|
Journal of Cancer Research and Therapeutics, Vol. 5, No. 2, April-June, 2009, pp. 78-84 Original Article Normal tissue complication probability: Does simultaneous integrated boost intensity-modulated radiotherapy score over other techniques in treatment of prostate adenocarcinoma Jothy Basu KS, Bahl Amit, Subramani V, Sharma DN, Rath GK, Julka PK Department of Radiation Oncology, Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi Code Number: cr09023 DOI: 10.4103/0973-1482.52789 Abstract Aim: The main objective of this study was to analyze the radiobiological effect of different treatment strategies on high-risk prostate adenocarcinoma.Materials and Methods: Ten cases of high-risk prostate adenocarcinoma were selected for this dosimetric study. Four different treatment strategies used for treating prostate cancer were compared. Conventional four-field box technique covering prostate and nodal volumes followed by three-field conformal boost (3D + 3DCRT), four-field box technique followed by intensity-modulated radiotherapy (IMRT) boost (3D + IMRT), IMRT followed by IMRT boost (IMRT + IMRT), and simultaneous integrated boost IMRT (SIBIMRT) were compared in terms of tumor control probability (TCP) and normal tissue complication probability (NTCP). The dose prescription except for SIBIMRT was 45 Gy in 25 fractions for the prostate and nodal volumes in the initial phase and 27 Gy in 15 fractions for the prostate in the boost phase. For SIBIMRT, equivalent doses were calculated using biologically equivalent dose assuming the α/β ratio of 1.5 Gy with a dose prescription of 60.75 Gy for the gross tumor volume (GTV) and 45 Gy for the clinical target volume in 25 fractions. IMRT plans were made with 15-MV equispaced seven coplanar fields. NTCP was calculated using the Lyman-Kutcher-Burman (LKB) model. Results: An NTCP of 10.7 ± 0.99%, 8.36 ± 0.66%, 6.72 ± 0.85%, and 1.45 ± 0.11% for the bladder and 14.9 ± 0.99%, 14.04 ± 0.66%, 11.38 ± 0.85%, 5.12 ± 0.11% for the rectum was seen with 3D + 3DCRT, 3D + IMRT, IMRT + IMRT, and SIBIMRT respectively. Conclusions: SIBIMRT had the least NTCP over all other strategies with a reduced treatment time (3 weeks less). It should be the technique of choice for dose escalation in prostate carcinoma. Keywords: Intensity-modulated radiotherapy, normal tissue complication probability, prostate Introduction Adenocarcinoma of the prostate is a slowly proliferating tumor with limited radio sensitivity, almost close to that of normal tissues. Higher doses of radiation have been associated with a better tumor control. [1] Four randomized trials have demonstrated improved prostate-specific antigen (PSA) relapse-free survival outcomes in patients treated with escalated radiation doses. [2],[3],[4],[5] Long-term results of single-institution trials using radiation doses up to 86.4 Gy have shown that higher doses are associated with an improved tumor control and a reduction in the risk of distant metastases. [6],[7],[8] But, the anatomic location of the prostate restricts the dose that can safely be delivered without exceeding the tolerance limits of surrounding critical structures. Conventionally, prostate tumors are treated with a two-phase treatment plan, with initial whole- pelvis treatment followed by a boost. Conventional techniques including three-dimensional conformal radiotherapy (3DCRT) are limited in their ability to deliver high doses to the prostate, due to the normal tissue complications. However, the advent of intensity-modulated radiotherapy (IMRT) has paved the way for dose escalation while keeping the critical structure doses within the tolerance limit. Zelefsky et al. have shown that IMRT significantly reduced the gastrointestinal (GI) and genitourinary (GU) toxicities compared to 3DCRT. Burman et al. have shown that IMRT significantly decreases the rectal morbidity even with high prescription doses.[9] Recently, Mohan et al. and Wu et al. suggested a single-phase approach called as simultaneous integrated boost IMRT (SIBIMRT) for the head and neck cancers, which increases the target dose conformity while simultaneously increasing the critical structure sparing. [10] SIBIMRT has certain advantages over a two-phase strategy and few of them are listed below:
