Historically, the standard surgical management of endometrial cancer included a routine full pelvic lymph node dissection. Because two trials have shown no survival advantage for full lymph node dissection, critics of this approach argue that it increases surgical morbidity without any appreciable diagnostic benefit and no survival gain, especially in “low-risk” disease.1–3
Although no consensus exists as to the definition of “low-risk” patients, some investigators have shown that patients with grade 1–2 endometrioid histology, tumor size 2 cm or less, and 50% or less myometrial invasion on frozen section have only a 5% chance of nodal metastasis and have concluded that lymph node dissection may be omitted in them.4,5 Critics of selective lymph node dissection based on these criteria argue against the accuracy of frozen section and that some women with advanced cancer will be inadequately staged.6 Sentinel lymph node mapping is an alternative strategy. Proponents say that it diminishes the risk of lymphedema with routine lymph node dissection and minimizes the risk of inadequately staging certain patients with selective lymph node dissection, whereas critics point out the risk of inaccurate mapping and the possibility of missing disease.3,7
As health care costs continue to rise, it is essential to scrutinize not only the benefit of different approaches, but also their costs. Prior cost-effectiveness analyses comparing routine and selective lymph node dissection have had conflicting results; however, no studies to date have assessed the cost-effectiveness of a sentinel lymph node algorithm in comparison with other management strategies.8,9 Our objective was to evaluate the cost-utility of sentinel lymph node mapping, selective lymph node dissection, and routine lymph node dissection in the management of low-risk endometrial carcinoma.
MATERIALS AND METHODS
Low-risk endometrial carcinoma was defined as clinical stage 1 disease with grade 1–2 endometrioid histology on preoperative endometrial biopsy. A decision analysis model (Fig. 1) was constructed comparing three lymph node dissection strategies in women undergoing minimally invasive surgery for low-risk carcinoma: 1) routine lymph node dissection in all patients; 2) selective lymph node dissection based on intraoperative frozen section (Mayo Clinic criteria), in which approximately 60% of patients undergo lymph node dissection4,10; and 3) sentinel lymph node mapping based on the Memorial Sloan Kettering algorithm.7 The sentinel lymph node mapping algorithm includes the pathologic ultrastaging of sentinel lymph nodes and the performance of a side-specific pelvic lymph node dissection if no mapping occurs on a hemipelvis. Based on their most recently published data, 80% of patients map to both sides of the pelvis, 15% map unilaterally and require a contralateral side-specific lymph node dissection, and 5% do not map and require a bilateral lymph node dissection.7,11 All patients in the three arms underwent hysterectomy and bilateral salpingo-oophorectomy. Minimally invasive surgery was chosen for the base-case analysis because that is the most common modality used to treat endometrial carcinoma in the United States. The University of Texas MD Anderson Cancer Center institutional review board exempted this study from approval.
The model was constructed using data from the published literature and clinical estimates. Clinical outcomes used are shown in Table 1. The rate of lymphedema after full lymph node dissection for endometrial cancer varies from 3% to 38% in the literature; thus, 20% was chosen for the base-case scenario.3,12–14 There are no published data on the risk of lymphedema with sentinel lymph node mapping; however, studies have shown that patients who had less than five lymph nodes excised had a 0–9% risk of developing it.14–16 Therefore, 5% was chosen as a conservative estimate for the sentinel lymph node mapping arm. Because not all patients who undergo sentinel lymph node map bilaterally successfully, “weighted sentinel lymph node mapping” adjusts for the fact that 15% of patients require a unilateral lymph node dissection and 5% require a bilateral lymph node dissection. The rate of lymphedema was thus estimated to be slightly higher in the sentinel lymph node arm, and “weighted sentinel lymph node” was used for all our analysis. The utility of lymphedema was considered to be 0.8 based on a large study by Cheville et al17 assessing the effect of lower extremity lymphedema on quality of life (QOL). Lymphedema was chosen as the main factor affecting QOL because it is often cited as the main complication from full lymph node dissection and can have a debilitating effect on patients when significant.14,16 Published 3-year disease-specific survival and overall survival rates were obtained from the report by Zahl Eriksson et al,18 which was a nonrandomized study comparing patients who underwent sentinel lymph node mapping and selective lymph node dissection at two institutions. In that study, patients in the sentinel lymph node mapping arm had a significant increase in 3-year disease-specific survival (100% vs 98.8%); however, there was a nonsignificant increase in overall survival (97.4% vs 95.4%). Additional operating room time needed was estimated for frozen section, lymph node dissection, and sentinel lymph node mapping.
