Measurement of Ambient Dose Equivalent in Compact Proton Therapy using In-house Neutron Moderator-based Poly Allyl Diglycol Carbonate.

Purpose: The high-energy proton produces the unwanted dose contribution from the secondary neutron. The main purpose of this study is to report the validation results of in-house neutron moderator based on poly allyl diglycol carbonate (CR-39) detector, Chulalongkorn University Neutron Moderator (CUMOD) through the ambient dose equivalent, H*(10) measurement. Materials and Methods: The Particle and Heavy Ion Transport code System (PHITS) Monte Carlo code was used to simulate the neutron response function. The CUMOD was calibrated with 241AmBe source calibrator in the range of 100-1000 µSv. The variation of neutron fields was generated employing different proton treatment plans covering most of the clinical scenarios. The ambient dose equivalents, H*(10), evaluated employing CUMOD were compared to those obtained with WENDI-II dosimeter. Results: The linear relationship between CUMOD and WENDI-II responses showed an R2 value close to 1. The H*(10) per Gy delivered dose was in the range of 22-105 µSv for a 10 cm × 10 cm field. Conclusion: The in-house CUMOD neutron moderator can expand the neutron detection dose range of CR-39 detector for ambient dose equivalent. The advantage of CUMODs is its capability to evaluate H*(10) in various positions simultaneously.

A multi-institutional evaluation of small field output factor determination following the recommendations of IAEA/AAPM TRS-483.

PURPOSE: The aim of this work was to test the implementation of small field dosimetry following TRS-483 and to develop quality assurance procedures for the experimental determination of small field output factors (SFOFs). MATERIALS AND METHODS: Twelve different centers provided SFOFs determined with various detectors. Various linac models using the beam qualities 6 MV and 10 MV with flattening filter and without flattening filter were utilized to generate square fields down to a nominal field size of 0.5 cm × 0.5 cm. The detectors were positioned at 10 cm depth in water. Depending on the local situation, the source-to-surface distance was either set to 90 cm or 100 cm. The SFOFs were normalized to the output of the 10 cm × 10 cm field. The spread of SFOFs measured with different detectors was investigated for each individual linac beam quality and field size. Additionally, linac-type specific SFOF curves were determined for each beam quality and the SFOFs determined using individual detectors were compared to these curves. Example uncertainty budgets were established for a solid state detector and a micro ionization chamber. RESULTS: The spread of SFOFs for each linac and field was below 5% for all field sizes. With the exception of one linac-type, the SFOFs of all investigated detectors agreed within 10% with the respective linac-type SFOF curve, indicating a potential inter-detector and inter-linac variability. CONCLUSION: Quality assurance on the SFOF measurements can be done by investigation of the spread of SFOFs measured with multiple detectors and by comparison to linac-type specific SFOFs. A follow-up of a measurement session should be conducted if the spread of SFOFs is larger than 5%, 3%, and 2% for field sizes of 0.5 cm × 0.5 cm, 1 cm × 1 cm, and field sizes larger than 2 cm × 2 cm, respectively. Additionally, deviations of measured SFOFs to the linac-type-curves of more than 7%, 3%, and 2% for field sizes 0.5 cm × 0.5 cm, 1 cm × 1 cm, and field sizes larger than 1 cm × 1 cm, respectively, should be followed up.

Validation of RapidPlan Knowledge-Based Model for Volumetric-Modulated Arc Therapy in Prostate Cancer.

Aims and Objectives: This study aims to investigate the viability of using RapidPlan (RP) knowledge-based (KB) treatment plans to initiate the new prostate volumetric-modulated arc therapy (VMAT) plans. Materials and Methods: The planning data for 120 prostate VMAT treatment plans were entered into the RP system's database. The database of previous VMAT plans was divided into four model groups for training in the RP system. The models were based on the numbers of 20, 60, and 120 prostate VMAT plans. The model of 120 plans used automated priority and manual priority for the optimization process. The models of 20 and 60 plans used only manual priority for optimization. Each model was validated on 15 cases of new prostate cancer patients by comparing RP model plans against manual clinical plans optimized according to the clinical dose constraints. Results: The RP models can estimate the dose comparable target volume to the manual clinical plan, which evaluated values of Dmax, D95%, D98%, HI, and CI and showed comparable results. For the normal organ doses of the bladder, rectum, penile bulb, and femoral head, all RP models exhibited a comparable or better dose than the manual clinical plan, except for the RP models using the automated priority for the optimization process, which cannot control the rectum dose below the dose constraints. Conclusions: The Varian RP KB planning can produce comparable doses or better doses with the clinical manual in a single optimization, although the RP model uses a minimum requirement of the planning number for the model training. The RP models can enhance the efficacy and quality of plans, which depend on the number of VMAT plans used in RP model training for prostate cancer.

