Gabriel Brooks, MD, MPH, Dartmouth‑Hitchcock Medical Center (Lebanon, NH), discusses results from a study evaluating the cost-effectiveness of DPYD genotyping to screen for dihydropyrimidine dehydrogenase (DPD) prior to adjuvant chemotherapy for colon cancer in the US.
These results were presented at the virtual 2021 ASCO Gastrointestinal Cancers Symposium.
Hi, my name is Gabe Brooks. I'm a medical oncologist at Dartmouth‑Hitchcock Medical Center and the Norris Cotton Cancer Center. I'm here to discuss the abstract that me and my team submitted and presented at the most recent ASCO GI Cancer Symposium.
In our abstract, it was about the cost‑effectiveness of DPYD genotyping to screen for dihydropyrimidine dehydrogenase or DPD deficiency prior to adjuvant chemotherapy for colon cancer. The background of this topic is that 5‑fluorouracil and capecitabine, which are the 2 fluoropyrimidine chemotherapies, are essential for the treatment of many cancers.
These drugs are well tolerated by most patients. However, severe and fatal toxicities can happen, even though they're uncommon. Those toxicities, when they happen, most commonly occur in patients who have deficiency of DPD. Those cases of DPD deficiency are most often caused by rare variants of the DPYD gene.
These include the well‑known *2A, *13, and 2846AT variants. Pretreatment screening for those variants has been shown in multiple prospective studies to be 1 way to reduce severe and even fatal toxicities related to DPD deficiency and deleterious DPYD polymorphisms.
Additionally, the question has come up about whether this is a cost‑effective practice. Multiple studies in European settings have suggested that this practice is cost‑effective or possibly cost‑saving in some populations because, by doing screening for DPD deficiency, we can prevent toxicities and prevent hospitalizations related to those toxicities.
However, those analyses have not been done in the United States setting. There has not been a good analysis to show if DPYD testing is cost‑effective in the United States setting where we have perhaps a lower prevalence of these DPYD variants than some European countries.
Our research involves construction of a Markov model, the cost‑effectiveness analysis model from where we took the US health care perspective. Then, we simulated the clinical decision of whether or not to use DPYD genotyping prior to adjuvant chemotherapy for Stage 3 colon cancer.
This is a situation where chemotherapy has been shown to reduce recurrence and improve survival. However, where there remains a risk from chemotherapy because these are patients who may be cured even without chemotherapy.
In our model, we used a 6‑month cycle length and a 5‑year horizon, and the future costs are discounted to 3% per year. We modeled the DPYD variant carriers to receive dose‑reduced chemotherapy under a DPYD‑guided genotyping dose adjustment strategy.
The primary outcome of our study was the incremental cost‑effectiveness ratio of pretreatment DPYD genotyping compared to no testing. In our model, patients with Stage 3 colon cancer either did or did not go on to have DPYD genotyping.
If they did have DPYD genotyping and if they were found to have a variant of the reduced gene function or enzyme function, then those patients would get a dose reduction as has been recommended in the prospective studies.
In the other arm where there was no screening of these patients received standard doses of chemotherapy, we then modeled their toxicities during treatment, including the risk of treatment and related death. We also modeled their post‑treatment survival.
1 of the most important things that we thought that our analysis apart is that we used estimates in our model that are all based on the United States data. Specifically, we used a probability of carrying the deleterious DPYD variant that comes from 1 of the United States cooperative group studies where there was a 2.2% prevalence of carriers of these deleterious DPYD gene variants.
We also used the United States‑based cost estimates for the DPYD genotyping including a cost of $174 for the test. That cost is based on the clinical laboratory fee schedule from the Centers for Medicare & Medicaid Services. That's the allowable amount that Medicare lists for this test specifically.
The results of our study were that compared with no screening for DPD deficiency, DPYD genotyping led to an incremental cost of $106 per patient, and you'll notice that that's less than $174 per test cost. That's because there were savings relative to the cost of the test. There was some savings with reduced hospitalizations and other toxicity with associated costs.
If this test were free, it would be cost‑saving. Because it's not free, the incremental cost is less than the total cost of the test. The incremental benefit was small. It was 0.0028 quality‑adjusted life‑years. However, that calculates to an incremental cost‑effectiveness of $37,300 per quality‑adjusted life‑year.
This is, we note, less than the commonly accepted threshold of $100,000 per quality‑adjusted life‑year suggesting that this intervention of DPYD genotyping is cost‑effective as an approach for preventing toxicity from DPD deficiency.
We did do sensitivity analyses to evaluate the robustness of our findings. We found that the incremental cost‑effectiveness ratio was greater than $100,000 per quality‑adjusted life‑year when the DPY gene variant carrier frequency was less than 1.2%. Unsurprisingly, the carrier frequency is an important variable in our model.
However, the lower limit of 1.2% is certainly lower than what we would expect to see in the US population. The incremental cost‑effectiveness ratio also exceeded $100,000 when the specificity of DPY genotyping was low.
The conclusion and takeaway of the study is that patients receiving adjuvant chemotherapy for Stage 3 colon cancer did show cost‑effectiveness of the intervention. The incremental costs were low.
The average benefit of screening, while small, is consistent with a small number of patients experiencing a large benefit. We recommend that based on these findings, it's reasonable and cost‑effective to implement this approach to DPD deficiency screening. Thank you