- Email: [email protected]
Introducing the Intergroup 0116 Protocol of Adjuvant Chemo-radiotherapy in Gastric Cancer into Clinical Practice
Clinical Oncology (2003) 15: 378–382 doi:10.1016/S0936-6555(03)00154-7
Original Article Introducing the Intergroup 0116 Protocol of Adjuvant Chemo-radiotherapy in Gastric Cancer into Clinical Practice M. F. Back, S. Premsenthil, C. J. Wynne, T. P. Shakespeare Department of Radiation Oncology, The Cancer Institute, National University Hospital, Singapore ABSTRACT: Aims: The results of techniques from a well-conducted clinical trial are often difficult to reproduce when implemented in community oncology practice. The U.S. Intergroup 0116 protocol of adjuvant chemo-radiotherapy in gastric cancer presented in mid-2000 produced a survival advantage over surgery alone. The current study aims to determine the adherence with protocol design and delivery of radiation therapy (radiotherapy) in the initial 20 patients managed with the Intergroup 0116 protocol at The National University Hospital, Singapore. Materials and methods: A formal quality assurance audit was performed on clinical features, radiotherapy treatment charts and simulation films of the first 20 patients treated with the Intergroup 0116 protocol from July 2000 to September 2001. Specific details were audited for their consistency with described protocol in domains of eligibility criteria, radiotherapy prescription, target volume coverage and adherence to dose-limiting normal tissue tolerances. Compliance and toxicity with the protocol was assessed by audit of delivered radiotherapy dose, treatment interruptions, inpatient admissions and weight loss during radiotherapy. Results: The 20 audited patients were appropriately selected on the basis of eligibility criteria of Intergroup 0116 protocol. There was only one minor variation of radiotherapy target volume coverage resulting from marginal coverage of the porta hepatis region. Adherence to the protocol was satisfactory, with 19 patients completing the radiotherapy protocol as planned and only one major variation in treatment delivery resulting from gastrointestinal toxicity. One major and one minor variation in normal tissue-dose constraints occurred on the heart and spinal cord, respectively. Compliance with treatment delivery was good, with only one patient failing to complete the prescribed radiotherapy dose owing to toxicity, although seven patients required treatment interruption. Conclusion: This audit showed good compliance with radiotherapy design and delivery. A formal medical quality assurance audit may provide a useful tool to assess complex new protocols introduced into routine departmental practice. Back M. F. et al. (2003). Clinical Oncology 15, 378–382 2003 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Radiotherapy, quality assurance, gastric cancer Received: 27 November 2002
Introduction
The Intergroup 0116 study showed that the addition of adjuvant chemo-radiotherapy after surgical resection of high-risk localised gastric cancer resulted in an improved relapse-free survival from 31% to 48% at 3 years [1]. Previous randomised studies have failed to show a significant survival benefit with intervention after surgery [2–4]. When these results were first presented at the American Society of Clinical Oncology in May 2000, the study protocol was quickly incorporated into many departments because of the poor outcome for patients
Address for correspondence: Dr Michael F. Back, National University Hospital, The Cancer Institute, Lower Kent Ridge Road, Singapore 119074, Singapore. Tel.: +65-6772-4869; Fax: +65-6779-6320; E-mail: [email protected] 0936-6555/03/70378+5 $30.00/0
Accepted: 1 May 2003
diagnosed with high-risk gastric cancer treated with surgery alone. One of the major features of the Intergroup 0116 Trial was the close attention to radiotherapy technique [5]. Not only was the complex target volume thoroughly outlined in the protocol, but all 281 patients in the experimental arm had a prospective quality assurance assessment conducted by the principal investigator before starting treatment [1,6]. Thirty-five per cent of treatment plans had major or minor deviations that required revision to comply satisfactorily with the described protocol, thus reflecting the potential complexity of the treatment. The concern of adopting this protocol into community oncology practice is that techniques may be incorrectly applied, with either inadequate coverage of target volume or excessive normal tissue exposure.
2003 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
379
CLINICAL ONCOLOGY
Table 1 – Quality assurance domains Eligibility criteria Gastric cancer TNM stage T3/T4 or node positive Dietary intake >1500 kcal per day
Radiation treatment prescription Radiation dose and fractionation: 45 Gy in 25 fractions Field size limitation <400 cm2 Dose specification point at isocentre with plan of distribution
In July 2000, the National University Hospital Singapore began to offer this protocol to patients with high-risk gastric cancer. This report outlines a quality assurance audit carried out on the first 20 patients managed on this protocol. Methods
We identified all patients managed with adjuvant radiation therapy for gastric malignancy from July 2000 to January 2002 from the LANTIS radiation oncology information system at the National University Hospital Singapore. We reviewed the medical records of the patients and isolated the first 20 consecutively managed patients under the Intergroup 0116 protocol. We retrospectively audited these patients’ medical records detailing patient, tumour and treatment-related factors. Quality Assurance Domains
Five quality assurance domains were determined from the Intergroup 0116 clinical trial protocol for formal assessment with the audit patients: eligibility criteria; radiation treatment prescription; radiation target volume coverage; quality assurance definition points used in audit; and normal tissue radiation dose constraints (Table 1). Radiation Target Volume Coverage
The target volume was designed to encompass the initial site of primary tumour and the regional lymph-node areas as defined by the Japanese Research Society for Gastric Cancer [7]. Quality assurance definition points were identified and audited for target volume coverage. The field arrangement recommended in the protocol and adopted in our institution was anterior-posterior opposed planar fields. The quality assurance definition points were obtained from the Intergroup 0116 protocol description of radiation field boundaries, and compared with individual patients’ simulation films. Variation was allowed for nodal coverage depending on the site of the primary tumour within the stomach to account for the different sites of potential relapse. This was specifically described in relation to no requirement for splenic hilar node coverage of antral tumours, and subpyloric node coverage of cardia tumours.
