- Email: [email protected]
The Impact of Gynaecological Tumour Board Rounds on Patient Care
GYNAECOLOGY
Impact of Gynaecological Tumour Board Rounds on Patient Care Darron A. Halliday, MD;1 Khadija Warfa, MD;2 Jill Nation, MD;3 Pamela Chu, MD;3 Gregg Nelson, MD, PhD;3 Sarah Glaze, MD;3 Jennifer Mateshaytis, MD;4 Máire A. Duggan, MD;5 Prafull Ghatage, MD3 1
Department of Obstetrics & Gynecology, Princess Margaret Hospital, Nassau, Bahamas
2
Department of Obstetrics & Gynecology, Aga Khan University Hospital, Nairobi, Kenya
D.A. Halliday
3
Department of Oncology, University of Calgary, Calgary, Canada
4
Department of Obstetrics & Gynecology, University of Alberta, Edmonton, Canada
5
Department of Pathology and Laboratory Medicine, University of Calgary, Calgary, Canada
Abstract
Résumé
Objective: The goal of this study was to determine the impact of tumour board rounds (TBRs) on the additional management of patients with gynaecologic malignancy.
Objectif : Cette étude visait à déterminer l’incidence des rencontres du comité de thérapie du cancer sur la prise en charge ultérieure des patientes atteintes de cancers gynécologiques.
Methods: A retrospective chart review of 1604 patients discussed between January 2011 and December 2013 at gynaecologic TBRs was conducted to determine the frequency and type of diagnostic discrepancies found post-TBRs and their potential impact on additional patient management. A discrepancy was defined as major if it affected patient management by cancelling, initiating, or modifying treatment; otherwise, the discrepancy was minor. Data collected included patients’ demographics, pre- and post-TBR diagnoses, and management.
Méthodologie : Nous avons mené une évaluation rétrospective des dossiers de 1 604 patientes ayant fait l’objet de rencontres du comité de thérapie du cancer gynécologique entre janvier 2011 et décembre 2013 pour déterminer la fréquence et le type des écarts diagnostiques relevés après les rencontres et leur incidence potentielle sur la prise en charge ultérieure des patientes. Un écart était jugé majeur s’il modifiait la prise en charge par l’annulation, l’initiation ou la modification d’un traitement; sinon, l’écart était jugé mineur. Les données recueillies comprenaient les données démographiques des patientes, le diagnostic avant et après la rencontre, et la prise en charge.
Results: The patients’ mean age was 57.6 ± 14.1. Endometrial disease accounted for (43%) of the TBRs. The remaining sites were ovarian (25%), cervical (23%), and others (9%). Overall, 13.2% (n = 212) had a discrepancy; 3.4% (n = 54) of these discrepancies were major, and 9.9% (n = 158) were minor. Most major discrepancies related to changes in the tumours’ primary site or stage, and most minor discrepancies were related to changes in tumour histotype. Among the 54 (25.5%) major discrepancies, 18 (33.3%) occurred in patients who had their additional management cancelled, 17 (31.5%) required chemotherapy, 4 (7.4%) required a change in the chemotherapy regimen, 10 (18.5%) required additional surgery, and 5 (9.3%) required chemoradiation. Conclusion: The 13% frequency of discrepancies, approximately 26% of which were major and resulted in changes in patient management, highlights the importance of TBRs as a quality tool.
Key words: Tumour board rounds, gynaecological cancer, patient management Corresponding Author: Dr. Darron Halliday, Department of Obstetrics & Gynecology, Princess Margaret Hospital, Nassau, Bahamas. [email protected] Competing interests: The authors declare that they have no competing interests. Received on March 17, 2018 Accepted on May 15, 2018
Résultats : L’âge moyen des patientes était de 57,6 ± 14,1 ans. Les cancers de l’endomètre étaient les plus fréquents, avec 43 % des dossiers. Les autres cancers touchaient les ovaires (25 %), le col de l’utérus (23 %) et d’autres organes (9 %). Dans l’ensemble, 13,2 % (n = 212) des cas présentaient un écart; dans 3,4 % (n = 54) des cas, l’écart était majeur, et il était mineur dans 9,9 % (n = 158) des cas. La plupart des écarts majeurs étaient liés à des changements du site primaire ou du stade de la tumeur, et la plupart des écarts mineurs étaient liés à des changements de l’histotype de la tumeur. Des 54 (25,5 %) écarts majeurs, 18 (33,3 %) ont entraîné l’annulation de la prise en charge ultérieure, 17 (31,5 %) ont entraîné une chimiothérapie, 4 (7,4%) ont nécessité un changement du régime de chimiothérapie, 10 (18,5 %) ont nécessité une chirurgie supplémentaire et 5 (9,3 %) ont nécessité une chimioradiothérapie. Conclusions : La fréquence des écarts (13 %), dont environ 26 % étaient majeurs et ont entraîné des changements dans la prise en charge des patientes, met en lumière l’importance des rencontres du comité de thérapie du cancer comme outil d’assurance de la qualité. Copyright © 2018 The Society of Obstetricians and Gynaecologists of Canada/La Société des obstétriciens et gynécologues du Canada. Published by Elsevier Inc. All rights reserved.
J Obstet Gynaecol Can ■■;■■(■■):■■–■■ https://doi.org/10.1016/j.jogc.2018.05.026
■■ JOGC ■■ ■■ • 1
GYNAECOLOGY
INTRODUCTION
T
umour board rounds (TBRs) are an integral part of oncology practice. Sometimes revisions of the pre-TBR diagnosis lead to changes in additional management, and different perspectives on management occur when multidisciplinary teams come together. The gynaecologic TBRs at the Tom Baker Cancer Centre (TBCC) in Calgary, Alberta, consist of gynaecologic and radiation oncologists, gynaecologic pathologists, and psychologists. This group meets weekly to discuss cases managed by the gynaecologic oncology service or referred for additional oncologic management by other physicians in southern Alberta. A typical TBR discussion involves an oral presentation summarizing the patient’s history, physical examination findings, imaging, laboratory investigations, surgical procedures, and pathology reports. The gynaecologic pathologist reviews the relevant and pertinent pathology of each case before the TBR and presents the results. Thereafter, the patient’s diagnosis and stage are discussed and agreed on, and a consensus regarding patient management is reached.
