Cost-effectiveness of Iterative Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy for the Treatment of Peritoneal Carcinomatosis

Discussion

CRS with HIPEC has become the standard of treatment worldwide for disseminated peritoneal malignancies, including DPAM, cancer of the appendix, and benign mesothelioma. There is also an increasing body of evidence that supports the use of iCRS in selected cases of disseminated colorectal, ovarian, and gastric cancer (10, 13, 16). As previously described, there is a high rate of recurrence of disseminated disease that is typically dependent on the nature and extent of the primary malignancy. As such, the use of iCRS for this indication by multiple centres has resulted in promising outcomes. We and others have described a wide range of overall median survival rates, from 15 to 90 months, following iCRS (11-13, 17). Similar to outcomes following pCRS, outcomes associated with iCRS are highly dependent on the nature of the primary malignancy. The highest overall survival rates following iCRS were those reported for patients diagnosed with DPAM, followed by those with cancer of the appendix and mesothelioma at 58.1 (95% CI=53.6-62.5) months, 48 (95% CI=38.4-57.5) months, and 36 (95% CI=32.7-39.2) months, respectively. The lowest overall survival rates were observed for those diagnosed with CRC or ovarian cancer, at 30.5 (95% CI=21.5-39.5) months and 28.5 (95% CI=18.7-38.3) months, respectively (Table II). The results of a recent systematic review indicated that median survival following iCRS for CRC ranged from 18 to 55.7 months, which are findings that are consistent with those reported by our center (18). There are very few reports that focus on survival after iCRS for malignancies other than DPAM, cancer of the appendix, or CRC. There were no significant differences between Eastern Cooperative Oncology Group and American Society of Anesthesiologists status of patients who were treated with pCRS or iCRS at our centre, nor were there any demographic differences noted between the two groups (Table I), further research will be needed to determine which factors are important when selecting patients who are most likely to respond effectively to iCRS. Of note, our findings included a significant difference in the initial PCI scores among those in the iCRS (PCI score of 6.5) group compared to those in the pCRS group (PCI score of 15.0; p<0.001). This finding may be explained by patient selection or because patients who had undergone a primary round of treatment received close follow-up and were most likely to experience early detection and treatment of any recurrence of disease. One criticism of iCRS is that it is associated with greater morbidity than would be anticipated in response to alternative treatment modalities, such as chemotherapy. Indeed, the results of this study revealed that 39.6% of patients undergoing iCRS developed grade III or IV complications that were life-threatening or required intervention although there was no difference in the morbidity grade between groups. It is difficult to take into account the precise role played by morbidity with respect to the overall therapeutic experience, although it is reassuring to note that ~79% of patients treated with iCRS reported that they would undergo additional cytoreductive procedures if necessary (19).

The breakdown is similar when comparing costs associated with pCRS and iCRS. The costs associated with the use of the operating theatre represent 25% of the overall total, followed by the costs associated with critical care at 20% (Figure 3). The fractions represented by these sub-costs are comparable to those reported for pCRS at our Institution in 2010 (14), at 22% and 26% of the total costs, respectively. Our finding that there were no significant differences in overall costs associated with the first vs. repeat cytoreduction procedures, and likewise our identification of similar cost breakdowns for these two procedures suggests that the patient experiences are also similar from an institutional perspective. However, the median LOS associated with iCRS at 21.2±16.3 days was significantly shorter than that after pCRS (25.51±21.11 days) (p=0.034). Patient selection as well as prior knowledge and expectations of the patients in this latter group may have contributed to the reduced LOS. It is important to recognise that this reflects a key component of the Enhanced Recovery After Surgery protocol, specifically, that perioperative counselling contributes to reduced LOS when compared to cases in which patients were not provided with similar interventions (20). Furthermore, the average PCI among those in the iCRS group was 6, which was significantly lower than the average PCI of 15 observed among those undergoing pCRS (p<0.001). The lower PCI value might be attributed to the fact that one cannot repeat some of the associated procedures, for example, splenectomy; this may have contributed to the reduced LOS associated with the secondary iCRS procedures.

The pCRS procedure is a cost-effective treatment when compared to non-surgical options for the treatment of some malignancies (21). To the best of our knowledge, this is the first study that has evaluated the cost-effectiveness of iCRS. Notably, we observed no significant differences between the average total costs associated with pCRS ($70,794) and iCRS ($68,688; p=0.486). In our evaluation of overall survival following pCRS and iCRS, we found that the costs per LY were $16,060 and $16,420, respectively. Furthermore, when compared to other standard treatments for disseminated malignancy, including chemotherapy, iCRS is cost-effective. In Australia, the treatment of stage IV CRC with standard chemotherapy (folinic acid/fluorouracil/irinotecan) with bevacizumab has an estimated total cost of $81,403 (22) and results in a median overall survival of 25.0 months (23). As such, a chemotherapeutic regimen based on these estimates for stage IV CRC results in a net cost of $39,073 per LY. Similar estimates can be made for advanced ovarian cancer and lung cancer (regardless of stage), in which standard chemotherapeutic regimens have been associated with net costs of $24,453 and $15,202 per LY, respectively (24).

Our study has several limitations. Our findings were based on patient data from a single institution in Australia and thus relies on the assumption that the overall costs associated with care at that institution are comparable to those found at similar centres both nationally and internationally. In Australia, the major costs of treatment for intraperitoneal malignancy, including the hospital stay and operating costs, are largely carried by the public sector. However, there are specific out-of-pocket costs, such as those that cover specialist appointments and out-patient medications, that are borne by the patient and thus not included in the analysis. Furthermore, the comparisons of costs associated with each treatment modality focused solely on cost per LY and did not account for morbidity following the procedure. Assessment of morbidity would require an objective, comparative evaluation of post-operative factors such as disability-adjusted LYs.

Given the substantial institutional support required, cytoreduction is typically performed at large centres. In New South Wales, Australia, there are two such centres which collectively serve more than 8 million people. This places significant economic burdens on these two centres, given the high upfront costs of CRS, the need for a surgical procedure and thus a prolonged hospital stay. This differs somewhat from the more distributed costs associated with non-surgical options, which typically focus on medications and involve shorter LOSs distributed over longer periods. Given the upfront cost burden placed on these specialised centres, it is crucial to collect data focused on iCRS as medically worthwhile and economically feasible in the long run.

To this end, we have shown here that iCRS is a cost-effective modality when used for the treatment of PC in selected patients.