gy is required for the same target TCP. These data were calculated on the basis of an alpha/beta ratio of 3.1 Gy. Greater dose escalations are required if the alpha/beta ratio is 1.5 Gy (e.g., 88 Gy for external beam radiotherapy or 180 Gy for (125)I implantation for the same treatment outcome). Our study results showed that it is important to cover the entire tumor volume, including all aggressive spots, with the desired prescription dose, especially for low-dose-rate brachytherapy. CONCLUSION: Dose escalation is necessary to offset the accelerated tumor cell repopulation during prolonged treatment courses. This study provides a preliminary estimate of the dose escalation for prostate cancer based on the in situ measurements of potential doubling time and radiobiologic models. The proposed dose prescriptions are technically feasible for clinical trials.

Dosimetric advantages of IMRT simultaneous integrated boost for high-risk prostate cancer.

PURPOSE: A sequential two-phase process, initial and boost irradiation, is the common practice for the radiotherapy management of high-risk prostate cancer. In this work, we explore the feasibility of using intensity modulated radiation therapy (IMRT) simultaneous integrated boost (SIB), a single-phase process, to simultaneously deliver high dose to the prostate and lower dose to the pelvic nodes. In addition, we introduce the concept of voxel-equivalent dose for the comparison of treatment plans. METHODS AND MATERIALS: The SIB is designed to deliver the same dose (e.g., 45 Gy, 25 x 1.8 Gy) as the conventional method to the pelvic nodes and to deliver higher doses to prostate in the same 25 fractions (i.e., hypofractionation). The equivalent uniform dose (EUD) was used to determine suitable SIB fractionations that deliver the biologically equivalent doses to prostate. For tumor, the EUD was estimated based on the linear quadratic (LQ) model. The most recent LQ parameters derived from clinical data for prostate cancer were used. The sensitivity of LQ parameters was evaluated. The EUD for normal tissue was computed based on the widely used Lyman model. To be able to consider biologic effectiveness spatially (e.g., voxel by voxel), we propose a new concept, termed the voxel-equivalent dose (VED). The calculation of VED was similar to that for EUD, except that it was done within a voxel. To demonstrate dosimetric feasibility and advantages of the proposed IMRT SIB, we have performed a retrospective planning study on selected patient cases using commercial IMRT and three-dimensional (3D) planning systems. Four treatment scenarios were considered: (1) the conventional 3D plan for initial whole-pelvic irradiation and subsequent conventional 3D boost plan for prostate gland, (2) the conventional 3D plan for initial whole-pelvic irradiation and subsequent IMRT boost plan for prostate, (3) IMRT plan for initial whole-pelvic irradiation and subsequent IMRT boost plan for prostate, and (4) IMRT SIB. EUDs and VED-based dose-volume histograms for prostate, pelvic nodes, small bowel, rectum, bladder, and other tissue for all 4 scenarios were compared. RESULTS: A series of equivalent hypofractionation regimens suitable for the IMRT SIB were obtained for high-risk prostate cancer. For example, the conventional treatment regimen of 42 x 1.8 Gy (EUD = 75.4 Gy) would be equivalent to a SIB regimen of 25 x 2.54 Gy. From the comparison of 3D VED dose distributions and dose-volume histograms between the SIB and the conventional two-phase irradiation, we found that the SIB offers better or equivalent dose conformity to prostate and pelvic nodes and better sparing to the critical structures. For example, for the 4 treatment scenarios with a prostate EUD of 75.4 Gy, the corresponding rectal EUDs are 67.1 (3D + 3D), 65.6 Gy (3D + IMRT), 63.7 Gy (IMRT + IMRT), and 62.0 Gy (SIB). CONCLUSIONS: A new IMRT simultaneous integrated boost strategy that irradiates prostate via hypofractionation while irradiating pelvic nodes with the conventional fractionation is proposed for high-risk prostate cancer. Compared to the conventional two-phase treatment, the proposed SIB technique offers potential advantages, including better sparing of critical structures, more efficient delivery, shorter treatment duration, and better biologic effectiveness.

Brachytherapy management of the retroverted uterus using ultrasound-guided implant applicator placement.