Quality assurance definition points Left lateral: medial two-thirds of left hemi-diaphragm Right lateral: >3 cm lateral to vertebral body Superior: dome of diaphragm/gastric fundus Inferior: L3/4 or L2 for selected cardiac lesions
Normal Tissue Radiation Dose Constraints
We identified and audited potential organs at risk for radiation therapy toxicity and respective dose constraints from the Intergroup 0116 protocol. The manner of calculation varied if patients were planned by computed tomography (CT) simulation or conventional X-ray simulation. When planned by CT, dose-volume histograms were used to calculate maximum dose received by organ or volume of organ receiving specified dose. Volume of organ spared referred to that volume receiving <10 Gy. Patients with X-ray simulation had two-dimensional contours taken with point-dose calculations. Organs within the field boundaries were presumed to receive target dose, and irradiated areas were calculated by direct measurement from simulation films. The calculated organ-dose constraints were as follows: renal (two-thirds of one kidney to be spared); spinal cord (maximum <45 Gy); liver (no more than 60% of liver volume to receive >30 Gy; cardiac (no more than 30% of cardiac area to receive >40 Gy). Compliance with Radiation Delivery
We calculated total radiotherapy dose delivered and treatment duration from the medical records. The Intergroup 0116 protocol permitted interruption to radiotherapy delivery in the event of toxicity. As an objective measure of treatment morbidity, this audit assessed the number of days interrupted as well as weight loss during radiotherapy. Quality Assurance Evaluation
The quality assurance guidelines from The TransTasman Radiation Oncology Group were modified to define major and minor variations to the Intergroup 0116 protocol [8] (Table 2).
Results
Twenty patients were managed under the Intergroup 0116 protocol from July 2000 to September 2001. Most were men (n=15), and the median age was 62 years (range 35–81 years). The tumours were located in the cardia (2), body (10) and antrum (8) of the stomach. Eight patients had extended nodal dissection as defined
380
INTERGROUP 0116 PROTOCOL
Table 2 – Quality assurance evaluation definition modified from Trans-Tasman Radiation Oncology Group criteria [8] Prescription Per protocol Minor variation Major variation
Within 5% 5–10% >10%
RT field placement
Normal tissue dose
RT delivery
Within 5% 5–10% >10%
Within 10% 11–20% >20%
Target volume covered with margins as defined by protocol (10 mm) Target covered by minimal (<10 mm) margin No margin around target
RT, radiotherapy.
Table 3 – Variations from protocol for each of the QA domains of audited patients
Table 4 – Renal volumes irradiated in audited patients Total number of patients (n=20)
Per protocol
Minor variation
Major variation
Eligibility Prescription RT target volume
20 20 19
— — 1
Normal tissue dose Renal Spinal cord Cardiac Liver
20 19 19 20
RT delivery Total dose RT duration
19 19
Two-thirds of one kidney spared
20
— — —
Combined kidney volume irradiated <20% 20–50% >50%
3 12 5
— 1 — —
— — 1 —
RT to > two-thirds of one kidney
— —
1 1
borders in 15 patients. There was one minor variation in coverage of the porta hepatis region, with a right lateral field border sited less than 3 cm margin from vertebral body.