Pathology reviews are valuable in establishing patients’ diagnoses. Kronz et al.1 reported 86 (1.4%) diagnostic revisions in a second review of 6171 surgical pathology cases. Two body sites were significantly more likely to result in a revised diagnosis: the serosal surfaces (9.5%) (P < 0.0001) and the female reproductive tract (5.1%) (P < 0.0001).1 Pathology reviews can be costly; however, the benefit of ensuring that patients receive the appropriate management for their tumour status2,3 outweighs the cost.1 Consequently, a review of pathology in all or selected patients, before commencement of definitive management, has become routine in most oncology practices. Diagnostic discrepancies may be found in 7% to 35% of gynaecologic oncology cases,4–8 and 20% of these discrepancies are categorized as major because they result in changes in patients’ management.6 A 6-month, prospective review of the value of gynaecologic TBR at the TBCC that was carried out in 20022 reported diagnostic discrepancies in 35 of 203 (17.2%) cases. Discrepancies were minor in 16 (7.9%) cases and major (including critical) in 19 (9.3%) cases. Because the impact of gynaecologic TBR at the TBCC had not been evaluated since 2002, this retrospective study of the frequency and additional management impact of discrepant diagnoses was carried out to monitor the quality of the gynaecologic oncology practice. MATERIALS AND METHODS
Approval for the study was obtained from the Health Research Ethics Board of the Alberta Cancer Committee.
2 • ■■ JOGC ■■ ■■
Thereafter, a retrospective chart review was completed on all 1604 cases discussed over a 3-year period at the gynaecologic TBRs from January 1, 2011 to December 31, 2013 at the TBCC. Variables collected included patients’ demographics, pre-TBR pathology diagnoses, post-TBR patients’ diagnoses, tumour grade, tumour stage, management recommendations, and resultant patients’ treatment. The information extracted from the patients’ TBCC charts, electronic medical records, and TBR presentations was entered using SPSS software version 17.0 for Windows (SPSS Inc., Chicago, IL). In this study, a discrepancy was defined as a change in tumour type, grade, site, or stage. A major discrepancy affected additional patient management, whereas a minor discrepancy did not.4 Descriptive statistics were used to analyze the data (SPSS software version 17.0 for Windows). RESULTS
The mean age of the 1604 patients was 57.6 (standard deviation ± 14.1) and ranged from 18 to 93. In total, there were 212 (13.2%) discrepancies; 54 (3.4%) major and 158 (9.9%) minor (Table 1). Reasons for most of the discrepancies were categorized as changes in the following: tumour primary site, histotype, grade, and stage; status of lymph vascular invasion; and pelvic wash cytology. Other reasons included a change from a malignant diagnosis to a pre-invasive or benign diagnosis, a change in the status of the surgical margins,
Table 1. The 212 discrepancies by type and major or minor category Discrepancies Minor n (%)
Type of discrepancy
Major n (%)
Total n (%)
Change in tumour stage
11 (20.4)
10 (6.3)
21 (9.9)
Change in tumour grade
-
35 (22.2)
35 (16.5)
Change in tumour histotype
9 (16.7)
39 (24.7)
48 (22.6)
Change in tumour primary site
11 (20.4)
6 (3.8)
17 (8.0)
Change in LVSI status
3 (5.6)
14 (8.9)
17 (8.0)
Change in peritoneal cytology status
8 (14.8)
16 (10.1)
24 (11.3)
Malignancy confirmed
6 (11.1)
12 (7.6)
18 (8.5)
—
1 (0.6)
1 (0.5)
Pre-invasive status confirmed
1 (1.9)
8 (5.1)
9 (4.2)
Benign tumour confirmed
2 (3.7)
3 (1.9)
5 (2.4)
Additional pathology/addition positive area/multiple change
3 (3.7)
11(7.0)
14 (6.6)
Change in margin status
—
3 (1.9)
3 (1.4)
Total
54
158
212
Borderline status confirmed
LVSI: lymph vascular space invasion.
Impact of Gynaecological Tumour Board Rounds on Patient Care
Table 2. The 1604 tumour board rounds: discrepancies by anatomic site
Anatomic site
Total TBRs n (%)
Total n (%)
Discrepancy Major n (%)
Minor n (%)
Fallopian tube
24 (1.5)
1 (0.5)
—
1 (0.6)
Metastatic and unknown primary
40 (2.5)
5 (2.4)
4 (7.4)
1 (0.6)
Müllerian tract NOS
171 (10.7)
3 (1.4)
1 (1.9)
2 (1.3)
Other
11 (0.7)
1 (0.5)
1(1.9)
—
Ovary
398 (24.8)
43 (20.3)
8 (14.8)
35 (22.2)
Peritoneum
19 (1.2)
4 (1.9)
—
4 (2.5)
Uterine endometrium
689 (43)
102 (48.1)
27 (50.0)
75 (47.5)
Uterine, other
52 (3.2)
7 (3.3)
3 (5.6)
4 (2.5)
Cervix
216 (13.5)
27 (12.7)
5 (9.3)
22 (13.9)
Dual sites NOS
32 (2.0)
5 (2.4)
3 (5.6)
2 (1.3)
Vagina
19 (1.2)
3 (1.4)
—
3 (1.9)
Vulva
87 (5.4)
11 (5.2)
2 (3.7)
9 (5.7)
Total
1604 (100.0)
212 (13.2)
54 (3.4)
158 (9.9)
TBRs: tumour board rounds; NOS: not otherwise specified.
additional areas of invasion noted, or partially completed pathology reports. Most of the total and minor discrepancies reflected changes in tumour histotype, whereas changes in the primary site or tumour stage accounted for most of the major discrepancies.