The analysis was performed from the perspective of a third-party payer. Year 2016 Medicare reimbursement rates were used to calculate estimates of costs. Cost categories consisted of hospital, physician, operating room, pathology, and lymphedema treatment (Table 2). The hospital cost of surgery was $11,697 based on diagnosis-related group code #740. Surgeon, anesthesia, and pathology fees were obtained from the Medicare Physician Fee Schedule using relevant Current Procedural Terminology codes.19,20 Pathology costs included the cost of ultrastaging and frozen section, and “weighted sentinel lymph node mappings” was used for both surgeon and pathology fees. Additional operating room time was estimated to cost $30 per minute.21 Lymphedema treatment costs were estimated at $2,500 per year based on a previous study.8 Other long-term costs were excluded because they were considered to be equivalent in the three arms. Costs were reported in U.S. dollars and were adjusted to 2016 currency values.
A cost-utility analysis was performed comparing the different strategies. A cost-utility analysis is a cost-effectiveness analysis that takes into account cost, survival, and QOL. Effectiveness was defined as 3-year disease-specific survival adjusted for the effect of lymphedema on QOL. Disease-specific survival was used because it was an outcome that was significantly different among lymph node dissection strategies in the analysis by Zahl Eriksson et al.18 The time horizon for our analysis was considered to be 3 years. This was chosen because we based our effectiveness outcome on Zahl Eriksson et al's study, which is the only one in the literature comparing survival between sentinel lymph node mapping and selective lymph node dissection and which published 3-year survival outcomes. We planned to calculate incremental cost-effectiveness ratios. Quality-adjusted life-years (QALYs) and costs were discounted at an annual 3% rate. Multiple deterministic sensitivity analyses were performed varying the surgical approach (open vs minimally invasive surgery), effectiveness outcome (overall survival vs disease-specific survival), the rates of lymphedema and lymph node dissection, lymphedema utility, the rate of successful sentinel lymph node mapping, and costs. Analysis was done using TreeAge Pro 2016.
In the base-case scenario, routine lymph node dissection had a cost of $18,041 per patient and an effectiveness of 2.79 QALYs. The cost was calculated as a summation of the following: hospital ($11,697), surgeon ($1,878), anesthesia ($592), pathology ($1,517), additional operating room time ($900), and lymphedema treatment ($1,457). Selective lymph node dissection had a cost of $17,036 per patient and an effectiveness of 2.81 QALYs, whereas sentinel lymph node mapping had a cost of $16,401 per patient and an effectiveness of 2.87 QALYs. When comparing selective with routine lymph node dissection, the selective lymph node dissection arm was both less costly ($1,005) and more effective (0.02 QALYs) than routine lymph node dissection, making it the dominant arm. However, when comparing all three strategies, the sentinel lymph node mapping arm had both the lowest cost and highest effectiveness, dominating the other modalities, and was thus the most cost-effective strategy (Table 3). No incremental cost-effectiveness ratio could be determined because the sentinel lymph node mapping arm was both less costly and more effective. For the estimated 40,000 women undergoing surgery for low-risk endometrial carcinoma each year in the United States, the annual cost would be $722 million, $681 million, and $656 million if all underwent routine lymph node dissection, selective lymph node dissection, and sentinel lymph node mapping, respectively (assuming all undergo minimally invasive surgery).