Determination of field output correction factors of radiophotoluminescence glass dosimeter and CC01 ionization chamber and validation against IAEA-AAPM TRS-483 code of practice.

PURPOSE: To determine the field output correction factors of the radiophotoluminescence glass dosimeter (RPLGD) in parallel and perpendicular orientations with reference to CC01, the ionization chamber. METHODS: The dose to a small water volume and the sensitive volume of the RPLGD and the IBA-CC01 were determined for 6-MV, 100-cm SAD, 10-cm depth using egs_chamber user-code. The RPLGD in perpendicular and parallel orientations to the beam axis were studied. The field output correction factors of each detector for 0.5 × 0.5 to 10 × 10 cm2 field sizes were determined. These field output correction factors were validated by comparing field output factors against data determined from IAEA-AAPM TRS-483 code of practice. RESULTS: The field output correction factors of all detectors were within 5% for field sizes down to 0.8 × 0.8 cm2. For 0.5 × 0.5 cm2, the field output correction factors of CC01, RPLGD in perpendicular and parallel orientations differed from unity by 14%, 19%, and 5%, respectively. The percentage difference between field output factors determined using RPLGD and CC01 data, corrected using the field output correction factors determined in this work and measurements with CC01 data corrected using TRS-483, was less than 3% for all field sizes, except for the smallest field size of RPLGD in perpendicular orientation and the CC01. CONCLUSIONS: The field output correction factors of RPLGD and CC01 are reported. The validation proves that RPLGD in parallel orientation combined with the field output correction factors is the most suitable for determining the field output factors for the smallest field used in this study.

Estimated radiation doses to ovarian and uterine organs in breast cancer irradiation using radio-photoluminescent glass dosimeters (RPLDs).

INTRODUCTION: The well-being of breast cancer patients is essential, especially fertility in patients of reproductive age. The objective of this study was to estimate the radiation doses to the ovaries and uterus for different treatment techniques of breast cancer irradiation using radio-photoluminescent glass dosimeters (RPLDs). METHODS: A Farmer-type ionisation chamber (IBA FC-65G) and RPLDs were used to measure in- and out-of-field radiation doses in a solid water phantom. The field sizes were set to 10 × 10 cm2 and 8 × 17 cm2 with the central axis at out-of-field measurement distances of 30 or 50 cm. The Rando phantom's left breast was planned using four different techniques: two tangential standard fields with and without electronic tissue compensator (E-comp) techniques, intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). The radiation doses in the ipsilateral ovary, contralateral ovary and uterus were measured using RPLDs. RESULTS: The percentage ratio of out of field to in field was affected by distance from the central axis to the point of measurement, in addition to the field sizes associated with collimator scatter. Advanced techniques such as IMRT and VMAT produced higher doses to the ovaries and uterus. The estimated results of the worst-case scenario for the ipsilateral ovary, contralateral ovary and uterus were 0.84% (42 cGy), 0.62% (31 cGy) and 0.76% (38 cGy), respectively, for a 5000 cGy prescription dose. CONCLUSION: The lowest to highest out-of-field radiation doses to the ovarian and uterine organs from breast irradiation were the two tangential field techniques, VMAT and IMRT. These advanced techniques yielded higher radiation leakage, which potentially contributed to the out-of-field radiation dose.

The impact of corrected field output factors based on IAEA/AAPM code of practice on small-field dosimetry to the calculated monitor unit in eclipse™ treatment planning system.