PURPOSE: Patients with a retroverted uterus present a dilemma for brachytherapy in gynecologic malignancies because of the challenges of the procedure and the risk of uterine perforation. The purpose of this study was to evaluate the efficacy and outcome of ultrasound-guided brachytherapy applicator placement and intraoperative uterine anteversion in patients with gynecologic malignancies, who have a retroverted uterus. METHODS AND MATERIALS: Thirty-three brachytherapy insertions were performed in 18 patients with retroverted uterus (cervical cancer, 17; vaginal cancer, 1). The endocervical canal was dilated, the intrauterine Fletcher tandem was inserted in retroverted fashion and then anteverted along with the uterus under continuous ultrasound guidance. The anteverted tandem position was secured with vaginal packing and use of a second and/or third flange on the tandem stem. Treatment was delivered with low-dose-rate brachytherapy using afterloading with 137Cs. Brachytherapy was combined with external beam radiation in all patients. Median post-therapy follow-up was 2.17 years (range, 0.75-9.25 years). RESULTS: Procedure. Ultrasound-guided dilation of the cervix was achieved in all procedures. Sounding of the retroverted uterus up to the fundus was accomplished successfully in all but one procedure (because severe retroflexion of the uterus and fixation of the fundus to the sacrum). Ultrasound-guided anteversion of the inserted tandem and uterus was achieved in all procedures. No ultrasonographic evidence of perforation was seen in any of the procedures. Intraoperative radiographs showed satisfactory position of the applicators in 31 of the 33 procedures; 2 cases were re-packed resulting in acceptable final applicator position. No backward rotation of the tandem was observed over the duration of the low-dose-rate brachytherapy application. The mean ratio between the dose to the rectum and Point A was 73%; the ratio between the dose to the bladder and Point A was 76%. Outcome. In the 17 patients with cervical cancer, 2-year pelvic tumor control rate was 100%, and 2-year actuarial disease-free survival was 73%. The patient with vaginal cancer has no evidence of disease 5 months post-therapy. There was one complication (1/18 patients, 5.5%): a rectal stricture in a patient with stage IVA cervical cancer requiring colostomy. CONCLUSIONS: The use of ultrasound-guided uterine anteversion for brachytherapy applicator placement is feasible and results in acceptable outcome and complication rates in a population otherwise difficult to manage and at high risk for uterine perforation. Based on these results, this method is likely preferable to brachytherapy with a retroverted tandem, or to the omission of brachytherapy.

How low is the alpha/beta ratio for prostate cancer?

PURPOSE: It has been suggested recently that the alpha/beta ratio for human prostate cancer is low (around 1.5 Gy), and much debate on the evidence for such a low value is ongoing. Analyses reported so far ignored the contribution of tumor repopulation. Extremely low alpha values and unrealistic cell numbers of tumor clonogens are found in these studies. In this paper, we present a comprehensive analysis of the updated clinical data to derive a self-consistent set of parameters for the linear-quadratic (LQ) model. METHODS AND MATERIALS: The generalized LQ model, considering the effects of dose rate, sublethal damage repair, and clonogenic proliferation, was used to analyze the recently reported clinical data for prostate cancer using either external-beam radiotherapy or brachytherapy. Three LQ parameters, alpha, alpha/beta, and the repair time, were determined based on the clinical finding that the external-beam radiotherapy and the 125I and (103)Pd permanent implants are biologically equivalent. The tumor control probability model was used also to analyze the clinical data to obtain an independent relationship of alpha/beta vs. alpha and to estimate clonogenic cell numbers for patients in different risk groups. RESULTS: Based on the analysis of clinical data and a consideration of repopulation effect, we have derived a self-consistent set of LQ parameters for prostate cancer: alpha = 0.15 +/- 0.04 Gy(-1), alpha/beta = 3.1 +/- 0.5 Gy. Our analysis indicates the half-time of sublethal damage repair to be in the range from 0 to 90 min with a best estimate of 16 min. The best estimate of clonogenic cell numbers in prostate tumors is found to range from 10(6) to 10(7) according to the patient risk level. These values are more realistic than those derived previously (only 10-100). CONCLUSION: The effect of tumor repopulation is not negligible in determining the LQ parameters for prostate cancer, especially for the low-dose-rate permanent implants. Analysis of clinical data for prostate cancer with corrections for damage repair and repopulation effects results in a low alpha/beta ratio of 3.1 Gy. Unrealistic clonogenic cell numbers and extremely small values of alpha reported in the literature can be resolved by correcting for repopulation effect. The LQ parameters derived presently from the clinical data are consistent with reports of intrinsic radiosensitivity in vitro.

Impact of prolonged fraction delivery times on tumor control: a note of caution for intensity-modulated radiation therapy (IMRT).

