S. Ayyagari

Variations of intracavitary applicator geometry during multiple HDR brachytherapy insertions in carcinoma cervix and its influence on reporting as per ICRU report 38

PURPOSE This paper examines the extent of variation in the applicator geometry during multiple high dose rate (HDR) intracavitary brachytherapy (ICBT) applications and its impact on reporting as per ICRU report 38. MATERIALS AND METHODS Eighty orthogonal radiographs from 20 consecutive patients of carcinoma cervix (FIGO stages, IIA-IIIB) having four HDR ICBT applications of 6 Gy each at weekly intervals following teletherapy were evaluated. The applicator consisted of a flexible intrauterine tandem (IUT) independent of the ovoid assembly. The applicator geometry was evaluated in terms of: alpha angle, beta angle, intrauterine length (IUTL), interovoid (IOV), os to right ovoid (ORT) and os to left ovoid (OLT) distances along with vertical (VDL) and anteroposterior displacements (ADL) of the os with respect to the ovoids. The Cartesian co-ordinates (X, Y, and Z) of the IUT tip, centre of both ovoids and os were also measured. Doses to right point A (ARD), left point A (ALD), along with a reference volume of 6 Gy for ICRU height (IRH), width (IRW), thickness (IRT) and volume (IRV) were estimated for each application. RESULTS Highly significant differences (P<0.001) between four insertions in any given patient across 20 patients for alpha angle, beta angle, IUTL, IOV, ORT, VDL, co-ordinates of the IUT, ovoids and os were observed, except for ADL (P=0.041) and OLT (P=0.247). As a consequence, variations were observed in ARD (P=0.027), ALD (P=0.017); IRH, IRW, IRT and IRV (all P<0.001). Applicator factors which influenced the various dose specification parameters were: beta angle and ORT for both ARD and ALD; UTLN, VDL and ORT for IRH; UTLN and IOV for IRW; UTLN for IRT and VDL for the 6 Gy IRV. CONCLUSIONS A significant variation of the applicator geometry and its movement was observed in patients undergoing multiple HDR ICBT. This could have implications for reporting dose and volume specifications as required by ICRU report 38.

Problems in reporting doses and volumes during multiple high-dose-rate intracavitary brachytherapy for carcinoma cervix as per ICRU Report 38: a comparative study using flexible and rigid applicators

OBJECTIVE The objective was to evaluate the extent of geometric and positional variations of two different applicators during multiple high-dose-rate (HDR) intracavitary brachytherapy (ICBT) in carcinoma cervix and its implication on reporting as per International Commission on Radiation Units and Measurement (ICRU) Report 38. METHODS Fifty patients, following teletherapy, were randomly allocated to two groups of 25 each. They received a dose of 6 Gy to point A during each of the three HDR ICBT applications by either a flexible Ralstron or a rigid geometry Rotterdam applicator. The various applicator components related to its geometry and their Cartesian coordinates were evaluated from orthogonal films. The doses to ICRU bladder, rectal, pelvic, lymphatic trapezoid points, and dimensions of 6-Gy ICRU height, width, thickness, and volume were estimated for each application. RESULTS Significant variation was observed with the three HDR ICBT applications for each group, for all components and for both applicators, although it was relatively more with the flexible Ralstron applicator. The average shift in each of the coordinates of os, uterine tip, and ovoids was around 10 mm for both groups. These resulted in significant variations in all the ICRU Report 38 reporting parameters for three insertions in any given patient and across 25 patients of both groups. CONCLUSIONS Multiple HDR ICBT applications led to significant variation in the applicator geometry and its positions in pelvis, irrespective of the applicator rigidity. This results in uncertainties in reporting as per ICRU Report 38 guidelines, and thus calls for its revision.

Feasibility of Non-Cisplatin-Based Induction Chemotherapy and Concurrent Chemoradiotherapy in Advanced Head and Neck Cancer

The purpose of the present study was to determine the safety and efficacy of induction chemotherapy followed by concomitant chemoradiotherapy. Thirty-eight patients were randomised to receive induction chemotherapy, consisting of cyclophosphamide and methotrexate followed by concomitant 5-fluorouracil and irradiation (study group) or irradiation alone (control group). There were non-significant differences in the initial tumor clearance rates in the two groups. Median disease-free survival (in complete responders) was 17 months (6-60+) vs 11 months (5-60+) (p = 0.407) and overall survival 11 months (1-60+) vs 14 months (2-60+) (p = 0.428) in the study and control groups respectively. Acute morbidity and deaths during intervention were higher in the study group (p = 0.007). This study suggests that induction along with concomitant chemoradiotherapy is too toxic for routine use and also fails to provide a survival benefit.