The objective of this study was to investigate the effect of field output factors (FOFs) according to the current protocol for small-field dosimetry in conjunction to treatment planning system (TPS) commissioning. The calculated monitor unit (MU) for intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT) plans in Eclipse™ TPS were observed. Micro ion chamber (0.01 CC) (CC01), photon field diode (shielded diode) (PFD), and electron field diode (unshielded diode) (EFD) were used to measure percentage depth doses, beam profiles, and FOFs from 1 × 1 cm2 to 10 × 10 cm2 field sizes of 6 MV photon beams. CC01 illustrated the highest percentage depth doses at 10 cm depth while EFD exhibited the lowest with the difference of 1.6% at 1 × 1 cm2 . CC01 also produced slightly broader penumbra, the difference with other detectors was within 1 mm. For uncorrected FOF of three detectors, the maximum percent standard deviation (%SD) was 5.4% at 1 × 1 cm2 field size. When the correction factors were applied, this value dropped to 2.7%. For the calculated MU in symmetric field sizes, beam commissioning group from uncorrected FOF demonstrated maximum %SD of 6.0% at 1 × 1 cm2 field size. This value decreased to 2.2% when the corrected FOF was integrated. For the calculated MU in IMRT-SRS plans, the impact of corrected FOF reduced the maximum %SD from 6.0% to 2.5% in planning target volume (PTV) less than 0.5 cm3 . Beam commissioning using corrected FOF also decreased %SD for VMAT-SRS plans, although it was less pronounced in comparison to other treatment planning techniques, since the %SD remained less than 2%. The use of FOFs based on IAEA/AAPM TRS 483 has been proven in this research to reduce the discrepancy of calculated MU among three beam commissioning datasets in Eclipse™ TPS. The dose measurement of both symmetric field and clinical cases comparing to the calculation illustrated the dependence of the types of detector commissioning and the algorithm of the treatment planning for small field size.

An effective patient training for deep inspiration breath hold technique of left-sided breast on computed tomography simulation procedure at King Chulalongkorn Memorial Hospital.

PURPOSE: To observe the effectiveness of the practical instruction sheet and the educational video for left-sided breast treatment in a patient receiving deep inspiration breath hold (DIBH) technique. Two parameters, simulation time and patient satisfaction, were assessed through the questionnaire. METHODS: Two different approaches, which were the instruction sheet and educational video, were combinedly used to assist patients during DIBH procedures. The guideline was assigned at least 1 week before the simulation date. On the simulation day, patients would fill the questionnaire regarding their satisfaction with the DIBH instruction. The questionnaire was categorized into five levels: extremely satisfied to dissatisfied, sequentially. The patients were divided into four groups: not DIBH technique, DIBH without instruction materials, the DIBH with instruction sheet or educational video, and DIBH with both of instruction sheet and educational video. RESULTS: Total number of 112 cases of left-sided breast cancer were analyzed. The simulation time during DIBH procedure significantly reduced when patients followed the instruction. There was no significant difference in simulation time on the DIBH procedures between patient compliance via instruction sheet or educational video or even following both of them. The excellent level was found at 4.6 ± 0.1 and 4.5 ± 0.1, for patients coaching via instruction sheet as well as on the educational video, respectively. CONCLUSION: Patient coaching before simulation could potentially reduce the lengthy time in the simulation process for DIBH technique. Practicing the DIBH technique before treatment is strongly advised.

Impact of testicular shielding in liposarcoma to scrotum by using radio-photoluminescence glass dosimeter (RPLGD): a case report.

Radiation protection in the scrotum to reduce the risk of genetic effect in the future is very important. This study aimed to measure the scrotal dose outside the treatment fields by using the radio-photoluminescence glass dosimeter (RPLGD). The characteristics of RPLGD model GD-302M were studied. Scattered dose to scrotum was measured in one liposarcoma case with the prescribed dose of 60 Gy. RPLGDs were placed in three different locations: one RPLGD was positioned at the posterior area which closer to the scrotum, and the other two RPLGDs were placed between the penis and the scrotum. Three RPLGDs were employed in each location. The scattered doses were measured in every fraction during the whole course of treatment. The entire number of 100 RPLGDs showed the uniformity within ±2%. The signal from RPLGD demonstrated linear proportion to the radiation dose (r = 0.999). The relative energy response correction factor was 1.05. The average scrotal dose was 4.1 ± 0.9 cGy per fraction. The results presented a wide range since there was a high uncertainty during RPLGD placement. The total scrotal dose for the whole course of treatment was 101.9 cGy (1.7% of the prescribed dose). The RPLGD model GD-302M could be used to measure scattered dose after applying the relative energy correction factor.