Materials and Methods Ten cases of high-risk prostate adenocarcinoma treated earlier were selected for this dosimetric study. Four different boost strategies were designed for this study. All the contouring and conventional and inverse planning were made with Eclipse TM TPS (Varian Associates, Palo Alto, CA, USA). The four different boost strategies used in this study were (i) simple four-field box technique covering the prostate and the regional lymph nodes followed by a 3DCRT boost treating the prostate alone (3D + 3DCRT); (ii) four-field box technique followed by an IMRT boost (3D + IMRT); (iii) IMRT designed to treat the contoured target volumes followed by an IMRT boost to the prostate (IMRT + IMRT); and (iv) simultaneous integrated boost IMRT (SIBIMRT). The clinical target volume (CTV) was given a uniform margin of 1 cm except posteriorly, where a margin of 0.5 cm was given to increase the rectal sparing. The volumes included in the respective techniques are given in [Table - 1]. CTV1 included the prostate, gross extracapsuar disease, and proximal seminal vesicles. CTV2 included distal seminal vesicles and periprostatic lymph nodes. The prescription doses were 45 Gy in 25 fractions for the initial phase and 27 Gy in 15 fractions for the boost phase for all the strategies except for the SIBIMRT. The equivalent doses for the SIBIMRT were calculated using the classical biologically equivalent dose (BED) equation (1) given below: where D is the total dose and d is the dose per fraction. The α/β value was assumed as 1.5 Gy, based on the most recent data available for prostate adenocarcinoma (1.5 Gy at 95% CI 0.8-2.2 and 1.49 at 95% CI 1.25-1.76). [11],[12] These reported values are substantially lower than the traditionally accepted value of 10 Gy for most tumor cells. The reevaluated value of the α/β ratio is now even lower than the best estimate value for normal tissue (α/β = 3 Gy). The low α/β ratio for prostate cancer might be explained by the low cell cycle rate, and high proportion of cells in the G 0 phase of the cell cycle. Treatment planning was done for the different techniques as follows. 3D + 3DCRT In this two-phase strategy, the initial phase was planned with the conventional four-field box technique (i.e., anteroposterior (AP), posteroanterior (PA), right lateral, and left lateral) with 15-MV X-rays. A dose of 45 Gy in 25 fractions was prescribed to the 95% isodose surface. This initial phase was followed by a boost dose of 27 Gy in 15 fractions for CTV1 with three 15-MV fields shaped with multi leaf collimator (MLC) (AP, right lateral and left lateral), avoiding the posterior field to decrease the rectal dose. The field portals and target volumes contoured are shown in [Figure 1a],[Figure 1b],[Figure 1c],[Figure 1d]. 3D + IMRT The initial phase was planned similar to the initial phase of the previous strategy. The boost phase was planned to be delivered with IMRT with seven equally spaced 15-MV coplanar fields. The dose--volume constraints used for this plan are given in [Table - 2]. The field placement is shown in [Figure - 2]. IMRT + IMRT This involved initial IMRT to CTV1 and CTV2 followed by a IMRT boost to the CTV1. The dose was kept same as in other techniques (45 Gy + 27 Gy). The field placement was same as in other IMRT plans used in this study. The dose-volume constraints used for the initial phase of IMRT are given in [Table - 2]. Even though it was possible to use the base dose option available in Eclipse TPS to incorporate the dose delivered in the initial phase of the treatment in the optimization of the boost plan, it was deliberately avoided as it might have produced unacceptable hot and cold spots within the target. SIBIMRT This single-phase strategy was planned to deliver a dose of 60.75 Gy (2.43 Gy/fraction) to CTV1 and 45 Gy (1.8 Gy/fraction) to CTV2 in 25 fractions. Field placements were similar as in other IMRT plans made in this study. The