Cost-Utility of PET/CT Surveillance Post Achieving First Remission in Patients with Diffuse Large B-Cell Lymphoma: A Systematic Review

Diffuse Large B-Cell Lymphoma (DLBCL) is one of the most common subtypes of Non-Hodgkin Lymphomas comprising 30-58% of all cases [1,2]. Family history, autoimmune disease, HIV infection are some of the risk factors identified with DLBCL. The treatment regimen of DLBCL, although varies with age, constitutes of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) therapy, along with the recent addition of rituximab (R-CHOP therapy) [1].


Introduction
Diffuse Large B-Cell Lymphoma (DLBCL) is one of the most common subtypes of Non-Hodgkin Lymphomas comprising 30-58% of all cases [1,2]. Family history, autoimmune disease, HIV infection are some of the risk factors identified with DLBCL. The treatment regimen of DLBCL, although varies with age, constitutes of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) therapy, along with the recent addition of rituximab (R-CHOP therapy) [1].
Previously, post achieving clinical remission, the guidelines used to recommend a computed tomography (CT) scans every 3-6 months for 5 years and annually thereafter with at 6-month intervals up to 2 years [1,3]. Later, the addition of Positron Emission Tomography (PET) to CT, led to the improvement in sensitivity of imaging surveillance, providing advantages in the detection of subclinical relapse. In a study comprising a cohort of 183 patients, the use of PET/CT led to an earlier diagnosis of relapse, suggesting that PET/CT may be a valid tool for the routine follow-up of lymphoma patients [4]. However, considering an increase in the cancer care cost nearly by ~27% by 2020, probably due to the expensive standard of care including diagnostic or imaging modalities, the cost-effectiveness of imaging surveillance needs to be carefully evaluated [5]. The present systematic review was therefore conducted to see if routine surveillance imaging with PET/CT in patients achieving complete remission following primary treatment for DLBCL offers any benefits in terms of costs and utilities along with an analysis of its effects on clinical outcomes. To our knowledge, this is the first of its kind review that systematically collates the currently available data highlighting the utility of conducting the PET/CT surveillance post achieving the first remission in patients with DLBCL.

Data sources and selection criteria
This systematic review was conducted as per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The following electronic databases were searched: PubMed/Medline and Google Scholar in June 2018. Two independent reviewers (MJ and PG) conducted an abstract review of all records. The Medline strategy is mentioned in Table 1.
Inclusion criteria were as follows: studies that evaluated the costs involved due to PET/CT in patients with DLBCL were included. Exclusion criteria were: experimental studies; mechanistic (association) studies; Letters/reviews/editorials, commentary, animal studies, in-vitro studies, case series (sample size<10 patients), case reports, pharmacodynamic/pharmacokinetic studies and studies with full-text published in a language other than English.

Summary of included studies
As shown in Figure 1 (PRISMA flow chart), the database searches initially yielded 12 results (seven from Medline search and five from a free hand Google search). After screening, 12 of these (review (2), a case report with different study design (1), and retrospective analysis with different study design (2) were excluded. Three studies were included for the systematic review.

Detection of relapses
Surveillance with PET/CT or CT led to the detection of asymptomatic relapses, however, the difference in detecting relapses was non-significant irrespective of surveillance technique used (PET/ CT or CT alone). Additionally, there were 2.9% and 1.5% incidences of false-positive imaging in the detection of relapses with PET/CT and CT, respectively. Further, the positive predictive value of PET/CT was calculated as 0.52. (Table 2).

PET/CT: costs and utilities involved
PET/CT surveillance had an insignificant benefit over no surveillance in terms of medical costs, quality-adjusted life years (QALYs) gained and led to high ICERs per QALY gained. PET/CT surveillance led to increased hospital visits, leading to increased costs owing to follow-up visits (Table 3). An incremental effectiveness of 0.025 was observed for performing surveillance with PET/CT. Furthermore, assuming the cost of performing the PET/CT to be as $1900, it was observed that avoiding PET/CT in asymptomatic patients achieving complete remission after first treatment would lead to a cost reduction of 27% or $102,600.

Results and Discussion
Based on the evidence of interim PET/CT scan in patients with Hodgkin lymphoma, and a few small studies wherein the PET/CT scan results correlated well with outcome in both HL and DLBCL, the predictive ability of PET/CT scan was given importance to evaluate the course of disease in patients with DLBCL, and based on the probability of relapse, to predict if the standard treatment could be altered prior for better prognosis. However, PET/CT surveillance is most commonly done after treatment completion in order to evaluate if complete remission has been achieved and to see if further therapy is required [8,9]. In routine practice, it was observed that more than 95% of patients treated at academic centres in the United States underwent at least one PET/CT scan during follow-up and about 50% of patients received at least one PET/CT during surveillance, despite guidelines suggesting otherwise in asymptomatic patients [10,11]. The present NCCN guideline for patients with DLBCL recommends the surveillance with PET/CT to be performed 8 weeks after treatment initiation to check the course of ongoing chemotherapy (NCCN guidelines 2018). Having said this, not much data is available that could suggest if interim surveillance or surveillance at the end of the treatment regimen using PET/CT could prevent (by altering the treatment regimen) or detect early relapses, and thus provides an advantage in terms of decreased costs in patients with DLBCL.
The results from the present systematic review were based on three retrospective studies across three very different geographies i.e. The US, South Korean and Germany.
With a very small clinical utility of 0.020-0.025 QALYs gained and similar relapse-free and overall survival (OS) with PET/CT surveillance, it can be suggested that PET/CT surveillance in patients with DLBCL in first complete remission does not provide any additional benefit in terms of clinical or economic outcomes. On the contrary, considering the high cure rates with DLBCL, there is a concern that routine surveillance with PET/CT, which is a common practice, may lead to an increase in treatment costs, evident with the results of this systematic review [12,13]. This boils down to the fact, that interim surveillance could be a better option wherein if the probability of relapse is observed, necessary measures in terms of changing the course of therapy can be taken upfront for a better prognosis. However, since the positive predictive values for the detection of relapse was estimated to as low   as 0.52, this will lead to patients undergoing further surveillance with PET/CT to confirm the results, further leading to increased costs [14].

Conclusion
As of now, the available data suggest no clinical utility of PET/ CT surveillance in patients with DLBCL achieving the first remission; hence the early detection of any sign and symptoms needs to be carefully observed. Probably PET/CT surveillance in asymptomatic patients may provide benefit, however considering the high probability of false positive, a willingness to pay approach could be an ideal way for performing PET/CT surveillance.
In the absence of data from randomized clinical trials, or for that matter any prospective economic study that could evaluate the effectiveness of PET/CT surveillance, the results of the present review suggest that the routine surveillance should be reduced in the interest of the patient and reimbursing agencies. The physician needs to ensure that surveillance should only be conducted when there is a high level of evidence that this would lead to a better clinical outcome or could help in better prognosis by changing the course of therapy.