To assess the robustness of these results, we performed several sensitivity analyses adjusting different parameters. In the first sensitivity analysis, we assumed all patients underwent open surgery and used cost estimates for that as opposed to minimally invasive surgery costs (Table 4). In the second analysis, we changed the effectiveness outcome and looked at quality-adjusted 3-year overall survival instead of disease-specific survival (Table 5). Sentinel lymph node mapping remained the most cost-effective strategy in each case.
We also performed multiple other deterministic sensitivity analyses on selected variables in the model across the ranges specified in Tables 1 and 2. We varied the risk of lymphedema in different lymph node strategies, lymphedema utility, the rate of lymph node dissection in the selective lymph node dissection arm, successful sentinel lymph node mapping rates, survival estimates, additional operating room time needed, and costs. For example, in one analysis, we assumed that the risk of lymphedema was 10%, 0%, and 0% for patients with routine lymph node dissection, sentinel lymph node mapping, and no lymph node dissection, respectively, which affected the total lymphedema treatment costs in each arm. In this scenario, sentinel lymph node mapping was still the least costly at $15,895 per patient vs $16,599 per patient for selective lymph node dissection and $17,312 per patient for routine lymph node dissection. With sentinel lymph node mapping having the highest effectiveness, it remained the dominant strategy.
In another analysis, we changed several variables simultaneously: we assumed different lymphedema risks (30%, 10%, and 5% for routine lymph node dissection, sentinel lymph node mapping, and no lymph node dissection, respectively), increased the lymphedema utility to 0.95, estimated that 50% of women in the selective lymph node dissection arm undergo lymph node dissection, increased additional operating room time needed (60 minutes for lymph node dissection, 30 minutes for sentinel lymph node mapping, and 25 minutes for frozen section), and assumed equivalent 3-year disease-specific survival for all three arms (95%). Routine lymph node dissection thus had a cost of $19,699 per patient and an effectiveness of 2.72 QALYs. Selective lymph node dissection had a cost of $17,757 per patient and an effectiveness of 2.74 QALYs, whereas sentinel lymph node mapping had a cost of $16,998 per patient and an effectiveness of 2.75 QALYs. In this analysis and other ones using different estimates for various parameters, sentinel lymph node mapping remained the most cost-effective approach.
In this cost-utility analysis, selective lymph node dissection was more cost-effective than routine lymph node dissection in patients undergoing surgery for low-risk endometrial carcinoma. However, compared with routine and selective lymph node dissection, sentinel lymph node mapping had the lowest costs and highest quality-adjusted survival, making it the most cost-effective strategy.
There is debate regarding the optimal approach to lymph node dissection in endometrial cancer. Although routine lymph node dissection was traditionally considered standard of care, some investigators have proposed a selective lymph node dissection approach after identifying a cohort of patients for whom lymph node dissection could be omitted.2,22 Others have advocated a sentinel lymph node mapping strategy, arguing that it decreases the risk of understaging some patients with selective lymph node dissection while minimizing the risk of lymphedema with routine lymph node dissection.3 A prospective study has suggested that sentinel lymph node mapping has a high degree of diagnostic accuracy in detecting metastasis and can safely replace routine lymph node dissection.23 Zahl Eriksson et al concluded that either selective lymph node dissection or sentinel lymph node mapping was acceptable for staging with no apparent detriment to the use of a sentinel lymph node algorithm.18
There are several reasons why the sentinel lymph node mapping strategy was most cost-effective. First, when looking at effectiveness, sentinel lymph node mapping was associated with improved 3-year disease-specific survival when compared with selective lymph node dissection.18 When adjusting for the effect of lymphedema on QOL, the difference became even more pronounced, because sentinel lymph node mapping theoretically leads to less lymphedema than both selective and routine lymph node dissection.17,24 With regard to costs, routine lymph node dissection had the highest costs as a result of having the highest surgeon ($1,878), pathology ($1,517), and lymphedema treatment ($1,457) costs. Compared with selective lymph node dissection, sentinel lymph node mapping had slightly higher surgeon ($1,621 vs $1,493) and pathology ($1,340 vs $1,240) fees, but significantly less operating room time ($645 vs $1,140) and lymphedema treatment ($506 vs $874) costs, making it the least costly arm. The selective lymph node dissection arm had the longest operating room time given the need for frozen section on all cases and lymph node dissection on 60% of the patients. If the time horizon of our study was longer than 3 years, it is possible the cost differences would be even more pronounced when accounting for lymphedema treatment.