by D2 dissection [7]. The remaining 12 patients had D1 dissections limited to removal of greater and lesser curve nodal groups. The median number of nodes described by the pathologist was 18 (range 10–39). Five radiation oncologists were subsequently involved in the treatment of these 20 patients. The audit showed good compliance with identified quality assurance domains taken from the Intergroup 0116 protocol (Table 3). Eligibility Criteria
All patients satisfied the clinical stage and performance status criteria. Eighteen patients were node positive, whereas two patients had T3 N0 disease. However, the dietary assessments were not specifically quantified in the medical records to determine if this was adequately followed. Radiation Treatment Prescription
All patients were managed with anterior-posterior opposed fields, with no use of lateral fields. Fields were weighted anteriorly to reduce the spinal cord dose. Prescription dose, field-size limitation and dose specification were adhered to. Radiation Target Volume Coverage
Radiation target volumes were determined by CT simulation in five patients and by X-ray simulation field
8
Normal Tissue Radiation Dose Constraints
One major variation occurred in a patient with a cardia lesion, in which the superior border of the field extended high into the thorax to cover the anastomosis. This resulted in a larger volume of the heart receiving the permitted 40 Gy. One minor variation occurred with a calculated spinal cord dose of 46 Gy exceeding the permitted 45 Gy. The renal volumes irradiated were within protocol, with all patients having at least two-thirds of one kidney spared from the radiotherapy field (or <10 Gy). The audit data demonstrated the large volumes of renal tissue that are routinely irradiated as part of an anteriorposterior gastric cancer field arrangement. Five patients had more than 50% of combined renal volume within the radiation portals (Table 4). Compliance with Radiation Delivery
The total radiotherapy dose was delivered as planned in 19 patients, with one major variation related to patientinitiated cessation of radiotherapy at 39.6 Gy of planned 45 Gy resulting from gastrointestinal toxicity. The radiotherapy was delivered in the planned 5 weeks in only 12 patients. Interruptions to radiotherapy occurred in seven patients at a median of 2 days. One major variation in treatment duration resulted from an 18-day treatment break due to gastrointestinal toxicity. This patient also experienced >10% of weight loss, but was
381
CLINICAL ONCOLOGY
Fig. 1 – Typical portal for gastric antrum tumour, with quality assurance definition points used in audit. (A) Medial twothirds of left hemi-diaphragm; (B) >3 cm lateral to vertebral body; (C) dome of diaphragm/gastric fundus; and (D) L3/4 or L2 for selected cardiac lesions.
the only individual to lose >5 kg or >10% body weight during radiotherapy. No patient required admission during the radiation therapy. Three patients reported Common Toxicity Criteria grade 3 or 4 acute radiotherapy toxicity. One patient had grade 3 vomiting (>6 episodes in 24 h); one had grade 3 diarrhoea (increase of >7 stools per day) and one had grade 4 anorexia (required nasogastric feeding). This was the patient that had the prolonged radiotherapy interruption. Only one patient failed to receive all three initial cycles of planned 5-fluorouracil. However, compliance with completion of the final two cycles was less, with only 11 out of 17 patients receiving all five planned cycles of chemotherapy. Data were incomplete for the final two cycles for three of the patients. Discussion
The Intergroup 0116 Trial showed a significant survival advantage for patients managed with adjuvant chemoradiotherapy for high-risk gastric cancer [1]. This multicentre study was well conducted by a large clinical trials group, with individual centres receiving group accreditation before patient enrolment. Furthermore, the study had a compulsory prospective quality assurance assessment before radiotherapy, as well as a quantitative measurement of performance status in the form of
calculated caloric intake [5]. The clinical trial protocol extensively detailed the radiation target volume to be managed and normal tissue-dose constraints. The extent to which these quality assurance features resulted in the positive study outcome will remain uncertain; however it is important to recognise the potential for geographical miss of target volume or enhanced toxicity when this study protocol is introduced to community practice. Promising data from wellconducted, single institution phase II studies have often floundered when tested in multicentre trials [9,10], possibly because of similar lowering in treatment delivery quality control or selection bias. Likewise validation of a positive phase III protocol may also be unsuccessful or impose greater morbidity if the criteria for selection and treatment technique are altered [11]. More commonly validated studies may not be performed, as is likely to be the scenario for the Intergroup 0116 study, and the results will remain unchallenged at a community oncology level. When a successful complex phase III protocol is introduced into clinical practice, it becomes important to consider the details of the protocol and the techniques adopted. Reproducibility of radiation technique needs to be carefully reviewed. Issues to consider are total radiotherapy dose, dose specification, selection criteria and normal tissue constraints. The electronic publication of complete original trial protocols may be one method of improving subsequent quality treatment delivery rather than relying on the publication of the final trial results. Formal quality assurance audit of technique, as conducted in this report, is useful if there is a time delay before clinical results are published. Although quality assurance processes are an integral part of radiotherapy delivery in radiation oncology departments, minimal data have been published on quality assurance of radiotherapy techniques in community practice. This quality assurance audit presents a structured method that follows the protocol from which the evidence to implement the treatment originated. In forthcoming years, with the passage of sophisticated intensity-modulated radiotherapy techniques from clinical trials into community practice, this reproducibility of technique will become even more important. A concern with the Intergroup 0116 study protocol is the potential for late normal tissue morbidity, especially with the large renal volumes irradiated. Although the dose constraints are well described in the protocol, the background for these restraints are based upon historical series using radiation alone, in which less than 20% of the population were long-term survivors [12]. Also the clinical presumption is that normal creatinine level reflects normal functioning of both kidneys, whereas it may be that one organ is maintaining overall renal function. The inclusion of this kidney in the radiotherapy portal, with sparing of an impaired kidney, may lead to subsequent renal dysfunction. A correlate of renal function may be obtained at time of contrast CT by reviewing extent of contrast enhancement of each
