FDG-PET provides the best correlation with the tumor specimen compared to MRI and CT in rectal cancer

Purpose

To compare CT-, MR- and PET-CT based tumor length measurements in rectal cancer with pathology.

Patients and Methods

Twenty-six rectal cancer patients underwent both MR and PET-CT imaging followed by short-course radiotherapy (RT 5 × 5 Gy) and surgery within 3 days after RT. Tumor length was measured manually and independently by 2 observers on CT, MR and PET. PET-based tumor length measurements were also generated automatically using the signal-to-background-ratio (SBR) method. All measurements were correlated with the tumor length on the pathological specimen.

Results

CT-based measurements did not show a valuable correlation with pathology. MR-based measurements correlated only weakly, but still significantly (Pearson correlation = 0.55 resp. 0.57; p < 0.001). Manual PET measurements reached a good correlation with pathology, but less strong (Pearson correlation 0.72 and 0.76 for the two different observers) than automatic PET-CT based measurements, which provided the best correlation with pathology (Pearson correlation of 0.91 (p < 0.001)). Bland-Altman analysis demonstrated in general an overestimation of the tumor diameter using manual measurements, while the agreement of automatic contours and pathology was within acceptable ranges. A direct comparison of the different modalities revealed a significant better precision for PET-based auto-contours as compared to all other measurements.

Conclusion

Automatically generated PET-CT based contours show the best correlation with the surgical specimen and thus provide a useful and powerful tool to accurately determine the largest tumor dimension in rectal cancer. This could be used as a quick and reliable tool for target delineation in radiotherapy. However, a 3D volume analysis is needed to confirm these results.     

Dose-effect relationship for local control of cervical cancer by magnetic resonance image-guided brachytherapy

Background and purpose

To analyse dose-response relationships for local control of cervical cancer after MR image-guided brachytherapy (IGBT) based on dose-volume histogram parameters.

Methods and materials

The analysis includes 141 patients with cervix cancer (stages IB-IVA) treated with 45-50.4 Gy EBRT ± cisplatin plus 4 × 7 Gy IGBT. Gross tumour volume (GTV), high risk clinical target volume (HR CTV) and intermediate risk CTV (IR CTV) were delineated and DVH parameters (D90, D100) were assessed. Doses were converted to the equivalent dose in 2 Gy (EQD2) using linear-quadratic model (α/β = 10 Gy). Groups of patients were formed according to tumour size at diagnosis (GTV_D) of 2-5 cm (group 1) or >5 cm (2), with subgroups of the latter for HR CTV size at first IGBT 2-5 cm (2a) or >5 cm (2b). Dose-response dependence for local recurrence was evaluated by logit analysis.

Results

Eighteen local recurrences in the true pelvis were observed. Dose-response analyses revealed a significant effect of HR CTV D100 (p = 0.02) and D90 (p = 0.005). The ED50-values for tumour control were 33 ± 15 Gy (D100) and 45 ± 19 Gy (D90). ED90-values were 67 Gy (95% confidence interval [50;104]) and 86 Gy [77;113], respectively.

Conclusions

A significant dependence of local control on D100 and D90 for HR CTV was found. Tumour control rates of >90% can be expected at doses >67 Gy and 86 Gy, respectively.     

Sigmoid dose using 3D imaging in cervical-cancer brachytherapy

Background and Purpose

To evaluate the proximity, variance, predictors of dose, and complications to the sigmoid in cervical-cancer brachytherapy using 3D planning.

Materials and methods

Over 36 months, 50 patients were treated for cervical cancer with either low-dose-rate (LDR) or high-dose-rate (HDR) brachytherapy. The distance from the central tandem to the sigmoid, the D0.1cc and the D2cc to the sigmoid, rectum and bladder doses, and toxicity were analyzed.