PURPOSE: Intensity-modulated radiation therapy (IMRT) allows greater dose conformity to the tumor target. However, IMRT, especially static delivery, usually requires more time to deliver a dose fraction than conventional external beam radiotherapy (EBRT). The purpose of this work is to explore the potential impact of such prolonged fraction delivery times on treatment outcome. METHODS AND MATERIALS: The generalized linear-quadratic (LQ) model, which accounts for sublethal damage repair and clonogen proliferation, was used to calculate the cell-killing efficiency of various simulated and clinical IMRT plans. LQ parameters derived from compiled clinical data for prostate cancer (alpha = 0.15 Gy(-1), alpha/beta = 3.1 Gy, and a 16-min repair half-time) were used to compute changes in the equivalent uniform dose (EUD) and tumor control probability (TCP) due to prolonged delivery time of IMRT as compared with conventional EBRT. EUD and TCP calculations were also evaluated for a wide range of radiosensitivity parameters. The effects of fraction delivery times ranging from 0 to 45 min on cell killing were studied. RESULTS: Our calculations indicate that fraction delivery times in the range of 15-45 min may significantly decrease cell killing. For a prescription dose of 81 Gy in 1.8 Gy fractions, the EUD for prostate cancer decreases from 78 Gy for a conventional EBRT to 69 Gy for an IMRT with a fraction delivery time of 30 min. The values of EUD are sensitive to the alpha/beta ratio, the repair half-time, and the fraction delivery time. The instantaneous dose-rate, beam-on time, number of leaf shapes (segments), and leaf-sequencing patterns given the same overall fraction delivery time were found to have negligible effect on cell killing. CONCLUSIONS: The total time to deliver a single fraction may have a significant impact on IMRT treatment outcome for tumors with a low alpha/beta ratio and a short repair half-time, such as prostate cancer. These effects, if confirmed by clinical studies, should be considered in designing IMRT treatments.

The low alpha/beta ratio for prostate cancer: what does the clinical outcome of HDR brachytherapy tell us?

PURPOSE: Accumulating evidence demonstrates that prostate cancer has a low alpha/beta ratio. However, several challenging issues have been raised from previous studies, including the biologic equivalence between external beam radiotherapy (EBRT) and brachytherapy, the effect of relative biologic effectiveness (RBE) for permanent implantation, and the systematic uncertainties of multi-institutional and multi-modality clinical data. The purpose of this study is to address these issues by reexamining a reported clinical outcome of high-dose-rate (HDR) brachytherapy and to confirm the low alpha/beta ratio for prostate cancer. METHODS AND MATERIALS: The generalized linear-quadratic (LQ) model with considerations of sublethal damage repair and clonogen repopulation was used to calculate the cell-killing efficiency of radiotherapy treatments for prostate cancer. Standard models of tumor cure based on Poisson statistics were used to bridge cell killing to treatment outcome. The data collected in a clinical trial using EBRT plus HDR brachytherapy boost for prostate cancer at William Beaumont Hospital (WBH) were reanalyzed. A 4-year post-treatment time endpoint was chosen as compared to the 3-year endpoint used in the previous study because of better maturity and stability of the data. The least chi-square method was employed to fit the clinical data to estimate the LQ parameters as well as their confidence intervals. The number of clonogens for prostate tumors derived in a separate study was used as a constraint for the data modeling to improve the confidence level. RESULTS: Our analysis demonstrates that only relationships among the LQ parameters, not their definitive and unique values, can be derived from the WBH data set alone. This is due to the large statistical uncertainties, i.e., the small numbers of sampled patients. By combining with the results obtained with the clinical data from Memorial Sloan-Kettering Cancer Center (MSKCC), a new set of LQ parameters (alpha = 0.14 +/- 0.05 Gy(-1), alpha/beta = 3.1(-1.6)(+2.6) Gy) was obtained from the current analysis of the WBH data without dealing with data from permanent implants. The results are consistent with a previous study based on the biologic equivalence between EBRT and permanent implants with a consideration of tumor repopulation. This set of LQ parameters provides a consistent interpretation of clinical data currently available for prostate cancer. CONCLUSIONS: This study provides further evidence to support that prostate cancer has a low alpha/beta ratio of about 3.1 Gy. This study shows that the RBE effect in permanent implantation may not be clinically significant for prostate cancer. The consistency found between this analysis and the previous reported study supports the general biologic equivalence between EBRT and brachytherapy treatments for prostate cancer. The low alpha/beta ratio opens the door to search for more effective radiotherapeutic approaches for prostate cancer, e.g., hypofractionation radiotherapy.

Evaluation of external beam radiotherapy and brachytherapy for localized prostate cancer using equivalent uniform dose.