INTERGROUP 0116 PROTOCOL
kidney. A renal perfusion study before treatment would more assess precisely the contribution of each kidney to renal function [13]. We have now introduced this formally for all patients under a clinical trial to assess the effect of radiotherapy on renal perfusion from baseline to 12 months after radiotherapy. It is unlikely that the Intergroup 0116 study will be repeated to confirm the survival advantage of adjuvant chemo-radiotherapy. Believers of more aggressive surgical intervention may debate whether an extensive lymph-node dissection can replace the adjuvant regional radiation therapy [14,15], especially as only 10% had D2 dissections in the Intergroup 0116 study. However, radiation therapy remains a standard in both arms for the next Radiation Therapy Oncology Group study, a randomised phase II study that is investigating potentially superior systemic agents, with the inclusion of cisplatin and paclitaxel to the systemic regimen [16]. For radiation oncologists, the Intergroup 0116 protocol field arrangements are remnants of an older radiotherapy philosophy, with field borders being recommended to cover potential sites of relapse. Threedimensional CT planning of gastric cancer, with specific contouring and targeting of nodal regions at risk, is more consistent with modern techniques. This offers potential benefits in reducing the dose to critical normal tissues; however, radiation oncologists will need to improve their understanding of the radiological anatomy, and incorporate advice from surgeons and radiologists in the planning process. Abandoning anterior–posterior fields for conformal radiotherapy may result in reduced renal irradiation, but potentially at the risk of a geographical target miss or increased treated liver volume. Whether this compromises locoregional control is at present unknown, and in the absence of a further randomised study, data from phase II studies will need to be compared with the experimental arm of Intergroup 0116. It is also important to reflect the lessons of surgical history, which, with increasing complexity of intervention, has reflected poorer results from clinicians who are either inexperienced at a technique or whose caseload is low compared with subspecialised surgeons [17]. Therefore, taking all these factors into account, the introduction of a new technique to a radiation oncology department should be undertaken with an appropriately audited process assessing the quality assurance of radiotherapy design and delivery. Conclusion
The Intergroup 0116 study will significantly influence radiation oncology practice. Careful attention to quality assurance of radiotherapy design and technique was an important feature of this study protocol. This audit
382
demonstrated satisfactory compliance with the protocol and revealed aspects of treatment design for further investigation. The use of a formal technical audit is an important method for radiation oncology departments to consider when incorporating a new clinical protocol into routine practice. References 1 Macdonald J, Smalley S, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach and gastroesophageal junction. N Engl J Med 2001; 345:725–730. 2 Hermans J, Bonenkamp JJ, Boon MC, et al. Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials. J Clin Oncol 1993;11:1441–1447. 3 Coombes RC, Schein PS, Chilvers CE, et al. A randomized trial comparing adjuvant fluorouracil, doxorubicin, and mitomycin with no treatment in operable gastric cancer. International Collaborative Cancer Group. J Clin Oncol 1990;8:1362–1369. 4 Songun I, Keizer HJ, Hermans J, et al. Chemotherapy for operable gastric cancer: results of the Dutch randomised FAMTX trial. The Dutch Gastric Cancer Group (DGCG). Eur J Cancer 1999; 35:558–562. 5 Macdonald J, Smalley S, Brown T, et al. Phase III Intergroup Trial of Adjuvant Chemoradiation after gastric resection for adenocarcinoma. NCI High Priority Study INT 0116. Edition April 20, 1998. 6 Smalley SR, Gunderson L, Tepper J, et al. Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int J Radiat Oncol Biol Phys 2002;52:283–293. 7 Kajitani T. Japanese Research Society for Gastric Cancer: the general rules for the gastric cancer study in surgery and pathology. Part I clinical classification. Jpn J Surg 1981;11:127–139. 8 TROG quality assurance statement of minimum requirements for clinical trials. Trans-Tasman Radiation Oncology Group handbook. Newcastle, Australia; 1999:9–10. 9 Back M, Delaney G, Denham J, et al. How should we introduce high-dose chemotherapeutic strategies into the adjuvant management of high-risk breast cancer in Australasia? Aust N Z J Surg 1998;68:10–15. 10 Dicato M. High-dose chemotherapy in breast cancer: where are we now? Semin Oncol 2002;29(suppl 8):16–20. 11 Piccart MJ, Lohrisch C, Duchateau L, et al. Taxanes in the adjuvant treatment of breast cancer: why not yet? J Natl Cancer Inst Monogr 2001;30:88–95. 12 Regine WF, Mohiuddin M. Impact of adjuvant therapy on locally advanced adenocarcinoma of the stomach. Int J Radiat Oncol Biol Phys 1992;24:921. 13 Degirmenci B, Uysal K, Bekis R, et al. Tc-99m MDP, thallium-201 chloride and Tc-99m MAG3 renal uptake in subacute and chronic radiation nephritis compared. Ann Nucl Med 2001;15:447–449. 14 Cuschieri A, Weeden S, Fielding J, et al. Patient survival after D1 and D2 resections for gastric cancer: long-term results of the MRC randomized surgical trial. Br J Cancer 1999;79:1422. 15 Sakamoto J, Yasue M. Extensive lymphadenectomy for gastric cancer patients: what can the results of one trial tell us? Lancet 1995;345:742–743. 16 Radiation Therapy Oncology Group Philadelphia. Study Protocol radiotherapyOG G-0114. A randomized phase II comparison of two cisplatin-paclitaxel containing chemoradiation regimens in resected gastric cancers. 17 Hermanek P, Hermanek PJ. Role of the surgeon as a variable in the treatment of rectal cancer. Semin Surg Oncol 2000;19:329–335.