Results

The median sigmoid EQD2 D0.1cc and D2cc were 84 Gy and 68.3 Gy for HDR versus 71.1 Gy and 65.9 Gy for LDR (p = 0.02 and 0.98, respectively). Twenty percent of the HDR fractions required manipulation of the superior dwell positions to decrease the sigmoid dose. The median distance from the sigmoid to the tandem was 1.7 cm (range [rg], 0.1-6.16 cm) for HDR and 2.7 cm (rg, 1.17-4.52 cm) for LDR; from the sigmoid to the 100% isodose region the median distances were −0.1 cm (rg, −1.4 to 2.5 cm) and 0.44 cm (rg. −0.73-5.2 cm), respectively. The proximity of the sigmoid to the tandem is significantly related to sigmoid dose (p < 0.0001). Within-patient (among-fraction) variation in sigmoid-to-tandem distance during HDR was substantial (coefficient of variation =40%). No grade 3-4 sigmoid toxicity was seen after a median 31-month follow-up period.

Conclusions

3D imaging in cervical-cancer brachytherapy shows the sigmoid in close proximity to the tandem. The sigmoid-to-tandem distance varies substantially between fractions, indicating the importance of sigmoid dose-volume evaluation with each fraction.     

Prostate-specific antigen kinetics following external-beam radiotherapy and temporary (Ir-192) or permanent (I-125) brachytherapy for prostate cancer

Background and purpose

The aim of the study was the evaluation of PSA kinetics after different radiotherapy methods.

Materials and methods

Two-hundred and ninety five patients received external-beam radiotherapy (EBRT; 70.2 Gy; n = 135), Ir-192 brachytherapy as a boost to EBRT (HDR-BT; 18 Gy + 50.4 Gy; n = 66) or I-125 brachytherapy (LDR-BT; 145 Gy; n = 94) as monotherapy. “PSA bounce” was defined as a PSA rise of ?0.2 ng/ml followed by spontaneous return to prebounce level or lower, biochemical failure as “nadir + 2 ng/ml”.

Results

Patients without biochemical failure reached a lower nadir after brachytherapy (median ?0.05 ng/ml after LDR- and HDR-BT without NHT) in comparison to EBRT (0.55 ng/ml without NHT; p < 0.01). Not a single patient without NHT and a nadir <0.1 ng/ml failed biochemically (0% vs. 45% with a nadir ?0.1 ng/ml; p < 0.01). PSA bounces were found predominantly in the LDR-BT group (42% vs. 23%/20% after HDR-BT/EBRT; p < 0.01). In a multivariate Cox regression analysis, LDR-BT and HDR-BT were associated with a significantly lower biochemical failure rate in comparison to EBRT.

Conclusions

PSA kinetics differ significantly following different radiotherapy methods. A lower nadir and a higher biochemical control rate suggest a higher radiobiological efficiency of brachytherapy in comparison to EBRT (with a dose of 70.2 Gy).     

Localization of the surgical bed using supine magnetic resonance and computed tomography scan fusion for planification of breast interstitial brachytherapy

Purpose

To evaluate the feasibility of supine breast magnetic resonance imaging (MR) for definition and localization of the surgical bed (SB) after breast conservative surgery. To assess the inter-observer variability of surgical bed delineation on computed tomography (CT) and supine MR.

Materials and methods

Patients candidate for breast brachytherapy and no contra-indications for MR were eligible for this study. Patients were placed in supine position, with the ipsilateral arm above the head in an immobilization device. All patients underwent CT and MR in the same implant/treatment position. Four points were predefined for CT-MRI image fusion. The surgical cavity was drawn on CT then MRI, by three independent observers. Fusion and analysis of CT and MR images were performed using the ECLIPSE treatment planning software.

Results

From September 2005 to November 2008, 70 patients were included in this prospective study. For each patient, we were able to acquire axial T1 and T2 images of good quality. Using the predefined fusion points, the median error following the fusion was 2.7 mm. For each observer, volumes obtained on MR were, respectively, 30%, 38% and 40% smaller than those derived from CT images. A highly significant inter-observer variability in the delineation of the SB on CT was demonstrated (p < 0.0001). On the contrary, all three observers agreed on the volume of the SB drawn on MR.

Conclusion

Supine breast MRI yields a more precise definition of the SB with a smaller inter-observer variability than CT and may obviate the need for surgical clips. The volume of the SB is smaller with MRI. In our opinion, CT-MRI fusion should be used for SB delineation, in view of partial breast irradiation.     

