Cost-effectiveness analysis of long-course oxaliplatin and bolus of fluorouracil based preoperative chemoradiotherapy vs. 5x5Gy radiation plus FOLFOX4 for locally advanced resectable rectal cancer

Rectal cancer (RC) is one of the most frequent malignancies in the world and represents a major socioeconomic and health issue [1]. In China, colorectal cancer (CRC) has attracted increasing attention over recent years, taking a second and fourth position in the incidence and mortality respectively among all malignant tumors in urban populations [2]. RC accounts for about 40% in the morbidity of CRC [2].

Surgery is the basic treatment for RC, but for advanced RC, adjuvant radiotherapy with or without chemotherapy has been used widely to improve outcomes. It is well known for locally advanced disease, postoperative chemoradiotherapy (CRT) significantly improves both local control and overall survival as compared with surgery alone or surgery plus irradiation [3, 4]. There are fewer studies on the preoperative treatments. Preoperative short-course radiotherapy (SCRT) plus chemotherapy consisting of FOLFOX4 or preoperative long-course chemoradiotherapy (LCCRT) with oxaliplatin and boluses of 5- fluorouracil and leucovorin in combination with conventional surgery are recommended depending on the tumor location, infiltration depth of the tumor, and lymph node involvement, which improves local control and survival [5]. Early results showed short-term preoperative radiotherapy decreased risk of local recurrence for irradiated patients at 2 years (2% vs 8%, p < 0.001) without a difference in overall survival (OS) for the patients with rectal cancer who undergo a standardized total mesorectal excision [6]. After a median follow-up of 6 years, the effect of radiotherapy on local recurrence persisted (6% vs 11%, p < 0.001), as well as the absence of a survival benefit for the patients with mobile rectal cancer undergoing surgery treated with preoperative short-term radiotherapy [7]. There is no international consensus on the use of these treatment schedules or the most appropriate patient selection for these schedules.

The phase III trial NCT00833131 comparing SCRT (5 × 5Gy) consisting of FOLFOX4 versus LCCRT (1.8 × 28Gy) in patients with fixed cT3 or cT4 rectal cancer with oxaliplatin and boluses of 5- fluorouracil and leucovorin, revealed that preoperative LCCRT, aimed at tumor shrinkage, is used to achieve R0 resection [8]. Nevertheless, improved OS and lower acute toxicity favours the 5 × 5Gy schedule with consolidation chemotherapy, with statistically-significant improvements found in disease-free survival (DFS) and OS with SCRT (p = 0.046) [8].

Several studies (e.g. CAO/ARO-094) have proved that preoperative radiotherapy has advantages in promoting the local control rates (LCR) and reducing toxicity compared with other treatment methods [9‐13]. So far, no economic evaluation is available that compares preoperative SCRT with consolidation chemotherapy versus preoperative LCCRT. It is reported that SCRT with consolidation chemotherapy could be considered as an effective option for preoperative management in advanced rectal cancer, especially in countries with low health care budgets or long waiting lists for radiotherapy [8]. Given the increasing emphasis on value-based care in oncology, the aim of this study was to compare SCRT plus chemotherapy and LCCRT for the management of cT4 or advanced cT3 RC using a decision analysis approach.

There have been plenty of studies published on the preoperational CRT regimens for RC which have been widely used in clinical practice. However, the research on economic analysis was rarely reported focusing the cost-effectiveness comparison. This is the first study to specifically investigate the cost-effectiveness between the preoperational SCRT plus chemotherapy and LCCRT for advanced RC patients.