Most (43%, n = 689) of the TBRs were for endometrial abnormalities, followed by ovarian (25%, n = 401) and cervical abnormalities (23%, n = 369) (Table 2). Lesser numbers of TBRs were to discuss “metastatic sites,” which included metastatic disease to gynaecologic sites such as breast (0.4%, n = 6), gastrointestinal (1.6%, n = 25), unknown primary (0.4%, n = 6), and other sites (0.6%, n = 9); “dual sites,” which were ovary and uterus (0.9%, n = 14), ovary and breast (0.9%, n = 14), and uterus and breast (0.1%, n = 1); and “müllerian NOS (not otherwise specified)” (0.7%, n = 11), which consisted of three serous tumours, seven adenocarcinomas, and one squamous tumour. “Other gynaecology” sites comprising the vulva and vagina and conditions such as gestational trophoblastic neoplasia were also reviewed. Other cases of non-gynaecologic tumour sites that were reviewed included pathology involving the urinary tract, anus, and appendix. Major discrepancies involved every site except for the fallopian tube, peritoneum, and vagina, and they most frequently involved the ovary. Minor discrepancies involved all sites and most frequently involved the endometrium. Among the 54 (25.5%) major discrepancies, 18 (33.3%) occurred in patients who had additional treatment cancelled, 17 (31.5%) required chemotherapy, 4 (7.4%) required a change in the chemotherapy regimen, 10 (18.5%) required additional surgery, and 5 (9.3%) required radiation or chemoradiation (Table 3). In comparison with the 2002 data, the overall and major discrepancy rates were lower (13.2% vs. 17.2% and 3.4% vs
Table 3. The 54 major discrepancies and changes in management
Major discrepancies
n (%)
Treatment cancelled n (%)
Chemotherapy n (%)
Management change Chemotherapy Additional changed surgery n (%) n (%)
Radiation n (%)
Chemoradiation n (%)
Changes in: Tumour stage
11(20.4)
5 (9.3)
6 (11.1)
0
0
0
0
Tumour histotype
9 (16.7)
1 (1.9)
1 (1.9)
1 (1.9)
4 (7.4)
1 (1.9)
1 (1.9)
Tumour primary site
1 (1.9)
11(20.4)
1 (1.9)
6 (11.1)
3 (5.6)
0
0
LVSI status
3 (5.6)
2 (3.7)
0
0
0
1 (1.9)
0
Peritoneal cytology
8 (14.8)
5 (9.3)
3 (5.6)
0
0
0
0
6 (11.1)
0
1 (1.9)
0
5 (9.3)
0
0
Confirmed: Malignancy Pre-invasion
1 (1.9)
1 (1.9)
0
0
0
0
0
Benign
2 (3.7)
2 (3.7)
0
0
0
0
0
Additional pathology/ additional positive area/ multiple changes
3 (5.6)
1 (1.9)
0
0
1 (1.9)
1 (1.9)
0
54 (100)
18 (33.3)
17 (31.5)
4 (7.4)
10 (18.5)
3 (5.6)
2 (3.7)
Total
LVSI: lymph vascular space invasion.
■■ JOGC ■■ ■■ • 3
GYNAECOLOGY
Table 4. Summary of the published literature on gynaecologic tumour board rounds discrepancies Number of TBRs
Discrepancies n (%)
Major discrepancies n (%)
Minor Discrepancies n (%)
Prospective
351
99 (28.2)
29 (8.3)
70 (19.9)
Tumour board
2004
Retrospective Chart review
459
32 (6.9%)
23 (5.0)
9 (1.9)
Tumour board
Gatcliffe et al.5
2008
Prospective
153
53 (34.6)
13 (8.4)
40 (26.1)
Tumour board
Cohen et al.4
2009
Retrospective Chart review
509
46 (9.0)
30 (5.9)
16 (3.1)
Tumour board
Greer et al.6
2010
Retrospective
526
140 (26.6)
104 (19.8)
36 (6.8)
Tumour board
Halliday et al.
2018
Retrospective
1604
212 (13.2)
54 (3.4)
158 (9.9)
Tumour board
3
1998
Retrospective
720
119 (16)
15 (2)
104 (14)
Review of pathology
Selman et al.15
1999
Retrospective
295
50 (16.9)
14 (4.7)
36 (12.2)
Review of pathology
Chan et al.
1999
Retrospective
569
108 (19)
37(6.5)
71 (12.5)
Review of pathology
Chafe et al.12
2000
Retrospective
599
200 (33)
63/514 (12)
—
Review of pathology
Eskander et al.14
2013
Retrospective
279
44 (15.8)
19 (6.8)
25 (9)
Review of pathology
Weir et al.
2003
Retrospective
1000
68 (6.8)
37 (3.7)
31 (31)
Review of pathology
Ogilvie et al.2
2003
Prospective
203
29 (14.3)
13 (6.4)
16 (7.8)
Review of pathology
Authors and reference
Publication year
Khalifa et al.7
2003
Santoso et al.8
Santoso et al. 11
16
Study type
Review format
TBRs: tumour board rounds.
9.3%, respectively), but the minor discrepancy frequency was higher (9.9% vs. 7.9%). DISCUSSION
This analysis represents one of the largest reported series on the frequency and type of diagnostic discrepancies found after gynaecologic oncology TBRs and on their impact on additional patient management. Of the 1604 cases reviewed, 212 (13.2%) were found to have discrepancies, 54(3.4%) major and 158 (9.9%) minor. Among the 54 (25.5%) major discrepancies, 18 (33.3%) caused additional treatment to be cancelled, and 36 (66.7%) required additional, or a change to additional, treatment. The study showed a reduction in the overall number of discrepancies and, in particular, major discrepancies since 2002. This finding highlights the continued improvement of the practice of gynaecologic oncology and the importance of TBRs to that improvement. The literature review suggests that the overall discrepancy frequency for gynaecologic oncology TBR ranges from 6.9% to 34.6%, and the major discrepancy frequency is 20% (Table 4). Although there was considerable heterogeneity across the studies, the criteria for major and minor discrepancies were similar, and thus comparisons across studies are feasible. A prior review of gynaecologic oncology cases at our institution in 2002,2 published in abstract form, found that discrepancies were identified in 35 of 203 (17.2%) patients and were categorized as 19 (9.3%) major (including critical) and 16 (7.9%) minor. At that time, the pathology of all patients with gynaecologic cancer was
4 • ■■ JOGC ■■ ■■
reviewed by a gynaecologic pathologist. Since that time, however, the pathology review practice at our institution changed, and only select patients had a pathology review. This change was brought about by improvements in the pathology quality assurance practices in the laboratories in Alberta that led to a reduction in the need to perform pathology reviews on all patients initially referred to the TBCC. Some of these practice improvements included the implementation of a Reporting and Learning System for Patient Safety, which standardized the definitions of degrees of harm (online Appendix Figure); the adoption of synoptic reporting; and performance of internal pathology reviews (i.e., second pathology opinion) within the laboratory before the pathology reports were finalized and reached the level of the patient and her physician. Our current overall discrepancy rate of 13.2% is towards the lower end of the reported range, and our major discrepancy rate of 3.4% is well below 20%. The rates are also considerably lower than our 2002 rates, and these very favorable results likely reflect the implementation of the aforementioned quality assurance measure. There are other possible explanations. Results of some of the reported literature and the 2002 review reflected the practice standards of the time, which would differ from those of the current review carried out nearly 10 years later. For example, the FIGO staging of gynaecologic cancers was revised in 2009.9 Thus, discrepancies in the staging of endometrial cancers as a FIGO 2A (cervical cancerization) and 2B (cervical stromal invasion) in the 1988 version10 would not be an issue in the current study because only cervical stromal invasion is a factor in stage 2 cancers in FIGO 2009.