Stereotactic body radiotherapy for local boost irradiation in unfavourable locally recurrent gynaecological cancer

Purpose

To evaluate outcome of radiotherapy for locally recurrent cervical and endometrial cancer.

Materials and methods

Nineteen patients were treated for a locally recurrent cervical (n = 12) or endometrial (n = 7) cancer median 26 months after initial surgery (n = 18) or radiotherapy (n = 1). The whole pelvis was irradiated with 50 Gy conventionally fractionated radiotherapy (n = 16). Because of large size of the recurrent cancer (median 4.5 cm) and peripheral location (n = 12), stereotactic body radiotherapy (SBRT; median 3 fractions of 5 Gy to 65%) was used for local dose escalation instead of (n = 16) or combined with (n = 3) vaginal brachytherapy.

Results

After median follow-up of 22 months, 3-year overall survival was 34% with systemic progression the leading cause of death (7/10). Median time to systemic progression was 16 months. Three local recurrences resulted in a local control rate of 81% at 3 years. No correlation between survival, systemic or local control and any patient or treatment characteristic was observed. The rate of late toxicity > grade II was 25% at 3 years: two patients developed a grade IV intestino-vaginal fistula and one patient suffered from a grade IV small bowel ileus.

Conclusion

Image-guided SBRT for local dose escalation resulted in high rates of local control but was associated with significant late toxicity.     

Applicator reconstruction for HDR cervix treatment planning using images from 0.35 T open MR scanner

Background and purposes

Magnetic resonance (MR) imaging is widely recognised as the modality of choice for imaging soft tissue such as the target volume and critical structures relevant to high dose rate (HDR) brachytherapy of the cervix. This work sets out to assess some of the issues faced when introducing this technique clinically compared to the more widely used computed tomography (CT). MR can be used either as the sole imaging modality, or in conjunction with CT.

Materials and methods

Distortion of the images produced by the MR scanner was assessed with a geometrical phantom. Distortion local to the titanium applicators, introduced by the susceptibility of the applicators themselves, was also measured. The technique used to reconstruct applicators is briefly described. An inter-operator study was performed to assess the variability of applicator reconstruction between operators when MR images are used alone to reconstruct the applicators.

Results

A 14-cm cube within which distortion was less than 2 mm at all points was identified. The inter-operator study showed some variability in applicator reconstruction with both MR and CT (median MR/CT 1.3 mm/0.9 mm, range 0-3.6 mm/0-3.3 mm). Inter-operator variation in planning target volume (PTV) V100% and PTV D90% for MR/CT was 6.1%/3.0% and 7.4%/6.3%, respectively, and D_2cc OAR doses varied by up to 1.0 Gy between operators for both MR and CT.

Conclusions

In this study distortion was minimal within a defined volume and inter-observer errors were comparable on MR and CT when used to define applicators and when applied to dose-volume histograms (DVHs). However this does not assure the technique is appropriate with other scanners and applicator sets without further commissioning.     

Interstitial brachytherapy using stereotactic implanted (125)iodine seeds for recurrent medulloblastoma

Aims

To evaluate the efficacy of interstitial brachytherapy using ¹²⁵iodine (¹²⁵I) seeds for the treatment of recurrent multimodal treated medulloblastoma.

Materials and methods

Between September 1989 and August 2009, 12 patients (female:male = 3:9, median age 19 years, range 7-55 years) with 23 recurrent medulloblastomas underwent interstitial brachytherapy using ¹²⁵I seeds. Before brachytherapy, all patients underwent microsurgical resection; six patients underwent a combined adjuvant treatment consisting of craniospinal irradiation and chemotherapy; three received craniospinal irradiation alone and two received chemotherapy alone. One patient was treated by surgery alone. The median tumour volume was 4.9 ml (range 0.4-44.2 ml), the median tumour surface dose 50 Gy (range 32-50 Gy) and the median implantation time 42 days (range 42-90 days). A median follow-up of 26 months was available (range 5-116 months).