Evaluation of equivalent dose from neutrons and activation products from a 15-MV X-ray LINAC.

A high-energy photon beam that is more than 10 MV can produce neutron contamination. Neutrons are generated by the [γ,n] reactions with a high-Z target material. The equivalent neutron dose and gamma dose from activation products have been estimated in a LINAC equipped with a 15-MV photon beam. A Monte Carlo simulation code was employed for neutron and photon dosimetry due to mixed beam. The neutron dose was also experimentally measured using the Optically Stimulated Luminescence (OSL) under various conditions to compare with the simulation. The activation products were measured by gamma spectrometer system. The average neutron energy was calculated to be 0.25 MeV. The equivalent neutron dose at the isocenter obtained from OSL measurement and MC calculation was 5.39 and 3.44 mSv/Gy, respectively. A gamma dose rate of 4.14 µSv/h was observed as a result of activations by neutron inside the treatment machine. The gamma spectrum analysis showed (28)Al, (24)Na, (54)Mn and (60)Co. The results confirm that neutrons and gamma rays are generated, and gamma rays remain inside the treatment room after the termination of X-ray irradiation. The source of neutrons is the product of the [γ,n] reactions in the machine head, whereas gamma rays are produced from the [n,γ] reactions (i.e. neutron activation) with materials inside the treatment room. The most activated nuclide is (28)Al, which has a half life of 2.245 min. In practice, it is recommended that staff should wait for a few minutes (several (28)Al half-lives) before entering the treatment room after the treatment finishes to minimize the dose received.

A method of setting limits for the purpose of quality assurance.

The result from any assurance measurement needs to be checked against some limits for acceptability. There are two types of limits; those that define clinical acceptability (action limits) and those that are meant to serve as a warning that the measurement is close to the action limits (tolerance limits). Currently, there is no standard procedure to set these limits. In this work, we propose an operational procedure to set tolerance limits and action limits. The approach to establish the limits is based on techniques of quality engineering using control charts and a process capability index. The method is different for tolerance limits and action limits with action limits being categorized into those that are specified and unspecified. The procedure is to first ensure process control using the I-MR control charts. Then, the tolerance limits are set equal to the control chart limits on the I chart. Action limits are determined using the Cpm process capability index with the requirements that the process must be in-control. The limits from the proposed procedure are compared to an existing or conventional method. Four examples are investigated: two of volumetric modulated arc therapy (VMAT) point dose quality assurance (QA) and two of routine linear accelerator output QA. The tolerance limits range from about 6% larger to 9% smaller than conventional action limits for VMAT QA cases. For the linac output QA, tolerance limits are about 60% smaller than conventional action limits. The operational procedure describe in this work is based on established quality management tools and will provide a systematic guide to set up tolerance and action limits for different equipment and processes.

Retrospective analysis of linear accelerator output constancy checks using process control techniques.

Shewhart control charts have previously been suggested as a process control tool for use in routine linear accelerator (linac) output verifications. However, a comprehensive approach to process control has not been investigated for linac output verifications. The purpose of this work is to investigate a comprehensive process control approach to linac output constancy quality assurance (QA). The RBA-3 dose constancy check was used to verify outputs of photon beams and electron beams delivered by a Varian Clinac 21EX linac. The data were collected during 2009 to 2010. Shewhart-type control charts, exponentially weighted moving average (EWMA) charts, and capability indices were applied to these processes. The Shewhart-type individuals chart (X-chart) was used and the number of data points used to calculate the control limits was varied. The parameters tested for the EWMA charts (smoothing parameter (λ) and the control limit width (L)) were λ = 0.05, L = 2.492; λ = 0.10, L = 2.703; and λ = 0.20, L = 2.860, as well as the number of points used to estimate the initial process mean and variation. Lastly, the number of in-control data points used to determine process capability (C(p)) and acceptability (C(pk)) were investigated, comparing the first in-control run to the longest in-control run of the process data. C(p) and C(pk) values greater than 1.0 were considered acceptable. The 95% confidence intervals were reported. The X-charts detected systematic errors (e.g., device setup errors). In-control run lengths on the X-charts varied from 5 to 30 output measurements (about one to seven months). EWMA charts showed in-control runs ranging from 9 to 33 output measurements (about two to eight months). The C(p) and C(pk) ratios are higher than 1.0 for all energies, except 12 and 20 MeV. However, 10 MV and 6, 9, and 16 MeV were in question when considering the 95% confidence limits. The X-chart should be calculated using 8-12 data points. For EWMA chart, using 4 data points is sufficient to calculate the initial mean and variance of the process. The EWMA limits should be calculated with λ = 0.10, L = 2.703. At least 25-30 in-control data points should be used to calculate the C(p) and C(pk) indices.