Our study has several limitations. Costs were based on Medicare reimbursement rates, which may or may not reflect commercial insurance reimbursement or the cost of delivering care from a hospital or health care provider perspective.19 Although published data were used for most parameters, clinical assumptions were made when data were unavailable. Indeed, our biggest assumption was that sentinel lymph node mapping is associated with a decreased risk of lymphedema compared with lymph node dissection. Although there are no data to support this, we based our assumption on several studies showing that having less than five lymph nodes excised was associated with a much smaller risk of developing lymphedema.13–16 Additionally, when looking at effectiveness, the QALY difference among the different lymph node dissection arms was small. This can be attributed to the fact that our time horizon was 3 years and that most patients with low-risk cancer tend to have good clinical outcomes. The 3-year limit was chosen because we based our effectiveness outcome on Zahl Eriksson et al's study, which was the only one in the literature comparing survival between sentinel lymph node mapping and selective lymph node dissection and which used 3-year survival outcomes.18 This can also be considered a limitation, because the conclusions regarding effectiveness were based on that study's reported survivals. Future studies that report survival may enhance or diminish our conclusions; therefore, they should be interpreted with caution until validated. Additionally, we did not account for the effect of adjuvant pelvic radiation therapy on costs or QOL.
The major strength of our study lies in the fact that we compared all three lymph node dissection strategies currently in practice in this population, and we assessed the cost-effectiveness of sentinel lymph node mapping in this setting. In addition, our effectiveness outcome was based on published survival data and quality-adjusted for the effect of lymphedema. Indeed, previous analyses comparing routine and selective lymph node dissection were cost-effectiveness analyses, whereas ours was a cost-utility study, which is a cost-effectiveness analysis that takes QOL into account.8,9 In addition, our findings were robust to multiple sensitivity analyses, which further support our results.
In conclusion, our study showed that sentinel lymph node mapping was the most cost-effective strategy in the upfront surgical management of low-risk endometrial carcinoma (clinical stage 1 disease with grade 1–2 endometrioid histology). This adds to the body of literature evaluating the clinical benefits of this strategy and may help health care providers in the decision-making process as they consider which approach to use. Continued research into the costs and effectiveness of different lymph node dissection strategies is needed.
1. Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100:1707–16.
2. ASTEC study group, Kitchener H, Swart AM, Qian Q, Amos C, Parmar MK. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet 2009;373:125–36.
3. Abu-Rustum NR. Sentinel lymph node mapping for endometrial cancer: a modern approach to surgical staging. J Natl Compr Canc Netw 2014;12:288–97.
4. Mariani A, Webb MJ, Keeney GL, Haddock MG, Calori G, Podratz KC. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506–19.
5. Dowdy SC, Borah BJ, Bakkum-Gamez JN, Weaver AL, McGree ME, Haas LR, et al. Prospective assessment of survival, morbidity, and cost associated with lymphadenectomy in low-risk endometrial cancer. Gynecol Oncol 2012;127:5–10.
6. Frumovitz M, Slomovitz BM, Singh DK, Broaddus RR, Abrams J, Sun CC, et al. Frozen section analyses as predictors of lymphatic spread in patients with early-stage uterine cancer. J Am Coll Surg 2004;199:388–93.
7. Barlin JN, Khoury-Collado F, Kim CH, Leitao MM Jr, Chi DS, Sonoda Y, et al. The importance of applying a sentinel lymph node mapping algorithm in endometrial cancer staging: beyond removal of blue nodes. Gynecol Oncol 2012;125:531–5.