Original Article Introducing the Intergroup 0116 Protocol of Adjuvant Chemo-radiotherapy in Gastric Cancer into Clinical Practice M. F. Back, S. Premsenthil, C. J. Wynne, T. P. Shakespeare Department of Radiation Oncology, The Cancer Institute, National University Hospital, Singapore ABSTRACT: Aims: The results of techniques from a well-conducted clinical trial are often difficult to reproduce when implemented in community oncology practice. The U.S. Intergroup 0116 protocol of adjuvant chemo-radiotherapy in gastric cancer presented in mid-2000 produced a survival advantage over surgery alone. The current study aims to determine the adherence with protocol design and delivery of radiation therapy (radiotherapy) in the initial 20 patients managed with the Intergroup 0116 protocol at The National University Hospital, Singapore. Materials and methods: A formal quality assurance audit was performed on clinical features, radiotherapy treatment charts and simulation films of the first 20 patients treated with the Intergroup 0116 protocol from July 2000 to September 2001. Specific details were audited for their consistency with described protocol in domains of eligibility criteria, radiotherapy prescription, target volume coverage and adherence to dose-limiting normal tissue tolerances. Compliance and toxicity with the protocol was assessed by audit of delivered radiotherapy dose, treatment interruptions, inpatient admissions and weight loss during radiotherapy. Results: The 20 audited patients were appropriately selected on the basis of eligibility criteria of Intergroup 0116 protocol. There was only one minor variation of radiotherapy target volume coverage resulting from marginal coverage of the porta hepatis region. Adherence to the protocol was satisfactory, with 19 patients completing the radiotherapy protocol as planned and only one major variation in treatment delivery resulting from gastrointestinal toxicity. One major and one minor variation in normal tissue-dose constraints occurred on the heart and spinal cord, respectively. Compliance with treatment delivery was good, with only one patient failing to complete the prescribed radiotherapy dose owing to toxicity, although seven patients required treatment interruption. Conclusion: This audit showed good compliance with radiotherapy design and delivery. A formal medical quality assurance audit may provide a useful tool to assess complex new protocols introduced into routine departmental practice. Back M. F. et al. (2003). Clinical Oncology 15, 378–382 2003 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Radiotherapy, quality assurance, gastric cancer Received: 27 November 2002
Introduction
The Intergroup 0116 study showed that the addition of adjuvant chemo-radiotherapy after surgical resection of high-risk localised gastric cancer resulted in an improved relapse-free survival from 31% to 48% at 3 years [1]. Previous randomised studies have failed to show a significant survival benefit with intervention after surgery [2–4]. When these results were first presented at the American Society of Clinical Oncology in May 2000, the study protocol was quickly incorporated into many departments because of the poor outcome for patients
Address for correspondence: Dr Michael F. Back, National University Hospital, The Cancer Institute, Lower Kent Ridge Road, Singapore 119074, Singapore. Tel.: +65-6772-4869; Fax: +65-6779-6320; E-mail: [email protected] 0936-6555/03/70378+5 $30.00/0
Accepted: 1 May 2003
diagnosed with high-risk gastric cancer treated with surgery alone. One of the major features of the Intergroup 0116 Trial was the close attention to radiotherapy technique [5]. Not only was the complex target volume thoroughly outlined in the protocol, but all 281 patients in the experimental arm had a prospective quality assurance assessment conducted by the principal investigator before starting treatment [1,6]. Thirty-five per cent of treatment plans had major or minor deviations that required revision to comply satisfactorily with the described protocol, thus reflecting the potential complexity of the treatment. The concern of adopting this protocol into community oncology practice is that techniques may be incorrectly applied, with either inadequate coverage of target volume or excessive normal tissue exposure.
2003 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
379
CLINICAL ONCOLOGY
Table 1 – Quality assurance domains Eligibility criteria Gastric cancer TNM stage T3/T4 or node positive Dietary intake >1500 kcal per day
Radiation treatment prescription Radiation dose and fractionation: 45 Gy in 25 fractions Field size limitation <400 cm2 Dose specification point at isocentre with plan of distribution
In July 2000, the National University Hospital Singapore began to offer this protocol to patients with high-risk gastric cancer. This report outlines a quality assurance audit carried out on the first 20 patients managed on this protocol. Methods
We identified all patients managed with adjuvant radiation therapy for gastric malignancy from July 2000 to January 2002 from the LANTIS radiation oncology information system at the National University Hospital Singapore. We reviewed the medical records of the patients and isolated the first 20 consecutively managed patients under the Intergroup 0116 protocol. We retrospectively audited these patients’ medical records detailing patient, tumour and treatment-related factors. Quality Assurance Domains
Five quality assurance domains were determined from the Intergroup 0116 clinical trial protocol for formal assessment with the audit patients: eligibility criteria; radiation treatment prescription; radiation target volume coverage; quality assurance definition points used in audit; and normal tissue radiation dose constraints (Table 1). Radiation Target Volume Coverage
The target volume was designed to encompass the initial site of primary tumour and the regional lymph-node areas as defined by the Japanese Research Society for Gastric Cancer [7]. Quality assurance definition points were identified and audited for target volume coverage. The field arrangement recommended in the protocol and adopted in our institution was anterior-posterior opposed planar fields. The quality assurance definition points were obtained from the Intergroup 0116 protocol description of radiation field boundaries, and compared with individual patients’ simulation films. Variation was allowed for nodal coverage depending on the site of the primary tumour within the stomach to account for the different sites of potential relapse. This was specifically described in relation to no requirement for splenic hilar node coverage of antral tumours, and subpyloric node coverage of cardia tumours.