Impact of Gynaecological Tumour Board Rounds on Patient Care
When all gynaecologic cancer sites are considered, discrepancies in uterine cancer pathology are the most common.5,7 Uterine endometrium was the most frequent (n = 689, 43%) cancer site discussed, and this site had most (48%) of the discrepancies and also had most (50%) of the major discrepancies. In many jurisdictions, surgery and further management are often performed when there is no regular tumour board review of the pathology. This could lead to either inadequate or excessive treatment postoperatively. It is important to note that every site in the müllerian tract had a discrepancy, although major discrepancies did not occur in the fallopian tube, vagina, and peritoneum. The most frequent discrepancies in this study were changes in tumour histotype (22.6%), grade (16.5%), and positive or negative status (11.3%) of the peritoneal cytology. The least frequent discrepancy was “borderline confirmed” (i.e., non-invasive components were noted). Changes to treatment or additional treatments were the two most (66.7%) frequent therapeutic decisions among women with major discrepancies, although it is notable that 33.3% had their additional treatment cancelled (Table 3). From the published literature, the most common changes to patient management that resulted from TBRs were the addition of chemotherapy and surgery.8,11 Possible limitations of this review include the retrospective design. This may have resulted in the reinterpretation of management by individuals initially not involved. We did not focus on the cost-effectiveness of the tumour board reviews.1,2,11 Although they are costly, the potential impact on patient management cannot be understated. TBRs provide a safe learning environment for clinicians and trainees and reinforce the need for quality assurance practices. This ultimately improves patient care.5,8,12 Because the tumour board review occurs before the initiation of additional therapies, there were no cases of harm as defined by health services “Clinical Impact Categories”13 (online Appendix Figure). The interaction with non-gynaecologic oncology faculty present at the tumour board is often the only opportunity for trainees to gain a broader perspective on the roles of other disciplines in the overall care of gynaecologic cancer patients and where the role of the gynaecologic pathologist is pivotal.5 CONCLUSION
The discrepancy frequency in this retrospective study was 13.2%, and nearly 26% led to a change in additional management. The study emphasizes the necessity of having TBRs as a quality tool in the management of gynaecologic tumours. It highlights the need to have these multidisciplinary rounds
before postoperative treatment of these gynaecologic malignant diseases. It also highlights the need for ongoing audits to determine the impact of prior quality assurance measures. SUPPLEMENTARY DATA
Supplementary data related to this article can be found at https://doi.org/10.1016/j.jogc.2018.05.026. REFERENCES 1. Kronz JD, Westra WH, Epstein JI. Mandatory second opinion surgical pathology at a large referral hospital. Cancer 1999;86:2426–35. 2. Ogilvie R, Craighead P, Ghatage P, et al. Assessment of diagnostic discrepancies identified by pathologic review of gynecologic oncology patients. Paper presented at: United States and Canadian Academy of Pathology 92nd Annual Meeting. March 22–28, 2003; Washington, DC: A1452. 3. Santoso JT, Coleman RL, Voet RL, et al. Pathology slide review in gynecologic oncology. Obstet Gynecol 1998;91:730–4. 4. Cohen P, Tan AL, Penman A. The multidisciplinary tumor conference in gynecologic oncology–does it alter management? Int J Gynecol Cancer 2009;19:1470–2. 5. Gatcliffe T, Coleman R. Tumor board: more than treatment planning–a 1-year prospective survey. J Cancer Educ 2008;23:235–7. 6. Greer HO, Frederick PJ, Falls NM, et al. Impact of a weekly multidisciplinary tumor board conference on the management of women with gynecologic malignancies. Int J Gynecol Cancer 2010;20:1321–5. 7. Khalifa MA, Dodge J, Covens A, et al. Slide review in gynecologic oncology ensures completeness of reporting and diagnostic accuracy. Gynecol Oncol 2003;90:425–30. 8. Santoso JT, Schwertner B, Coleman RL, et al. Tumor board in gynecologic oncology. Int J Gynecol Cancer 2004;14:206–9. 9. Pecorelli S. Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium. Int J Gynaecol Obstet 2009;105:103–4. 10. Shepherd JH. Revised FIGO staging for gynaecological cancer. Br J Obstet Gynaecol 1989;96:889–92. 11. Chan YM, Cheung AN, Cheng DK, et al. Pathology slide review in gynecologic oncology: routine or selective? Gynecol Oncol 1999;75:267– 71. 12. Chafe S, Honore L, Pearcey R, et al. An analysis of the impact of pathology review in gynecologic cancer. Int J Radiat Oncol Biol Phys 2000;48:1433–8. 13. Zarbo RJ, Meier FA, Raab SS. Error detection in anatomic pathology. Arch Pathol Lab Med 2005;129:1237–45. 14. Eskander RN, Baruah J, Nayak R, et al. Outside slide review in gynecologic oncology: impact on patient care and treatment. Int J Gynecol Pathol 2013;32:293–8. 15. Selman AE, Niemann TH, Fowler JM, et al. Quality assurance of second opinion pathology in gynecologic oncology. Obstet Gynecol 1999;94:302– 6. 16. Weir MM, Jan E, Colgan TJ. Interinstitutional pathology consultations. A reassessment. Am J Clin Pathol 2003;120:405–12.
■■ JOGC ■■ ■■ • 5
GYNAECOLOGY
Appendix FIGURE
Quick reference guide: degree of harm. ER: emergency room; TYSH: type & screen. (From the Alberta Health Services, Edmonton, AB.)