8. Clements AE, Tierney BJ, Cohn DE, Straughn JM Jr. Is selective lymphadenectomy more cost-effective than routine lymphadenectomy in patients with endometrial cancer? Gynecol Oncol 2013;128:166–70.
9. Lee JY, Cohn DE, Kim Y, Lee TJ, Barnett JC, Kim JW, et al. The cost-effectiveness of selective lymphadenectomy based on a preoperative prediction model in patients with endometrial cancer: insights from the US and Korean healthcare systems. Gynecol Oncol 2014;135:518–24.
10. Milam MR, Java J, Walker JL, Metzinger DS, Parker LP, Coleman RL, et al. Nodal metastasis risk in endometrioid endometrial cancer. Obstet Gynecol 2012;119:286–92.
11. Jewell EL, Huang JJ, Abu-Rustum NR, Gardner GJ, Brown CL, Sonoda Y, et al. Detection of sentinel lymph nodes in minimally invasive surgery using indocyanine green and near-infrared fluorescence imaging for uterine and cervical malignancies. Gynecol Oncol 2014;133:274–7.
12. Todo Y, Yamamoto R, Minobe S, Suzuki Y, Takeshi U, Nakatani M, et al. Risk factors for postoperative lower-extremity lymphedema in endometrial cancer survivors who had treatment including lymphadenectomy. Gynecol Oncol 2010;119:60–4.
13. Yost KJ, Cheville AL, Al-Hilli MM, Mariani A, Barrette BA, McGree ME, et al. Lymphedema after surgery for endometrial cancer: prevalence, risk factors, and quality of life. Obstet Gynecol 2014;124:307–15.
14. Beesley VL, Rowlands IJ, Hayes SC, Janda M, O'Rourke P, Marquart L, et al. Incidence, risk factors and estimates of a woman's risk of developing secondary lower limb lymphedema and lymphedema-specific supportive care needs in women treated for endometrial cancer. Gynecol Oncol 2015;136:87–93.
15. Abu-Rustum NR, Alektiar K, Iasonos A, Lev G, Sonoda Y, Aghajanian C, et al. The incidence of symptomatic lower-extremity lymphedema following treatment of uterine corpus malignancies: a 12-year experience at Memorial Sloan-Kettering Cancer Center. Gynecol Oncol 2006;103:714–8.
16. Rowlands IJ, Beesley VL, Janda M, Hayes SC, Obermair A, Quinn MA, et al. Quality of life of women with lower limb swelling or lymphedema 3-5 years following endometrial cancer. Gynecol Oncol 2014;133:314–8.
17. Cheville AL, Almoza M, Courmier JN, Basford JR. A prospective cohort study defining utilities using time trade-offs and the Euroqol-5D to assess the impact of cancer-related lymphedema. Cancer 2010;116:3722–31.
18. Zahl Eriksson AG, Ducie JA, Ali N, McGree M, Weaver AL, Bogani G, et al. Comparison of a sentinel lymph node and a selective lymphadenectomy algorithm in patients with endometrioid endometrial carcinoma and limited myometrial invasion. Gynecol Oncol 2016;140:394–9.
22. Mariani A, Dowdy SC, Wa Cliby, Gostout BS, Jones MB, Wilson TO, et al. Prospective assessment of lymphatic dissemination in endometrial cancer: a paradigm shift in surgical staging. Gynecol Oncol 2008;109:11–8.
23. Rossi EC, Kowalski LD, Scalici J, Cantrell L, Schuler K, Hanna RK, et al. A comparison of sentinel lymph node biopsy to lymphadenectomy for endometrial cancer staging (FIRES trial): a multicentre, prospective, cohort study. Lancet Oncol 2017;18:384–92.
© 2018 by the American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
24. Abu-Rustum NR. The increasing credibility of sentinel lymph node mapping in endometrial cancer. Ann Surg Oncol 2013;20:353–4.