Quality assurance definition points Left lateral: medial two-thirds of left hemi-diaphragm Right lateral: >3 cm lateral to vertebral body Superior: dome of diaphragm/gastric fundus Inferior: L3/4 or L2 for selected cardiac lesions
Normal Tissue Radiation Dose Constraints
We identified and audited potential organs at risk for radiation therapy toxicity and respective dose constraints from the Intergroup 0116 protocol. The manner of calculation varied if patients were planned by computed tomography (CT) simulation or conventional X-ray simulation. When planned by CT, dose-volume histograms were used to calculate maximum dose received by organ or volume of organ receiving specified dose. Volume of organ spared referred to that volume receiving <10 Gy. Patients with X-ray simulation had two-dimensional contours taken with point-dose calculations. Organs within the field boundaries were presumed to receive target dose, and irradiated areas were calculated by direct measurement from simulation films. The calculated organ-dose constraints were as follows: renal (two-thirds of one kidney to be spared); spinal cord (maximum <45 Gy); liver (no more than 60% of liver volume to receive >30 Gy; cardiac (no more than 30% of cardiac area to receive >40 Gy). Compliance with Radiation Delivery
We calculated total radiotherapy dose delivered and treatment duration from the medical records. The Intergroup 0116 protocol permitted interruption to radiotherapy delivery in the event of toxicity. As an objective measure of treatment morbidity, this audit assessed the number of days interrupted as well as weight loss during radiotherapy. Quality Assurance Evaluation
The quality assurance guidelines from The TransTasman Radiation Oncology Group were modified to define major and minor variations to the Intergroup 0116 protocol [8] (Table 2).
Results
Twenty patients were managed under the Intergroup 0116 protocol from July 2000 to September 2001. Most were men (n=15), and the median age was 62 years (range 35–81 years). The tumours were located in the cardia (2), body (10) and antrum (8) of the stomach. Eight patients had extended nodal dissection as defined
380
INTERGROUP 0116 PROTOCOL
Table 2 – Quality assurance evaluation definition modified from Trans-Tasman Radiation Oncology Group criteria [8] Prescription Per protocol Minor variation Major variation
Within 5% 5–10% >10%
RT field placement
Normal tissue dose
RT delivery
Within 5% 5–10% >10%
Within 10% 11–20% >20%
Target volume covered with margins as defined by protocol (10 mm) Target covered by minimal (<10 mm) margin No margin around target
RT, radiotherapy.
Table 3 – Variations from protocol for each of the QA domains of audited patients
Table 4 – Renal volumes irradiated in audited patients Total number of patients (n=20)
Per protocol
Minor variation
Major variation
Eligibility Prescription RT target volume
20 20 19
— — 1
Normal tissue dose Renal Spinal cord Cardiac Liver
20 19 19 20
RT delivery Total dose RT duration
19 19
Two-thirds of one kidney spared
20
— — —
Combined kidney volume irradiated <20% 20–50% >50%
3 12 5
— 1 — —
— — 1 —
RT to > two-thirds of one kidney
— —
1 1
borders in 15 patients. There was one minor variation in coverage of the porta hepatis region, with a right lateral field border sited less than 3 cm margin from vertebral body.
by D2 dissection [7]. The remaining 12 patients had D1 dissections limited to removal of greater and lesser curve nodal groups. The median number of nodes described by the pathologist was 18 (range 10–39). Five radiation oncologists were subsequently involved in the treatment of these 20 patients. The audit showed good compliance with identified quality assurance domains taken from the Intergroup 0116 protocol (Table 3). Eligibility Criteria
All patients satisfied the clinical stage and performance status criteria. Eighteen patients were node positive, whereas two patients had T3 N0 disease. However, the dietary assessments were not specifically quantified in the medical records to determine if this was adequately followed. Radiation Treatment Prescription
All patients were managed with anterior-posterior opposed fields, with no use of lateral fields. Fields were weighted anteriorly to reduce the spinal cord dose. Prescription dose, field-size limitation and dose specification were adhered to. Radiation Target Volume Coverage
Radiation target volumes were determined by CT simulation in five patients and by X-ray simulation field
8
Normal Tissue Radiation Dose Constraints
One major variation occurred in a patient with a cardia lesion, in which the superior border of the field extended high into the thorax to cover the anastomosis. This resulted in a larger volume of the heart receiving the permitted 40 Gy. One minor variation occurred with a calculated spinal cord dose of 46 Gy exceeding the permitted 45 Gy. The renal volumes irradiated were within protocol, with all patients having at least two-thirds of one kidney spared from the radiotherapy field (or <10 Gy). The audit data demonstrated the large volumes of renal tissue that are routinely irradiated as part of an anteriorposterior gastric cancer field arrangement. Five patients had more than 50% of combined renal volume within the radiation portals (Table 4). Compliance with Radiation Delivery
The total radiotherapy dose was delivered as planned in 19 patients, with one major variation related to patientinitiated cessation of radiotherapy at 39.6 Gy of planned 45 Gy resulting from gastrointestinal toxicity. The radiotherapy was delivered in the planned 5 weeks in only 12 patients. Interruptions to radiotherapy occurred in seven patients at a median of 2 days. One major variation in treatment duration resulted from an 18-day treatment break due to gastrointestinal toxicity. This patient also experienced >10% of weight loss, but was
381
CLINICAL ONCOLOGY
Fig. 1 – Typical portal for gastric antrum tumour, with quality assurance definition points used in audit. (A) Medial twothirds of left hemi-diaphragm; (B) >3 cm lateral to vertebral body; (C) dome of diaphragm/gastric fundus; and (D) L3/4 or L2 for selected cardiac lesions.