One of the strengths of this study is that we included that preoperative SCRT plus chemotherapy consisting 3 cycles of FOLFOX4 may be a more cost-effective for locally advanced resectable RC, especially for the patients in the DFS state. The SCRT plus chemotherapy had more total cost than LCCRT with 5-fluorouracil and leucovorin, which maybe resulted from the long survival in PD state after the treatment of DFS state and the high cost of targeted therapies in PD state. But for the patients in DFS, the whole cost for SCRT plus chemotherapy and LCCRT were $11,490 and $10,794 with an effectiveness of 21.70 QALMs and 19.65 QALMs, respectively. In other words, SCRT plus chemotherapy increased cost and QALM by $695.97 and 2.05 relative to LCCRT, resulting in a ICER of $339.50/QALM gained, which below the WTP. Besides, our result was markedly consistent on deterministic and probabilistic sensitivity analyses, strengthening the argument that this novel technology is actually cost-effectiveness. The challenges facing healthcare worldwide is the incremental cost-effectiveness and the threshold for using or rejecting specific drugs. Based on the trial, SCRT with consolidation chemotherapy can be considered as an effective option for preoperative management in advanced RC, especially in countries with low health care budgets or long waiting lists for radiotherapy. Similar results were concluded by Mandy Van Den et al that short-term preoperative radiotherapy in patients with rectal cancer undergoing total mesorectal excision (TME) is both effective and cost-effective based on the trial conducted by the Dutch Colorectal Cancer Group [18]. And Dahlberg M et al found 5 × 5 Gy preoperative radiotherapy was in the range of other well-accepted medical interventions with reduced local recurrence rates and improved overall survival in a Swedish Rectal Cancer Trial [19].

On sensitivity analysis, the utility of DFS state was the most influential factor for the robustness of the model, followed by transition probabilities in SCRT plus chemotherapy and LCCRT. Especially, the costs in perioperative period had the greater impact on the results than the radiotherapy only. Furthermore, a key factor distinguishing between the two groups was the time arrangement for radiotherapy and chemotherapy, except for the surgery. Since SCRT plus chemotherapy costs much more than LCCRT ($78,937 vs. $38,140), cost for DFS state was higher in SCRT plus chemotherapy than in LCCRT, while cost for PD state was distinctly higher in SCRT plus chemotherapy than in LCCRT. There more chemotherapy drugs were given by infusion in LCCRT, which affected the stay-days in hospital. The costs of supportive drugs also showed a profound impact on the analysis model, which were related to patients’ length of stay. The longer the hospital stay, the more supportive drugs were administrated. It was suggested that the hospitals losses could be reduced by shortening the length of patients’ hospitalization. Dramatically, QALMs in the PD state for SCRT plus chemotherapy was distinctly higher than the data in LCCRT (8.22 vs. 3.33). There may be a balance between the cost and effectiveness after surgery. The results were consistent with Nishimura et al. [20], which suggested that the hospitals losses could be reduced by shortening the length of patients’ hospitalization, furthermore, increasing the dose per fraction is a good choice for the patient with RC.

Several limitations may not been ignored in this study. Principally, this study was based on a published clinical trial, rather than a randomized prospective research. It reflects the actual resource use of the subjects’ care but, of course, is subject to bias. Furthermore, the information about the PD state in the NCT00833131 trial was limited, the costs in this setting were based on the cost-effectiveness analysis from Chinese FIRE-3 study, which has been analyzed by our group previously. Data from many different sources could be reasonably combined into a single model to yield useful results, but this may have determined multiple biases in the analysis. For example, the adverse events maybe changed because of the different nationality in NCT00833131 trial and FIRE-3(Polish vs. Germany and Austria), leading to little cost discrepancy in the PD state. It’s worth noting that treatments to PD state is palliative care for advanced rectal cancer, so when the patients in the NCT00833131 trial translated to PD state, the strategies could be the same as patients in FIRE-3. At the same time, cost-effectiveness analysis for FIRE-3 study was from Chinese perspective, which was the same in the current study, so there was no nationality difference. Undeniably, although we assessed the economic costs in this study, it was reminded that some other factors biasing the analysis result would influence the choice of fit treatment protocol for the patients with RC, containing the time away from home, lost productivity costs, education, religion. Meanwhile, as Dahlberg et al concluded that the long-term side effects of radiotherapy and chemotherapy before surgery should be of future prime concern [19]. Based on these limitations, we still believe that our study will encourage decision makers to take a more comprehensive view of treatment- related costs.

Compared to LCCRT, SCRT plus chemotherapy was more cost-effective for locally advanced resectable RC, despite the higher cost with SCRT plus chemotherapy has been shown in this study. Our analysis sharply suggested that it should be preferred to hypofractional radiotherapy in a cost-conscious environment for the patients in the DFS state as well as in the all states when the cost of PD state below $1920.