dose-volume constraints used for this planning are given in [Table - 2]. Radiobiological models were used to evaluate the results. We used equivalent uniform dose (EUD)-based mathematical models, due to the simplicity, versatility, and also the ability to use the same model for NTCP calculations. The original definition of the EUD was based on the mechanistic formulation using the linear-quadratic (LQ) cell survival model. The concept of the EUD introduced by Niemierko is defined as ′the dose which when distributed uniformly across the target volume causes the survival of the same number of clonogens as the non-uniform dose distribution.′ Niemierko extended the concept of the EUD to apply to normal tissues and tumors as well, and it is known as the generalized EUD. [13] The mathematical expression of the generalized EUD is given in Equation 2: where V i is a unitless partial volume receiving dose D i in Gy and a is a unitless tumor or normal tissue-specific parameter that describes the dose-volume effect. For a tumor, it indicates the dose, which results in the same cell-kill as the nonuniform dose distribution; for normal tissues, it indicates the dose that results in the same probability of complications. The EUD approaches the maximum dose when a increases to a large positive number and approaches the minimum dose when it decreases to a large negative number; it approaches the mean dose when a equals 1 and approaches the geometric mean when a equals 0. EUDs were calculated from differential dose volume histograms (DVH) with a = 8.33 for the rectum and a = 2 for the bladder. For SIBIMRT, the concept of voxel equivalent dose was used to calculate the EUD to compare the nonstandard fractionation dose to the standard fractionation dose (2 Gy/fraction) by converting the individual voxel doses to conventional fractionation doses using the BED with an α/β value of 1.5, 6, and 3.9 for the prostate tumor, bladder, and rectum respectively. The TCP was calculated using the formula given below (equation 3) with D 50 = 46.3 and γ50 = 0.95 as estimated in the work of Okunieff et al , [14] where D ijk is the dose received by the ijk th voxel and N is the total number of voxels in the structure of interest. For SIBIMRT, D ijk is the equivalent dose calculated using the BED to account for the nonstandard fractionation: The NTCP was calculated using the Lyman-Kutcher-Burman (LKB) model as follows (equation 4): In the above equation (5), D is the EUD delivered to the structure of interest (calculated using Equation 2), D 50 is the dose in Gy required to produce a complication probability of 50% after 5 years, and m is a unitless parameter accounting for the volume effect (determining the steepness of the model curve). The parameters used in the calculation of TCP and NTCP are given in [Table - 3]. Results The results are summarized in [Table - 4],[Table - 5],[Table - 6] for bladder, rectum and prostate respectively. The equivalent dose-volume histograms for four techniques are given in [Figure - 1],[Figure - 2],[Figure - 3] for the bladder, rectum, and prostate respectively. The mean percent difference in the EUD for the bladder is −0.87% ( P -value 0.199), −2.92% ( P = 0.003), −11.58% ( P < 0.001) for 3D + IMRT, IMRT + IMRT, and SIBIMRT, respectively, compared with 3D + 3DCRT. The mean percent difference in the EUD for the rectum is −0.39% ( P = 0.019), −2.91% ( P = 0.003), and −10.75% ( P = 0.002) for 3D + IMRT, IMRT + IMRT, and SIBIMRT, respectively, compared with 3D + 3DCRT. The mean percent difference for PTV1 is −0.87% ( P -value 0.199), −2.92% ( P = 0.003), and −11.58% ( P < 0.001) for 3D + IMRT, IMRT + IMRT, and SIBIMRT, respectively, compared with 3D + 3DCRT. Among all the four boost strategies, the SIBIMRT plan showed the least NTCP for the bladder and rectum with a very small increment in the TCP of the prostate tumor. The mean percent difference in the NTCP was −16.32%, −32.41%, and −85.13% for the bladder and −6.02%, −23.91%, and −64.29% for the rectum, respectively, for 3D + IMRT, IMRT + IMRT, and SIBIMRT compared with 3D + 3DCRT. The mean