Results

After brachytherapy, nine of 23 tumours (39%) presented a complete remission, nine (39%) a partial remission and five (22%) stable disease on magnetic resonance images. The neurological status improved in six patients and remained unchanged in four. Two patients deteriorated: one developed spinal metastasis and another a treatment-related adverse radiation effect. Ten patients died due to disseminated disease despite local tumour control. The median survival after treatment was 15 months (range 5-68 months).

Conclusions

Our results show a good response of recurrent medulloblastoma after interstitial brachytherapy. High rates of tumour remission were yielded with low rates of treatment-related morbidity. Thus, ¹²⁵I seed brachytherapy should be considered as a treatment option for recurrent medulloblastoma.     

Clinical outcome of protocol based image (MRI) guided adaptive brachytherapy combined with 3D conformal radiotherapy with or without chemotherapy in patients with locally advanced cervical cancer

Background

To analyse the overall clinical outcome and benefits by applying protocol based image guided adaptive brachytherapy combined with 3D conformal external beam radiotherapy (EBRT) ± chemotherapy (ChT).

Methods

Treatment schedule was EBRT with 45-50.4 Gy ± concomitant cisplatin chemotherapy plus 4 × 7 Gy High Dose Rate (HDR) brachytherapy. Patients were treated in the “protocol period” (2001-2008) with the prospective application of the High Risk CTV concept (D90) and dose volume constraints for organs at risk including biological modelling. Dose volume adaptation was performed with the aim of dose escalation in large tumours (prescribed D90 > 85 Gy), often with inserting additional interstitial needles. Dose volume constraints (D_2cc) were 70-75 Gy for rectum and sigmoid and 90 Gy for bladder.Late morbidity was prospectively scored, using LENT/SOMA Score. Disease outcome and treatment related late morbidity were evaluated and compared using actuarial analysis.

Findings

One hundred and fifty-six consecutive patients (median age 58 years) with cervix cancer FIGO stages IB-IVA were treated with definitive radiotherapy in curative intent. Histology was squamous cell cancer in 134 patients (86%), tumour size was >5 cm in 103 patients (66%), lymph node involvement in 75 patients (48%). Median follow-up was 42 months for all patients.Interstitial techniques were used in addition to intracavitary brachytherapy in 69/156 (44%) patients. Total prescribed mean dose (D90) was 93 ± 13 Gy, D_2cc 86 ± 17 Gy for bladder, 65 ± 9 Gy for rectum and 64 ± 9 Gy for sigmoid.Complete remission was achieved in 151/156 patients (97%). Overall local control at 3 years was 95%; 98% for tumours 2-5 cm, and 92% for tumours >5 cm (p = 0.04), 100% for IB, 96% for IIB, 86% for IIIB. Cancer specific survival at 3 years was overall 74%, 83% for tumours 2-5 cm, 70% for tumours >5 cm, 83% for IB, 84% for IIB, 52% for IIIB. Overall survival at 3 years was in total 68%, 72% for tumours 2-5 cm, 65% for tumours >5 cm, 74% for IB, 78% for IIB, 45% for IIIB.In regard to late morbidity in total 188 grade 1 + 2 and 11 grade 3 + 4 late events were observed in 143 patients. G1 + 2/G3 + 4 events for bladder were n = 32/3, for rectum n = 14/5, for bowel (including sigmoid) n = 3/0, for vagina n = 128/2, respectively.

Interpretation

3D conformal radiotherapy ± chemotherapy plus image (MRI) guided adaptive intracavitary brachytherapy including needle insertion in advanced disease results in local control rates of 95-100% at 3 years in limited/favourable (IB/IIB) and 85-90% in large/poor response (IIB/III/IV) cervix cancer patients associated with a moderate rate of treatment related morbidity. Compared to the historical Vienna series there is relative reduction in pelvic recurrence by 65-70% and reduction in major morbidity. The local control improvement seems to have impact on CSS and OS. Prospective clinical multi-centre studies are mandatory to evaluate these challenging mono-institutional findings.     

Comparison of DVH parameters and loading patterns of standard loading, manual and inverse optimization for intracavitary brachytherapy on a subset of tandem/ovoid cases

Purpose

Comparison of inverse planning with the standard clinical plan and with the manually optimized plan based on dose-volume parameters and loading patterns.