Electron beam therapy at extended SSDs: an analysis of output correction factors for a Mitsubishi linear accelerator

The effects of extended source-to-surface distance (SSD) on the electron beam dose profiles were evaluated for various electron beam energies--6, 9, 12, 15 and 20 MeV-and the accuracy of various output correction methods was analysed on a Mitsubishi linear accelerator using a radiation field analyser (RFA). The dose fall-off region of the central axis depth-dose curves was nearly independent for SSDs up to 120 cm where as in the build-up region, a marginal reduction of surface dose was observed, particularly for lower energies and for smaller field sizes. Effective SSDs and virtual source distances were evaluated for field sizes ranging from 5 x 5 to 15 x 15 cm2 for various energies. Curve fitting was done with the measured outputs with various equations and coefficients were evaluated. The accuracy of the derived output correction factors by effective SSD, virtual source distance and curve-fit methods was assessed by evaluating correlation coefficients between the calculated and the measured values. The correlation coefficient was best with the linear-quadratic equation followed by the effective SSD method and the virtual source method. The output correction based on the linear-quadratic equation showed the best estimate of electron beam output at extended SSDs with an accuracy well within +/- 1%. The rapid reduction of dose due to the applicator-scattered component at d(max) point with an extended SSD was significant for the 5 x 5 cm2 applicator and lower energies.

A non-randomized comparison of two radiotherapy protocols in inoperable squamous cell carcinoma of the oesophagus

We report an audit of two different telebrachytherapy schedules in inoperable carcinoma of the oesophagus. Between October 1990 and December 1996, 108 patients with a Karnofsky performance status > or = 50 were selected from our database on the basis of intention to treat by telebrachytherapy. Teletherapy in the low dose group L (55 patients) consisted of 35 Gy in 15 fractions over 3 weeks, while that in the high dose group H (53 patients) consisted of 50 Gy in 25 fractions over 5 weeks. The choice of teletherapy dose was based on physician preference. The high dose rate intraluminal radiotherapy that followed 2 weeks later was identical in both groups and consisted of two applications of 6 Gy, a week apart. The pretreatment disease characteristics of the patients in both arms were similar. Relief of dysphagia was obtained in 49% of the patients in group L and in 75% of those in group H (chi2: P = 0.004). The median dysphagia-free interval was 0 and 7 months in groups L and H respectively (log-rank: P = 0.06), while the median overall survival was identical at 8 months (log-rank: P = 0.21) for both groups. The probability of survival at 1, 2 and 5 years was 34.8% versus 35.8%, 14.5% versus 13.9% and 0% versus 10% for groups L and H respectively. Morbidity in the form of ulcers, strictures and fistulae were observed in 9%, 7% and 5% of patients in group L compared with 8%, 8% and 13% in groups H respectively. This audit suggests that the protocol used in group H, when compared with group L, results in a greater proportion of patients being rendered dysphagia free, with a statistical trend towards a greater sustainment of dysphagia relief on follow-up.

Partial Transmission Block for Optimization of Anterior Supraclavicular-Posterior Axillary Boost in the Radiation Therapy of CArcinoma Breast

Radiation therapy of breast often involves an anterior supraclavicular-axillary (SC-AX) portal to irradiate the supraclavicular and axillary contents. However, the fall-off of the dose in this region leads to an inhomogeneity that could result either during the use of a single anterior SC-AX field or even with the concomitant use of a posterior axillary boost. An attempt has been made to circumvent this inhomogeneity by the use of a partial transmission block that could be placed in the anterior SC-AX portal corresponding to the posterior axillary boost field. The details of the quantum of partial transmission block to be used for different axillary separations and the reference depths of the dose prescription has been evaluated and presented.

Midline shield for radiation therapy of carcinoma of the uterine cervix: Should it be “midline” or “individualized”?