the only individual to lose >5 kg or >10% body weight during radiotherapy. No patient required admission during the radiation therapy. Three patients reported Common Toxicity Criteria grade 3 or 4 acute radiotherapy toxicity. One patient had grade 3 vomiting (>6 episodes in 24 h); one had grade 3 diarrhoea (increase of >7 stools per day) and one had grade 4 anorexia (required nasogastric feeding). This was the patient that had the prolonged radiotherapy interruption. Only one patient failed to receive all three initial cycles of planned 5-fluorouracil. However, compliance with completion of the final two cycles was less, with only 11 out of 17 patients receiving all five planned cycles of chemotherapy. Data were incomplete for the final two cycles for three of the patients. Discussion
The Intergroup 0116 Trial showed a significant survival advantage for patients managed with adjuvant chemoradiotherapy for high-risk gastric cancer [1]. This multicentre study was well conducted by a large clinical trials group, with individual centres receiving group accreditation before patient enrolment. Furthermore, the study had a compulsory prospective quality assurance assessment before radiotherapy, as well as a quantitative measurement of performance status in the form of
calculated caloric intake [5]. The clinical trial protocol extensively detailed the radiation target volume to be managed and normal tissue-dose constraints. The extent to which these quality assurance features resulted in the positive study outcome will remain uncertain; however it is important to recognise the potential for geographical miss of target volume or enhanced toxicity when this study protocol is introduced to community practice. Promising data from wellconducted, single institution phase II studies have often floundered when tested in multicentre trials [9,10], possibly because of similar lowering in treatment delivery quality control or selection bias. Likewise validation of a positive phase III protocol may also be unsuccessful or impose greater morbidity if the criteria for selection and treatment technique are altered [11]. More commonly validated studies may not be performed, as is likely to be the scenario for the Intergroup 0116 study, and the results will remain unchallenged at a community oncology level. When a successful complex phase III protocol is introduced into clinical practice, it becomes important to consider the details of the protocol and the techniques adopted. Reproducibility of radiation technique needs to be carefully reviewed. Issues to consider are total radiotherapy dose, dose specification, selection criteria and normal tissue constraints. The electronic publication of complete original trial protocols may be one method of improving subsequent quality treatment delivery rather than relying on the publication of the final trial results. Formal quality assurance audit of technique, as conducted in this report, is useful if there is a time delay before clinical results are published. Although quality assurance processes are an integral part of radiotherapy delivery in radiation oncology departments, minimal data have been published on quality assurance of radiotherapy techniques in community practice. This quality assurance audit presents a structured method that follows the protocol from which the evidence to implement the treatment originated. In forthcoming years, with the passage of sophisticated intensity-modulated radiotherapy techniques from clinical trials into community practice, this reproducibility of technique will become even more important. A concern with the Intergroup 0116 study protocol is the potential for late normal tissue morbidity, especially with the large renal volumes irradiated. Although the dose constraints are well described in the protocol, the background for these restraints are based upon historical series using radiation alone, in which less than 20% of the population were long-term survivors [12]. Also the clinical presumption is that normal creatinine level reflects normal functioning of both kidneys, whereas it may be that one organ is maintaining overall renal function. The inclusion of this kidney in the radiotherapy portal, with sparing of an impaired kidney, may lead to subsequent renal dysfunction. A correlate of renal function may be obtained at time of contrast CT by reviewing extent of contrast enhancement of each
INTERGROUP 0116 PROTOCOL