Materials and methods

Twenty-eight patients who underwent MRI based HDR brachytherapy for cervix cancer were selected for this study. Three plans were calculated for each patient: (1) standard loading, (2) manual optimized, and (3) inverse optimized. Dosimetric outcomes from these plans were compared based on dose-volume parameters. The ratio of Total Reference Air Kerma of ovoid to tandem (TRAK_O/T) was used to compare the loading patterns.

Results

The volume of HR CTV ranged from 9-68 cc with a mean of 41(±16.2)cc. Mean V100 for standard, manual optimized and inverse plans was found to be not significant (p = 0.35, 0.38, 0.4). Dose to bladder (7.8 ± 1.6 Gy) and sigmoid (5.6 ± 1.4 Gy) was high for standard plans; Manual optimization reduced the dose to bladder (7.1 ± 1.7 Gy p = 0.006) and sigmoid (4.5 ± 1.0 Gy p = 0.005) without compromising the HR CTV coverage. The inverse plan resulted in a significant reduction to bladder dose (6.5 ± 1.4 Gy, p = 0.002). TRAK was found to be 0.49(±0.02), 0.44(±0.04) and 0.40(±0.04)cGy m⁻² for the standard loading, manual optimized and inverse plans, respectively. It was observed that TRAK_O/T was 0.82(±0.05), 1.7(±1.04) and 1.41(±0.93) for standard loading, manual optimized and inverse plans, respectively, while this ratio was 1 for the traditional loading pattern.

Conclusions

Inverse planning offers good sparing of critical structures without compromising the target coverage. The average loading pattern of the whole patient cohort deviates from the standard Fletcher loading pattern.     

Comparative study of respiratory motion correction techniques in cone-beam computed tomography

Background and purpose

To validate the clinical usefulness of motion-compensated (MC) cone-beam (CB) computed tomography (CT) for image-guided radiotherapy (IGRT) in comparison to four-dimensional (4D) CBCT and three-dimensional (3D) CBCT.

Material and methods

Forty-eight stereotactic body radiation therapy (SBRT) patients were selected. Each patient had 5-12 long CB acquisitions (4 min) and 1-7 short CB acquisitions (1 min), with a total of 349 and 150 acquisitions, respectively. 3D, 4D and MC CBCT images of every acquisition were reconstructed. Image quality, tumor positioning accuracy and tumor motion amplitude were quantified.

Results

The mean image quality of long short acquisitions, measured using the correlation ratio with the planning CT, was 74%/70%, 67%/47% and 79%/74% for 3D, 4D and MC CBCT, respectively; both 4D and MC CBCT were corrected for respiratory motion artifacts but 4D CBCTs suffered from streak artifacts. Tumor positioning with MC CBCT was significantly closer to 4D CBCT than 3D CBCT (p < 0.0001). Detailed patient analysis showed that motion correction was not required for tumors with less than 1 cm motion amplitude.

Conclusions

4D and MC CBCT both allow accurate tumor position analysis under respiratory motion but 4D CBCT requires longer acquisition time than MC CBCT for adequate image quality. MC CBCT can therefore advantageously replace 4D CBCT in clinical protocols for patients with large motion to improve image quality and reduce acquisition time.     

Determining DVH parameters for combined external beam and brachytherapy treatment: 3D biological dose adding for patients with cervical cancer

Purpose

To compare two methods of DVH parameter determination for combined external beam and brachytherapy treatment of cervical cancer.

Materials and methods

Clinical treatment plans from five patients were used in this study. We simulated two applications given with PDR (32 × 60 cGy per application, given hourly) or HDR (4 × 7 Gy in two applications; each application of two fractions of 7 Gy, given within 17 h) standard and optimised treatment plans, all combined with IMRT (25 × 1.8 Gy). Additionally, we simulated an external beam (EBRT) boost to pathological lymph nodes or the parametrium (7 × 2 Gy).We determined D90 of the high-risk CTV (HR-CTV) and D_2cc of bladder and rectum in EQD2 in two ways. (1) ‘Parameter adding’: assuming a uniform contribution of the EBRT dose distribution and adding the values of DVH parameters for the two brachytherapy insertions, and (2) ‘distributions adding’: summing 3D biological dose distributions of IMRT and brachytherapy plans and subsequently determining the values of the DVH parameters. We took α/β = 10 Gy for HR-CTV, α/β = 3 Gy otherwise and half-time of repair 1.5 h.