This study attempts to evaluate the advantage of individualized midline shield (IMLS) constructed on the basis of uterine geometry and applicator position in terms of the dosimetric consequences to points A-right (AR) and left (AL) as compared to standard midline shield (SMLS) in radiation therapy of carcinoma cervix. Twenty consecutive patients of carcinoma cervix (Stage I, II and III) were treated by external beam radiotherapy (EBRT) (50 Gy/5 weeks/25 fractions) and high-dose rate intracavitary brachytherapy (24 Gy/4 weeks/4 fractions) prescribed at point A. At the completion of 40 Gy by EBRT (phase I), IMLS (5 cm wide) at isocenter was fabricated on the basis of uterine geometry as ascertained by a dummy intracavitary application. The remaining 10 Gy of EBRT was delivered using IMLS (phase II) to effectively minimize and optimize the dose to point A. The dose profiles of IMLS were compared against the corresponding dose profile of a 5 cm SMLS and were found to be dependent on the positional variation of AR and AL with respect to the midline. With IMLS, the dose to AR and AL for the 10 Gy of phase II varied between 21.7-38.87% (30.01 +/- 4.87) and 22.42-35.72% (28.12 +/- 3.79) respectively. However with SMLS, the AR and AL doses would have ranged from 22.03% to 77.26% (34.55 +/- 15.94) for AR and from 20.59% to 96.2% (46.93 +/- 28.15) for AL leading to considerable inhomogeneity. Thus, in protocols incorporating midline shield for radiotherapy of carcinoma cervix, IMLS in place of SMLS could be preferred for achieving a definitive and homogeneous dose to the points AR and AL.

Biological integral dose: an alternate method for numerical scoring of rival plans.

Numerical scoring of rival plans (NSRP) are usually based either on basis of dose-volume histograms (DVH) or the relative values of corresponding normal tissue complication probabilities (NTCP) and tumor control probabilities (TCP). An alternative method for NSRP based on biological integral dose (BID) is being proposed, which is illustrated using a case of pituitary tumor planned to receive a dose of 50 Gy in 25 fractions over 5 weeks. BID for the various alternate plans -2-field (2F), 3-field (3F), 220 degrees arc (ARC) and 3-field static multileaf collimator (MLC) were calculated using the integration of the product of extrapolated response dose and the corresponding mass of the tissue enclosed separately for tumor and the normal brain in the entire planned target volume or a selected range of dose (approximately 90% and above of the normalized dose). Ratios of the BID for the brain versus the tumor were obtained and the plans were ranked on the basis of the least value of this ratio. In all of these plans, although the DVHs for normal brain were different, the DVHs for tumor were almost identical. However, the BID values for brain for 2F, 3F, ARC, and MLC were 22.53 Joules (J), 21.176 J, 21.991 J, and 10.608 J, respectively, and for tumor 0.561 J, 0.552 J, 0.555 J, and 0.556 J, respectively. The corresponding brain/tumor values were 40.16 (2F), 38.36 (3F), 39.62 (ARC), and 19.08 (MLC), thus ranking the plans in order of merit as MLC, 3F, ARC, and 2F. The BID for volumes encompassed by 90% and more of the normalized dose magnified the differences between the plans, with 2F being 29.99, compared to 3.82 for MLC. Rankings of rival plans could be based on the concept of BID. It requires a lesser number of uncertain variables and therefore could be used as an alternative technique in evaluation of the different plans in routine clinical practice.

Spatial information on dose distribution using multisectional dose-volume histograms

Dose-volume histograms are useful tools to summarize the information on the dose profiles resulting within a target volume. However, the spatial relationships of the hot and the cold spots are blunted in the dose-volume histograms. This study tries to circumvent this problem using multisectional dose-volume histograms and highlight the utility of these in the optimization of a radiation therapy plan.

Multisectional planning for external beam radiotherapy: A “poor person's” alternative to three-dimensional treatment planning

Recent technical advances in the field of computers have led to the use of three-dimensional dose computation for optimizing a radiation therapy plan. However, in centers which lack such a state-of-the-art technology, one could explore the use of multisectional planning to obtain information about the dose profiles all along the target volume. This article highlights the utility of multisectional planning as an alternative to three-dimensional treatment planning systems for external beam radiation therapy.