Statistical process control analysis for patient-specific IMRT and VMAT QA.

This work applied statistical process control to establish the control limits of the % gamma pass of patient-specific intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT) quality assurance (QA), and to evaluate the efficiency of the QA process by using the process capability index (Cpml). A total of 278 IMRT QA plans in nasopharyngeal carcinoma were measured with MapCHECK, while 159 VMAT QA plans were undertaken with ArcCHECK. Six megavolts with nine fields were used for the IMRT plan and 2.5 arcs were used to generate the VMAT plans. The gamma (3%/3 mm) criteria were used to evaluate the QA plans. The % gamma passes were plotted on a control chart. The first 50 data points were employed to calculate the control limits. The Cpml was calculated to evaluate the capability of the IMRT/VMAT QA process. The results showed higher systematic errors in IMRT QA than VMAT QA due to the more complicated setup used in IMRT QA. The variation of random errors was also larger in IMRT QA than VMAT QA because the VMAT plan has more continuity of dose distribution. The average % gamma pass was 93.7% ± 3.7% for IMRT and 96.7% ± 2.2% for VMAT. The Cpml value of IMRT QA was 1.60 and VMAT QA was 1.99, which implied that the VMAT QA process was more accurate than the IMRT QA process. Our lower control limit for % gamma pass of IMRT is 85.0%, while the limit for VMAT is 90%. Both the IMRT and VMAT QA processes are good quality because Cpml values are higher than 1.0.

A two-year experience of implementing 3 dimensional radiation therapy and intensity-modulated radiation therapy for 925 patients in King Chulalongkorn Memorial Hospital.

BACKGROUND AND OBJECTIVE: Three dimensional conformal radiation therapy (3D CRT) and intensity-modulated radiation therapy (IMRT) have been implemented at Department of Therapeutic Radiation and Oncology, King Chulalongkorn Memorial Hospital (KCMH) since July 2005. This is the first study in Thailand to evaluate the pattern of care and utilization of 3D CRT and IMRT for treatment in each individual cancer. MATERIAL AND METHOD: Between July 2005 and July 2007, 925 newly diagnosed cancer patients underwent IMRT or 3D CRT at KCMH. The authors retrospectively reviewed the experience and utilization of 3D CRT and IMRT for each disease site and region. RESULTS: There were 471 males and 454 females. There were 332 patients (35.9%) treated with IMRT. Among the 332 IMRT patients, there were 100, 32 and 27 nasopharyngeal, lung and prostate cancers, respectively. On the contrary, 593 patients (64.1%) were treated with 3D CRT. Among these, breast, cervix and lung cancers were the most common diseases. Except for head and neck as well as genitourinary cancer 3D CRT was still the main technique used in more than 60% of the patients at KCMH. CONCLUSION: 3D CRT and IMRT have been successfully implemented at KCMH for 2 years. Three dimensional conformal radiation therapy was still the main technique used in more than 60% of the patients at KCMH. Prospective studies evaluating tumor control and treatment sequelae are expected.

Radiation dose to medical staff in interventional radiology.