5.e1 • ■■ JOGC ■■ ■■
Impact of Gynaecological Tumour Board Rounds on Patient Care Darron A. Halliday, MD;1 Khadija Warfa, MD;2 Jill Nation, MD;3 Pamela Chu, MD;3 Gregg Nelson, MD, PhD;3 Sarah Glaze, MD;3 Jennifer Mateshaytis, MD;4 Máire A. Duggan, MD;5 Prafull Ghatage, MD3 1
Department of Obstetrics & Gynecology, Princess Margaret Hospital, Nassau, Bahamas
2
Department of Obstetrics & Gynecology, Aga Khan University Hospital, Nairobi, Kenya
D.A. Halliday
3
Department of Oncology, University of Calgary, Calgary, Canada
4
Department of Obstetrics & Gynecology, University of Alberta, Edmonton, Canada
5
Department of Pathology and Laboratory Medicine, University of Calgary, Calgary, Canada
Abstract
Résumé
Objective: The goal of this study was to determine the impact of tumour board rounds (TBRs) on the additional management of patients with gynaecologic malignancy.
Objectif : Cette étude visait à déterminer l’incidence des rencontres du comité de thérapie du cancer sur la prise en charge ultérieure des patientes atteintes de cancers gynécologiques.
Methods: A retrospective chart review of 1604 patients discussed between January 2011 and December 2013 at gynaecologic TBRs was conducted to determine the frequency and type of diagnostic discrepancies found post-TBRs and their potential impact on additional patient management. A discrepancy was defined as major if it affected patient management by cancelling, initiating, or modifying treatment; otherwise, the discrepancy was minor. Data collected included patients’ demographics, pre- and post-TBR diagnoses, and management.
Méthodologie : Nous avons mené une évaluation rétrospective des dossiers de 1 604 patientes ayant fait l’objet de rencontres du comité de thérapie du cancer gynécologique entre janvier 2011 et décembre 2013 pour déterminer la fréquence et le type des écarts diagnostiques relevés après les rencontres et leur incidence potentielle sur la prise en charge ultérieure des patientes. Un écart était jugé majeur s’il modifiait la prise en charge par l’annulation, l’initiation ou la modification d’un traitement; sinon, l’écart était jugé mineur. Les données recueillies comprenaient les données démographiques des patientes, le diagnostic avant et après la rencontre, et la prise en charge.
Results: The patients’ mean age was 57.6 ± 14.1. Endometrial disease accounted for (43%) of the TBRs. The remaining sites were ovarian (25%), cervical (23%), and others (9%). Overall, 13.2% (n = 212) had a discrepancy; 3.4% (n = 54) of these discrepancies were major, and 9.9% (n = 158) were minor. Most major discrepancies related to changes in the tumours’ primary site or stage, and most minor discrepancies were related to changes in tumour histotype. Among the 54 (25.5%) major discrepancies, 18 (33.3%) occurred in patients who had their additional management cancelled, 17 (31.5%) required chemotherapy, 4 (7.4%) required a change in the chemotherapy regimen, 10 (18.5%) required additional surgery, and 5 (9.3%) required chemoradiation. Conclusion: The 13% frequency of discrepancies, approximately 26% of which were major and resulted in changes in patient management, highlights the importance of TBRs as a quality tool.
Key words: Tumour board rounds, gynaecological cancer, patient management Corresponding Author: Dr. Darron Halliday, Department of Obstetrics & Gynecology, Princess Margaret Hospital, Nassau, Bahamas. [email protected] Competing interests: The authors declare that they have no competing interests. Received on March 17, 2018 Accepted on May 15, 2018
Résultats : L’âge moyen des patientes était de 57,6 ± 14,1 ans. Les cancers de l’endomètre étaient les plus fréquents, avec 43 % des dossiers. Les autres cancers touchaient les ovaires (25 %), le col de l’utérus (23 %) et d’autres organes (9 %). Dans l’ensemble, 13,2 % (n = 212) des cas présentaient un écart; dans 3,4 % (n = 54) des cas, l’écart était majeur, et il était mineur dans 9,9 % (n = 158) des cas. La plupart des écarts majeurs étaient liés à des changements du site primaire ou du stade de la tumeur, et la plupart des écarts mineurs étaient liés à des changements de l’histotype de la tumeur. Des 54 (25,5 %) écarts majeurs, 18 (33,3 %) ont entraîné l’annulation de la prise en charge ultérieure, 17 (31,5 %) ont entraîné une chimiothérapie, 4 (7,4%) ont nécessité un changement du régime de chimiothérapie, 10 (18,5 %) ont nécessité une chirurgie supplémentaire et 5 (9,3 %) ont nécessité une chimioradiothérapie. Conclusions : La fréquence des écarts (13 %), dont environ 26 % étaient majeurs et ont entraîné des changements dans la prise en charge des patientes, met en lumière l’importance des rencontres du comité de thérapie du cancer comme outil d’assurance de la qualité. Copyright © 2018 The Society of Obstetricians and Gynaecologists of Canada/La Société des obstétriciens et gynécologues du Canada. Published by Elsevier Inc. All rights reserved.
J Obstet Gynaecol Can ■■;■■(■■):■■–■■ https://doi.org/10.1016/j.jogc.2018.05.026
■■ JOGC ■■ ■■ • 1
GYNAECOLOGY
INTRODUCTION
T
umour board rounds (TBRs) are an integral part of oncology practice. Sometimes revisions of the pre-TBR diagnosis lead to changes in additional management, and different perspectives on management occur when multidisciplinary teams come together. The gynaecologic TBRs at the Tom Baker Cancer Centre (TBCC) in Calgary, Alberta, consist of gynaecologic and radiation oncologists, gynaecologic pathologists, and psychologists. This group meets weekly to discuss cases managed by the gynaecologic oncology service or referred for additional oncologic management by other physicians in southern Alberta. A typical TBR discussion involves an oral presentation summarizing the patient’s history, physical examination findings, imaging, laboratory investigations, surgical procedures, and pathology reports. The gynaecologic pathologist reviews the relevant and pertinent pathology of each case before the TBR and presents the results. Thereafter, the patient’s diagnosis and stage are discussed and agreed on, and a consensus regarding patient management is reached.