Results

Without EBRT boost, ‘parameter adding’ yielded a good approximation. With an EBRT boost to lymph nodes, the total D90 HR-CTV was underestimated by 2.6 (SD 1.3)% for PDR and 2.8 (SD 1.4)% for HDR. This was even worse with a parametrium boost: 9.1 (SD 6.2)% for PDR and 9.9 (SD 6.2)% for HDR.

Conclusion

Without an EBRT boost ‘parameter adding’, as proposed by the GEC-ESTRO, yielded accurate results for the values for DVH parameters. If an EBRT boost is given ‘distributions adding’ should be considered.     

Increased use of regional radiotherapy is associated with improved outcome in a population-based cohort of women with breast cancer with 1-3 positive nodes

Purpose

To examine use of and outcomes from adjuvant locoregional radiotherapy (LRRT) after breast-conserving surgery (BCS) for women with breast cancer with 1-3 positive nodes (1-3 N+) before and after the 1997 publication of randomised trial evidence of a survival advantage from post-mastectomy LRRT.

Methods

Data were analysed for 2768 women diagnosed between 1989 and 2005 and referred to the British Columbia Cancer Agency with newly diagnosed pT1-3 breast cancer with 1-3 N+, treated with BCS and RT. LRRT use was analysed over time. Ten-year Kaplan-Meier locoregional control (LRC), breast cancer-specific survival (BCSS) and overall survival (OS) curves were compared using the log-rank test. Cox regression modeling of LRC and BCSS were performed.

Results

LRRT use in patients with 1-3 N+ increased from 23% before 1997 to 57% after 1997. LRRT was associated with significant improvements in LRC, but not in DRFS, BCSS, or OS. 10--year LRC was 89% with local RT alone and 93% with LRRT (p = 0.006). On multivariable analysis, LRRT was associated with improved LRC compared to local RT alone (HR 0.55, 95% CI: 0.40-0.77), but not with significant BCSS differences. Margin status, grade, % positive nodes, and hormonal therapy were significant predictors for LRC, while tumour size, grade, % positive nodes, and hormonal therapy significantly affected BCSS.

Conclusion

Post-BCS LRRT use in British Columbia increased almost threefold in patients with 1-3 N+ after 1997. Adjuvant LRRT was associated with improved LRC, but not with improved BCSS compared to breast RT alone.     

Applicator reconstruction in MRI 3D image-based dose planning of brachytherapy for cervical cancer

Background and purpose

To elaborate a method for applicator reconstruction for MRI-based brachytherapy for cervical cancer.

Materials and methods

Custom-made plastic catheters with a copper sulphate solution were made for insertion in the source channels of MR-CT compatible applicators: plastic and titanium tandem ring applicators, and titanium needles. The applicators were CT and MR scanned in a phantom for accurate 3D assessment of applicator visibility and geometry. A reconstruction method was developed and evaluated in 19 patient MR examinations with ring applicator (plastic: 14, titanium: 5). MR applicator reconstruction uncertainties related to inter-observer variation were evaluated.

Results

The catheters were visible in the plastic applicator on T1-weighted images in phantom and in 14/14 clinical applications. On T2-weighted images, the catheters appeared weaker but still visible in phantom and in 13/14 MR clinical applications. In the titanium applicator, the catheters could not be separated from the artifacts from the applicator itself. However, these artifacts could be used to localize both titanium ring applicator (5/5 clinical applications) and needles (6/6 clinical applications). Standard deviations of inter-observer differences were below 2 mm in all directions.

Conclusion

3D applicator reconstruction based on MR imaging could be performed for plastic and titanium applicators. Plastic applicators proved well to be suited for MRI-based reconstruction. For improved practicability of titanium applicator reconstruction, development of MR applicator markers is essential. Reconstruction of titanium applicator and needles at 1.5 T MR requires geometric evaluations in phantoms before using the applicator in patients.