OBJECTIVE: The purposes of the present study were to determine the dose to medical staff in interventional radiology at different locations on the body measured by thermoluminescent dosimeter (TLD) and to relate the medical staff dose to patient dose measured by the dose-area product (DAP) meter. MATERIAL AND METHOD: The present study covered 42 patients in three interventional radiology procedures with three x-ray machines. Thermoluminescent dosimeters were stuck at eight positions on the radiologist's skin during the procedure. In addition, direct reading from the DAP meter placed in front of the collimator of the x-ray tube, was recorded to estimate the patient radiation dose. RESULTS: The surface dose to the primary radiologist showed maximum value at the left forearm of 407 microGy. The ratios between the maximum interventional radiologist surface dose and patient dose are 12.88 microGy per 10 Gycm2 for transarterial oily chemoembolization TOCE (Siemens Polystar), 22.58 microGy per 10 Gycm2 for transarterial oily chemoembolization TOCE (Siemens Neurostar), 148.29 microGy per 10 Gycm2 for percutaneous transhepatic biliary drainage PTBD (Siemens Polystar) and 100.46 microGy per 10 Gycm2for endoscopic retrograde cholangiopancreatography ERCP (GE Advantx). CONCLUSION: The interventional radiologist surface dose can be estimated from the mentioned ratio if the patient dose is measured. This will help the radiologists to avoid receiving an excess dose during their work.

Intensity-modulated radiation therapy in head-and-neck cancer, first report in Thailand.

OBJECTIVE: This is the first report in Thailand to evaluate the efficacy of using intensity-modulated radiotherapy (IMRT) in the primary treatment of head-and-neck cancer. MATERIAL AND METHOD: From July 2005 to March 2006, eighteen patients with head and neck cancer were treated with IMRT, fourteen of which were nasopharyngeal cancer. The median age at diagnosis was 52 years (range 23-58 years). The treatment plan composed of two sequential plans for PTV-low risk (50Gy in 25 fractions) and PTV-high risk (20Gy in 10 fractions). Chemotherapy was given to 13 patients with locoregionally advanced disease (stage T3/T4 and N2/3) using cisplatin (n = 3) or carboplatin (n = 10) every 3 weeks during the course of radiation therapy. RESULTS: The median overall treatment time was 49 days (range, 43-57 days), and 77.8 percent of the patients completed 35 fractions within 50 days. The clinical complete response and partial response rates at 3 months after complete radiation were 71.4% and 28.6%, respectively. However at the median follow-up of 5.6 months, the complete response rate increased to 89%. Treatment break during RT range from 3 to 7 days, was observed in three patients. All of them received concurrent chemoradiation. No distant metastasis was noted. CONCLUSION: The authors' experience of using concurrent chemotherapy with IMRT for a cohort of patients with head and neck carcinoma showed a very high rate response rate at early follow-up. Long-term clinical outcome is expected.

Phase III randomized trial comparing LDR and HDR brachytherapy in treatment of cervical carcinoma.

PURPOSE: Intracavitary brachytherapy plays an important role in the treatment of cervical carcinoma. Previous results have shown controversy between the effect of dose rate on tumor control and the occurrence of complications. We performed a prospective randomized clinical trial to compare the clinical outcomes between low-dose-rate (LDR) and high-dose-rate (HDR) intracavitary brachytherapy for treatment of invasive uterine cervical carcinoma. METHODS AND MATERIALS: A total of 237 patients with previously untreated invasive carcinoma of the uterine cervix treated at King Chulalongkorn Memorial Hospital were randomized between June 1995 and December 2001. Excluding ineligible, incomplete treatment, and incomplete data patients, 109 and 112 patients were in the LDR and HDR groups, respectively. All patients were treated with external beam radiotherapy and LDR or HDR intracavitary brachytherapy using the Chulalongkorn treatment schedule. RESULTS: The median follow-up for the LDR and HDR groups was 40.2 and 37.2 months, respectively. The actuarial 3-year overall and relapse-free survival rate for all patients was 69.6% and 70%, respectively. The 3-year overall survival rate in the LDR and HDR groups was 70.9% and 68.4% (p = 0.75) and the 3-year pelvic control rate was 89.1% and 86.4% (p = 0.51), respectively. The 3-year relapse-free survival rate in both groups was 69.9% (p = 0.35). Most recurrences were distant metastases, especially in Stage IIB and IIIB patients. Grade 3 and 4 complications were found in 2.8% and 7.1% of the LDR and HDR groups (p = 0.23). CONCLUSION: Comparable outcomes were demonstrated between LDR and HDR intracavitary brachytherapy. Concerning patient convenience, the lower number of medical personnel needed, and decreased radiation to health care workers, HDR intracavitary brachytherapy is an alternative to conventional LDR brachytherapy. The high number of distant failure suggests that other modalities such as systemic concurrent or adjuvant chemotherapy might lower this high recurrence, especially in Stage IIB and IIIB.