Pathology reviews are valuable in establishing patients’ diagnoses. Kronz et al.1 reported 86 (1.4%) diagnostic revisions in a second review of 6171 surgical pathology cases. Two body sites were significantly more likely to result in a revised diagnosis: the serosal surfaces (9.5%) (P < 0.0001) and the female reproductive tract (5.1%) (P < 0.0001).1 Pathology reviews can be costly; however, the benefit of ensuring that patients receive the appropriate management for their tumour status2,3 outweighs the cost.1 Consequently, a review of pathology in all or selected patients, before commencement of definitive management, has become routine in most oncology practices. Diagnostic discrepancies may be found in 7% to 35% of gynaecologic oncology cases,4–8 and 20% of these discrepancies are categorized as major because they result in changes in patients’ management.6 A 6-month, prospective review of the value of gynaecologic TBR at the TBCC that was carried out in 20022 reported diagnostic discrepancies in 35 of 203 (17.2%) cases. Discrepancies were minor in 16 (7.9%) cases and major (including critical) in 19 (9.3%) cases. Because the impact of gynaecologic TBR at the TBCC had not been evaluated since 2002, this retrospective study of the frequency and additional management impact of discrepant diagnoses was carried out to monitor the quality of the gynaecologic oncology practice. MATERIALS AND METHODS
Approval for the study was obtained from the Health Research Ethics Board of the Alberta Cancer Committee.
2 • ■■ JOGC ■■ ■■
Thereafter, a retrospective chart review was completed on all 1604 cases discussed over a 3-year period at the gynaecologic TBRs from January 1, 2011 to December 31, 2013 at the TBCC. Variables collected included patients’ demographics, pre-TBR pathology diagnoses, post-TBR patients’ diagnoses, tumour grade, tumour stage, management recommendations, and resultant patients’ treatment. The information extracted from the patients’ TBCC charts, electronic medical records, and TBR presentations was entered using SPSS software version 17.0 for Windows (SPSS Inc., Chicago, IL). In this study, a discrepancy was defined as a change in tumour type, grade, site, or stage. A major discrepancy affected additional patient management, whereas a minor discrepancy did not.4 Descriptive statistics were used to analyze the data (SPSS software version 17.0 for Windows). RESULTS
The mean age of the 1604 patients was 57.6 (standard deviation ± 14.1) and ranged from 18 to 93. In total, there were 212 (13.2%) discrepancies; 54 (3.4%) major and 158 (9.9%) minor (Table 1). Reasons for most of the discrepancies were categorized as changes in the following: tumour primary site, histotype, grade, and stage; status of lymph vascular invasion; and pelvic wash cytology. Other reasons included a change from a malignant diagnosis to a pre-invasive or benign diagnosis, a change in the status of the surgical margins,
Table 1. The 212 discrepancies by type and major or minor category Discrepancies Minor n (%)
Type of discrepancy
Major n (%)
Total n (%)
Change in tumour stage
11 (20.4)
10 (6.3)
21 (9.9)
Change in tumour grade
-
35 (22.2)
35 (16.5)
Change in tumour histotype
9 (16.7)
39 (24.7)
48 (22.6)
Change in tumour primary site
11 (20.4)
6 (3.8)
17 (8.0)
Change in LVSI status
3 (5.6)
14 (8.9)
17 (8.0)
Change in peritoneal cytology status
8 (14.8)
16 (10.1)
24 (11.3)
Malignancy confirmed
6 (11.1)
12 (7.6)
18 (8.5)
—
1 (0.6)
1 (0.5)
Pre-invasive status confirmed
1 (1.9)
8 (5.1)
9 (4.2)
Benign tumour confirmed
2 (3.7)
3 (1.9)
5 (2.4)
Additional pathology/addition positive area/multiple change
3 (3.7)
11(7.0)
14 (6.6)
Change in margin status
—
3 (1.9)
3 (1.4)
Total
54
158
212
Borderline status confirmed
LVSI: lymph vascular space invasion.
Impact of Gynaecological Tumour Board Rounds on Patient Care
Table 2. The 1604 tumour board rounds: discrepancies by anatomic site
Anatomic site
Total TBRs n (%)
Total n (%)
Discrepancy Major n (%)
Minor n (%)
Fallopian tube
24 (1.5)
1 (0.5)
—
1 (0.6)
Metastatic and unknown primary
40 (2.5)
5 (2.4)
4 (7.4)
1 (0.6)
Müllerian tract NOS
171 (10.7)
3 (1.4)
1 (1.9)
2 (1.3)
Other
11 (0.7)
1 (0.5)
1(1.9)
—
Ovary
398 (24.8)
43 (20.3)
8 (14.8)
35 (22.2)
Peritoneum
19 (1.2)
4 (1.9)
—
4 (2.5)
Uterine endometrium
689 (43)
102 (48.1)
27 (50.0)
75 (47.5)
Uterine, other
52 (3.2)
7 (3.3)
3 (5.6)
4 (2.5)
Cervix
216 (13.5)
27 (12.7)
5 (9.3)
22 (13.9)
Dual sites NOS
32 (2.0)
5 (2.4)
3 (5.6)
2 (1.3)
Vagina
19 (1.2)
3 (1.4)
—
3 (1.9)
Vulva
87 (5.4)
11 (5.2)
2 (3.7)
9 (5.7)
Total
1604 (100.0)
212 (13.2)
54 (3.4)
158 (9.9)
TBRs: tumour board rounds; NOS: not otherwise specified.
additional areas of invasion noted, or partially completed pathology reports. Most of the total and minor discrepancies reflected changes in tumour histotype, whereas changes in the primary site or tumour stage accounted for most of the major discrepancies.