kidney. A renal perfusion study before treatment would more assess precisely the contribution of each kidney to renal function [13]. We have now introduced this formally for all patients under a clinical trial to assess the effect of radiotherapy on renal perfusion from baseline to 12 months after radiotherapy. It is unlikely that the Intergroup 0116 study will be repeated to confirm the survival advantage of adjuvant chemo-radiotherapy. Believers of more aggressive surgical intervention may debate whether an extensive lymph-node dissection can replace the adjuvant regional radiation therapy [14,15], especially as only 10% had D2 dissections in the Intergroup 0116 study. However, radiation therapy remains a standard in both arms for the next Radiation Therapy Oncology Group study, a randomised phase II study that is investigating potentially superior systemic agents, with the inclusion of cisplatin and paclitaxel to the systemic regimen [16]. For radiation oncologists, the Intergroup 0116 protocol field arrangements are remnants of an older radiotherapy philosophy, with field borders being recommended to cover potential sites of relapse. Threedimensional CT planning of gastric cancer, with specific contouring and targeting of nodal regions at risk, is more consistent with modern techniques. This offers potential benefits in reducing the dose to critical normal tissues; however, radiation oncologists will need to improve their understanding of the radiological anatomy, and incorporate advice from surgeons and radiologists in the planning process. Abandoning anterior–posterior fields for conformal radiotherapy may result in reduced renal irradiation, but potentially at the risk of a geographical target miss or increased treated liver volume. Whether this compromises locoregional control is at present unknown, and in the absence of a further randomised study, data from phase II studies will need to be compared with the experimental arm of Intergroup 0116. It is also important to reflect the lessons of surgical history, which, with increasing complexity of intervention, has reflected poorer results from clinicians who are either inexperienced at a technique or whose caseload is low compared with subspecialised surgeons [17]. Therefore, taking all these factors into account, the introduction of a new technique to a radiation oncology department should be undertaken with an appropriately audited process assessing the quality assurance of radiotherapy design and delivery. Conclusion
The Intergroup 0116 study will significantly influence radiation oncology practice. Careful attention to quality assurance of radiotherapy design and technique was an important feature of this study protocol. This audit
382
demonstrated satisfactory compliance with the protocol and revealed aspects of treatment design for further investigation. The use of a formal technical audit is an important method for radiation oncology departments to consider when incorporating a new clinical protocol into routine practice. References 1 Macdonald J, Smalley S, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach and gastroesophageal junction. N Engl J Med 2001; 345:725–730. 2 Hermans J, Bonenkamp JJ, Boon MC, et al. Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials. J Clin Oncol 1993;11:1441–1447. 3 Coombes RC, Schein PS, Chilvers CE, et al. A randomized trial comparing adjuvant fluorouracil, doxorubicin, and mitomycin with no treatment in operable gastric cancer. International Collaborative Cancer Group. J Clin Oncol 1990;8:1362–1369. 4 Songun I, Keizer HJ, Hermans J, et al. Chemotherapy for operable gastric cancer: results of the Dutch randomised FAMTX trial. The Dutch Gastric Cancer Group (DGCG). Eur J Cancer 1999; 35:558–562. 5 Macdonald J, Smalley S, Brown T, et al. Phase III Intergroup Trial of Adjuvant Chemoradiation after gastric resection for adenocarcinoma. NCI High Priority Study INT 0116. Edition April 20, 1998. 6 Smalley SR, Gunderson L, Tepper J, et al. Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int J Radiat Oncol Biol Phys 2002;52:283–293. 7 Kajitani T. Japanese Research Society for Gastric Cancer: the general rules for the gastric cancer study in surgery and pathology. Part I clinical classification. Jpn J Surg 1981;11:127–139. 8 TROG quality assurance statement of minimum requirements for clinical trials. Trans-Tasman Radiation Oncology Group handbook. Newcastle, Australia; 1999:9–10. 9 Back M, Delaney G, Denham J, et al. How should we introduce high-dose chemotherapeutic strategies into the adjuvant management of high-risk breast cancer in Australasia? Aust N Z J Surg 1998;68:10–15. 10 Dicato M. High-dose chemotherapy in breast cancer: where are we now? Semin Oncol 2002;29(suppl 8):16–20. 11 Piccart MJ, Lohrisch C, Duchateau L, et al. Taxanes in the adjuvant treatment of breast cancer: why not yet? J Natl Cancer Inst Monogr 2001;30:88–95. 12 Regine WF, Mohiuddin M. Impact of adjuvant therapy on locally advanced adenocarcinoma of the stomach. Int J Radiat Oncol Biol Phys 1992;24:921. 13 Degirmenci B, Uysal K, Bekis R, et al. Tc-99m MDP, thallium-201 chloride and Tc-99m MAG3 renal uptake in subacute and chronic radiation nephritis compared. Ann Nucl Med 2001;15:447–449. 14 Cuschieri A, Weeden S, Fielding J, et al. Patient survival after D1 and D2 resections for gastric cancer: long-term results of the MRC randomized surgical trial. Br J Cancer 1999;79:1422. 15 Sakamoto J, Yasue M. Extensive lymphadenectomy for gastric cancer patients: what can the results of one trial tell us? Lancet 1995;345:742–743. 16 Radiation Therapy Oncology Group Philadelphia. Study Protocol radiotherapyOG G-0114. A randomized phase II comparison of two cisplatin-paclitaxel containing chemoradiation regimens in resected gastric cancers. 17 Hermanek P, Hermanek PJ. Role of the surgeon as a variable in the treatment of rectal cancer. Semin Surg Oncol 2000;19:329–335.