percent difference in the TCP was 0.31%, 0.95% and 2.68% for 3D + IMRT, IMRT + IMRT, and SIBIMRT, respectively, compared with 3D + 3DCRT. The dose distributions for different techniques are shown in [Figure 2a],[Figure 2b],[Figure 2c],[Figure 2d]. The error bars and DVHs for different techniques are shown in [Figure - 3],[Figure - 4],[Figure - 5].Discussion The radiation dose escalation in prostate carcinoma has largely been determined by the dose received by the bladder and rectum. The TD 5/5 (the probability of 5% complications within 5 years of treatment) parameters used for the conventional treatment of whole bladder and rectum are 65 and 60 Gy, respectively. [15] The reported rates of cumulative acute Grade ≥ 2 bladder and bowel toxicity seen using conformal radiotherapy of a dose of 64 Gy/32 fractions and 74 Gy/37 fractions are 38% and 30%, and 39% and 33%, respectively.[5] Other studies have reported Grade ≥2 bladder morbidity ranging from 45 to 53% and bowel morbidity ranging from 55 to 56% using standard doses.[16],[17] Late bladder and bowel toxicity of 25 and 21% has been reported after conformal radiotherapy. [18] IMRT has provided a means for escalating doses up to 86 Gy in treatment of prostate adenocarcinoma. [19] However, the evaluation of normal tissue toxicity both acute and late is important at such high doses. IMRT has shown to produce the most conformal doses when delivered as SIB. [10] The TCP and NTCP are useful and essential parameters when planning to deliver very high doses as they provide the risk associated with the treatment. Sometimes such parameters may guide us how much dose we can deliver safely to the tumor. Our results in this study show that SIBIMRT can produce a better dose distribution and the least NTCP when compared with 3D conformal radiotherapy plans and conventional IMRT plans. The slight increase in the TCP values of IMRT plans compared to 3DCRT may be attributed to the increased uniformity and coverage achieved with IMRT plans. Even though SIBIMRT did not increase the TCP much, it has shown a considerable drop in the NTCP of both bladder and rectum. A better sparing of normal tissues has also been reported when using this technique in gynecological malignancies. [20] Grade 2 GU toxicity of 41% and GI toxicity of 11% have been reported in dose escalation studies in prostate cancer using SIBIMRT. [21] This compares favorably with results reported at lower doses using more conventional treatments. However, Stavrev et al. have reported a reduction of 5-7% in the TCP when comparing SIBIMRT delivered in the same number of fractionations as the two-phase technique in prostate treatments and suggested that SIB may not be the best treatment strategy unless fractionation schedules are carefully designed. SIBIMRT appears to be a promising technique in prostatic adenocarcinoma. Apart from the enhanced radiobiological effectiveness, SIBIMRT offers an increased machine throughput as the entire treatment is completed in a much shorter time (3 weeks less for prostate cases in our study) and lesser planning and verification time. However, further evaluation using parameters like TCP and NTCP is warranted in bigger trials.Conclusion Our study showed that SIBIMRT would produce the least normal tissue complications with almost the same amount of tumor control compared to conventional two-phase treatments. Moreover, SIBIMRT can produce a better physical dose distribution by finding better mathematical solutions by inverse planning techniques. The TCP and NTCP can play a vital role in planning and evaluation when delivering very high doses in individual patients. SIBIMRT can also increase the machine throughput as the treatments are delivered in a shorter period compared to a two-phase treatment. References
Copyright 2009 - Journal of Cancer Research and Therapeutics The following images related to this document are available:Photo images[cr09023t4.jpg] [cr09023f4.jpg] [cr09023f2.jpg] [cr09023t3.jpg] [cr09023f1.jpg] [cr09023f3.jpg] [cr09023t5.jpg] [cr09023t2.jpg] [cr09023t6.jpg] [cr09023f5.jpg] [cr09023t1.jpg] |
|