Most (43%, n = 689) of the TBRs were for endometrial abnormalities, followed by ovarian (25%, n = 401) and cervical abnormalities (23%, n = 369) (Table 2). Lesser numbers of TBRs were to discuss “metastatic sites,” which included metastatic disease to gynaecologic sites such as breast (0.4%, n = 6), gastrointestinal (1.6%, n = 25), unknown primary (0.4%, n = 6), and other sites (0.6%, n = 9); “dual sites,” which were ovary and uterus (0.9%, n = 14), ovary and breast (0.9%, n = 14), and uterus and breast (0.1%, n = 1); and “müllerian NOS (not otherwise specified)” (0.7%, n = 11), which consisted of three serous tumours, seven adenocarcinomas, and one squamous tumour. “Other gynaecology” sites comprising the vulva and vagina and conditions such as gestational trophoblastic neoplasia were also reviewed. Other cases of non-gynaecologic tumour sites that were reviewed included pathology involving the urinary tract, anus, and appendix. Major discrepancies involved every site except for the fallopian tube, peritoneum, and vagina, and they most frequently involved the ovary. Minor discrepancies involved all sites and most frequently involved the endometrium. Among the 54 (25.5%) major discrepancies, 18 (33.3%) occurred in patients who had additional treatment cancelled, 17 (31.5%) required chemotherapy, 4 (7.4%) required a change in the chemotherapy regimen, 10 (18.5%) required additional surgery, and 5 (9.3%) required radiation or chemoradiation (Table 3). In comparison with the 2002 data, the overall and major discrepancy rates were lower (13.2% vs. 17.2% and 3.4% vs
Table 3. The 54 major discrepancies and changes in management
Major discrepancies
n (%)
Treatment cancelled n (%)
Chemotherapy n (%)
Management change Chemotherapy Additional changed surgery n (%) n (%)
Radiation n (%)
Chemoradiation n (%)
Changes in: Tumour stage
11(20.4)
5 (9.3)
6 (11.1)
0
0
0
0
Tumour histotype
9 (16.7)
1 (1.9)
1 (1.9)
1 (1.9)
4 (7.4)
1 (1.9)
1 (1.9)
Tumour primary site
1 (1.9)
11(20.4)
1 (1.9)
6 (11.1)
3 (5.6)
0
0
LVSI status
3 (5.6)
2 (3.7)
0
0
0
1 (1.9)
0
Peritoneal cytology
8 (14.8)
5 (9.3)
3 (5.6)
0
0
0
0
6 (11.1)
0
1 (1.9)
0
5 (9.3)
0
0
Confirmed: Malignancy Pre-invasion
1 (1.9)
1 (1.9)
0
0
0
0
0
Benign
2 (3.7)
2 (3.7)
0
0
0
0
0
Additional pathology/ additional positive area/ multiple changes
3 (5.6)
1 (1.9)
0
0
1 (1.9)
1 (1.9)
0
54 (100)
18 (33.3)
17 (31.5)
4 (7.4)
10 (18.5)
3 (5.6)
2 (3.7)
Total
LVSI: lymph vascular space invasion.
■■ JOGC ■■ ■■ • 3
GYNAECOLOGY
Table 4. Summary of the published literature on gynaecologic tumour board rounds discrepancies Number of TBRs
Discrepancies n (%)
Major discrepancies n (%)
Minor Discrepancies n (%)
Prospective
351
99 (28.2)
29 (8.3)
70 (19.9)
Tumour board
2004
Retrospective Chart review
459
32 (6.9%)
23 (5.0)
9 (1.9)
Tumour board
Gatcliffe et al.5
2008
Prospective
153
53 (34.6)
13 (8.4)
40 (26.1)
Tumour board
Cohen et al.4
2009
Retrospective Chart review
509
46 (9.0)
30 (5.9)
16 (3.1)
Tumour board
Greer et al.6
2010
Retrospective
526
140 (26.6)
104 (19.8)
36 (6.8)
Tumour board
Halliday et al.
2018
Retrospective
1604
212 (13.2)
54 (3.4)
158 (9.9)
Tumour board
3
1998
Retrospective
720
119 (16)
15 (2)
104 (14)
Review of pathology
Selman et al.15
1999
Retrospective
295
50 (16.9)
14 (4.7)
36 (12.2)
Review of pathology
Chan et al.
1999
Retrospective
569
108 (19)
37(6.5)
71 (12.5)
Review of pathology
Chafe et al.12
2000
Retrospective
599
200 (33)
63/514 (12)
—
Review of pathology
Eskander et al.14
2013
Retrospective
279
44 (15.8)
19 (6.8)
25 (9)
Review of pathology
Weir et al.
2003
Retrospective
1000
68 (6.8)
37 (3.7)
31 (31)
Review of pathology
Ogilvie et al.2
2003
Prospective
203
29 (14.3)
13 (6.4)
16 (7.8)
Review of pathology
Authors and reference
Publication year
Khalifa et al.7
2003
Santoso et al.8
Santoso et al. 11
16
Study type
Review format
TBRs: tumour board rounds.
9.3%, respectively), but the minor discrepancy frequency was higher (9.9% vs. 7.9%). DISCUSSION
This analysis represents one of the largest reported series on the frequency and type of diagnostic discrepancies found after gynaecologic oncology TBRs and on their impact on additional patient management. Of the 1604 cases reviewed, 212 (13.2%) were found to have discrepancies, 54(3.4%) major and 158 (9.9%) minor. Among the 54 (25.5%) major discrepancies, 18 (33.3%) caused additional treatment to be cancelled, and 36 (66.7%) required additional, or a change to additional, treatment. The study showed a reduction in the overall number of discrepancies and, in particular, major discrepancies since 2002. This finding highlights the continued improvement of the practice of gynaecologic oncology and the importance of TBRs to that improvement. The literature review suggests that the overall discrepancy frequency for gynaecologic oncology TBR ranges from 6.9% to 34.6%, and the major discrepancy frequency is 20% (Table 4). Although there was considerable heterogeneity across the studies, the criteria for major and minor discrepancies were similar, and thus comparisons across studies are feasible. A prior review of gynaecologic oncology cases at our institution in 2002,2 published in abstract form, found that discrepancies were identified in 35 of 203 (17.2%) patients and were categorized as 19 (9.3%) major (including critical) and 16 (7.9%) minor. At that time, the pathology of all patients with gynaecologic cancer was
4 • ■■ JOGC ■■ ■■
reviewed by a gynaecologic pathologist. Since that time, however, the pathology review practice at our institution changed, and only select patients had a pathology review. This change was brought about by improvements in the pathology quality assurance practices in the laboratories in Alberta that led to a reduction in the need to perform pathology reviews on all patients initially referred to the TBCC. Some of these practice improvements included the implementation of a Reporting and Learning System for Patient Safety, which standardized the definitions of degrees of harm (online Appendix Figure); the adoption of synoptic reporting; and performance of internal pathology reviews (i.e., second pathology opinion) within the laboratory before the pathology reports were finalized and reached the level of the patient and her physician. Our current overall discrepancy rate of 13.2% is towards the lower end of the reported range, and our major discrepancy rate of 3.4% is well below 20%. The rates are also considerably lower than our 2002 rates, and these very favorable results likely reflect the implementation of the aforementioned quality assurance measure. There are other possible explanations. Results of some of the reported literature and the 2002 review reflected the practice standards of the time, which would differ from those of the current review carried out nearly 10 years later. For example, the FIGO staging of gynaecologic cancers was revised in 2009.9 Thus, discrepancies in the staging of endometrial cancers as a FIGO 2A (cervical cancerization) and 2B (cervical stromal invasion) in the 1988 version10 would not be an issue in the current study because only cervical stromal invasion is a factor in stage 2 cancers in FIGO 2009.