Various radiotherapy (RT) modalities, such as external beam radiotherapy (EBRT) and permanent/high-dose-rate (HDR) brachytherapy, have been used for the management of localized prostate cancer. Using the linear-quadratic (LQ) model, we compared the relative merits of these modalities in terms of equivalent uniform dose (EUD) and tumor control probability (TCP). The LQ parameters (alpha = 0.15 Gy(-1) and alpha/beta = 3.1 Gy) determined recently from compiled clinical data, as well as other sets of LQ parameters for prostate cancer, were used to carry out the EUD and TCP calculations. A computer code was developed for this purpose. We calculate the EUD for some common RT modalities, and present the corresponding TCP data predicted for a sample patient group (high-risk). Biological equivalence of treatment outcome among various RT modalities is demonstrated. The model suggests that the hypofractionation is preferred in terms of tumor control, due to the lower alpha/beta ratio. Also, the current combined treatment schemes (initial EBRT + permanent/HDR brachytherapy boost) provide higher EUD and TCP than these monotherapies. The study shows that EUD is less sensitive to model parameters than TCP, and EUD can be used to compare and to optimize treatment plans involving different RT modalities. Techniques to further optimize and/or to combine external beams with brachytherapy for better treatment outcomes are proposed.

Equivalence of the linear-quadratic and two-lesion kinetic models.

Double strand breaks (DSBs) are widely accepted as the main type of DNA damage responsible for cell killing in the range of doses and dose rates relevant to radiation therapy. Although the standard linear-quadratic (LQ) model with one first-order repair term often suffices to explain the results of some radiobiological experiments, converging lines of evidence suggest that DSBs are rejoined at two or more distinct rates. A two-lesion kinetic (TLK) model has been proposed to provide a direct link between biochemical processing of the DSBs and cell killing. A defining feature of the TLK model is that the family of all possible DSBs is subdivided into simple and complex DSBs, and each kind may have its own unique repair characteristics. Break-ends associated with both kinds of DSB are allowed to interact in pairwise fashion to form irreversible lethal and non-lethal chromosome aberrations. This paper examines the theoretical and practical linkages between the TLK and LQ models. The TLK formalism is used to derive an LQ formula with two first-order repair terms (dose protraction factors) and to relate the intrinsic radiosensitivity parameters used in one model to the parameters used in the other. Two extensive radiobiological datasets, one for CHO 10B2 cells and one for C3H 10T1/2 cells, are analysed using the TLK and LQ models. The LQ with two repair terms and the TLK are equally capable of explaining the CHO 10B2 and C3H 10T1/2 cell survival data. For the doses and dose rates most relevant to radiation therapy, tests of model equivalence indicate that an LQ formula with two first-order repair terms is an excellent approximation to the TLK model. We find the LQ and TLK models useful complementary tools for the analysis and prediction of radiobiological effects.

Comparison of in vitro and in vivo alpha/beta ratios for prostate cancer.

Parallel in vitro and in vivo studies provide insight into the relationship between clinical response and intrinsic cellular radiosensitivity and may aid in the development of predictive assays. Compilations of radiosensitivity parameters from in vitro experiments can also be used to examine the potential effectiveness of alternative or new treatment plan designs until enough clinical data become available to directly estimate the requisite radiosensitivity parameters. In this work, survival data for six prostate cancer cell lines (ten datasets total) have been extracted from the literature and re-analysed using the linear-quadratic (LQ) survival model. The paired bootstrap technique for regression is used to compute 95% confidence intervals for the estimated radiosensitivity parameters. LQ radiosensitivity parameters derived from the in vitro data are then compared to radiosensitivity parameters derived from clinical data for prostate cancer. Estimates of alpha range from 0.09 to 0.35 Gy(-1) (all cell lines), and the alpha/beta ratio ranges from 1.09 to 6.29 Gy (all cell lines). Point estimates of the repair half-time (PPC-1, TSU-Pr1, PC-3 and DU-145 cell lines) range from 5.7 to 8.9 h (95% confidence interval from 0.26 h to 10.7 h). Differences in the radiosensitivity parameters determined from the data reported by different laboratories are as large as or larger than the differences in radiosensitivity parameters observed among the various prostate cell lines. The reported studies demonstrate that even seemingly small corrections for dose rate effects, such as those expected in high dose rate (HDR) experiments, can sometimes have a significant impact on estimates of alpha and alpha/beta. By neglecting dose rate effects in the analysis of HDR experiments, estimates of the alpha/beta, ratio may be too high by factors as large as 1.3 to 6.2. The half-time for repair derived from the in vitro experiments appears significantly larger (slower repair rate) than estimates derived from the clinical data. However, the prostate radiosensitivity parameters alpha and alpha/beta may be approximately the same in vitro and in vivo. Most of the in vitro data are consistent with an alpha/beta ratio for prostate cancer less than 3 or 4 Gy.