Impact of Gynaecological Tumour Board Rounds on Patient Care
When all gynaecologic cancer sites are considered, discrepancies in uterine cancer pathology are the most common.5,7 Uterine endometrium was the most frequent (n = 689, 43%) cancer site discussed, and this site had most (48%) of the discrepancies and also had most (50%) of the major discrepancies. In many jurisdictions, surgery and further management are often performed when there is no regular tumour board review of the pathology. This could lead to either inadequate or excessive treatment postoperatively. It is important to note that every site in the müllerian tract had a discrepancy, although major discrepancies did not occur in the fallopian tube, vagina, and peritoneum. The most frequent discrepancies in this study were changes in tumour histotype (22.6%), grade (16.5%), and positive or negative status (11.3%) of the peritoneal cytology. The least frequent discrepancy was “borderline confirmed” (i.e., non-invasive components were noted). Changes to treatment or additional treatments were the two most (66.7%) frequent therapeutic decisions among women with major discrepancies, although it is notable that 33.3% had their additional treatment cancelled (Table 3). From the published literature, the most common changes to patient management that resulted from TBRs were the addition of chemotherapy and surgery.8,11 Possible limitations of this review include the retrospective design. This may have resulted in the reinterpretation of management by individuals initially not involved. We did not focus on the cost-effectiveness of the tumour board reviews.1,2,11 Although they are costly, the potential impact on patient management cannot be understated. TBRs provide a safe learning environment for clinicians and trainees and reinforce the need for quality assurance practices. This ultimately improves patient care.5,8,12 Because the tumour board review occurs before the initiation of additional therapies, there were no cases of harm as defined by health services “Clinical Impact Categories”13 (online Appendix Figure). The interaction with non-gynaecologic oncology faculty present at the tumour board is often the only opportunity for trainees to gain a broader perspective on the roles of other disciplines in the overall care of gynaecologic cancer patients and where the role of the gynaecologic pathologist is pivotal.5 CONCLUSION
The discrepancy frequency in this retrospective study was 13.2%, and nearly 26% led to a change in additional management. The study emphasizes the necessity of having TBRs as a quality tool in the management of gynaecologic tumours. It highlights the need to have these multidisciplinary rounds
before postoperative treatment of these gynaecologic malignant diseases. It also highlights the need for ongoing audits to determine the impact of prior quality assurance measures. SUPPLEMENTARY DATA
Supplementary data related to this article can be found at https://doi.org/10.1016/j.jogc.2018.05.026. REFERENCES 1. Kronz JD, Westra WH, Epstein JI. Mandatory second opinion surgical pathology at a large referral hospital. Cancer 1999;86:2426–35. 2. Ogilvie R, Craighead P, Ghatage P, et al. Assessment of diagnostic discrepancies identified by pathologic review of gynecologic oncology patients. Paper presented at: United States and Canadian Academy of Pathology 92nd Annual Meeting. March 22–28, 2003; Washington, DC: A1452. 3. Santoso JT, Coleman RL, Voet RL, et al. Pathology slide review in gynecologic oncology. Obstet Gynecol 1998;91:730–4. 4. Cohen P, Tan AL, Penman A. The multidisciplinary tumor conference in gynecologic oncology–does it alter management? Int J Gynecol Cancer 2009;19:1470–2. 5. Gatcliffe T, Coleman R. Tumor board: more than treatment planning–a 1-year prospective survey. J Cancer Educ 2008;23:235–7. 6. Greer HO, Frederick PJ, Falls NM, et al. Impact of a weekly multidisciplinary tumor board conference on the management of women with gynecologic malignancies. Int J Gynecol Cancer 2010;20:1321–5. 7. Khalifa MA, Dodge J, Covens A, et al. Slide review in gynecologic oncology ensures completeness of reporting and diagnostic accuracy. Gynecol Oncol 2003;90:425–30. 8. Santoso JT, Schwertner B, Coleman RL, et al. Tumor board in gynecologic oncology. Int J Gynecol Cancer 2004;14:206–9. 9. Pecorelli S. Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium. Int J Gynaecol Obstet 2009;105:103–4. 10. Shepherd JH. Revised FIGO staging for gynaecological cancer. Br J Obstet Gynaecol 1989;96:889–92. 11. Chan YM, Cheung AN, Cheng DK, et al. Pathology slide review in gynecologic oncology: routine or selective? Gynecol Oncol 1999;75:267– 71. 12. Chafe S, Honore L, Pearcey R, et al. An analysis of the impact of pathology review in gynecologic cancer. Int J Radiat Oncol Biol Phys 2000;48:1433–8. 13. Zarbo RJ, Meier FA, Raab SS. Error detection in anatomic pathology. Arch Pathol Lab Med 2005;129:1237–45. 14. Eskander RN, Baruah J, Nayak R, et al. Outside slide review in gynecologic oncology: impact on patient care and treatment. Int J Gynecol Pathol 2013;32:293–8. 15. Selman AE, Niemann TH, Fowler JM, et al. Quality assurance of second opinion pathology in gynecologic oncology. Obstet Gynecol 1999;94:302– 6. 16. Weir MM, Jan E, Colgan TJ. Interinstitutional pathology consultations. A reassessment. Am J Clin Pathol 2003;120:405–12.
■■ JOGC ■■ ■■ • 5
GYNAECOLOGY
Appendix FIGURE
Quick reference guide: degree of harm. ER: emergency room; TYSH: type & screen. (From the Alberta Health Services, Edmonton, AB.)
5.e1 • ■■ JOGC ■■ ■■