Fitting-Room Approach for Diabetes Meds: Try It First

Akshay B. Jain, MD


January 30, 2023

As we start 2023, perhaps it might be time to rethink several "conventional" methods we have been using to manage diabetes and see if it's possible to change the age-old paradigms.

Picture this: A person with diabetes who is not meeting glycemic targets comes to your clinic. You discuss several options for therapy escalation. On the basis of the patient's preference, your clinical judgment, available access, and ruling out contraindications, you narrow it down to two to three options. Then you decide to pick one of the surviving contenders on the list. Doing so allows the patient to only fill the prescription of that one medication. Subsequently, only if efficacy targets are not met at the next appointment a few months later, or if side effects occur, do we move on to the next medication on the list. That's pretty much how most of us have been managing diabetes across the world.

Now let's visualize a parallel metaverse: What if there were a "fitting room" approach whereby patients could try each of the two to three medications on the narrowed-down list for some time and then see which one suits them the best? They will be able to see which drug best helps achieve the desired glycemic targets with the fewest side effects, suits their lifestyle, and falls within their budget. They will then fill the prescription of only that "best fitting" medication and continue on it for an ongoing basis.

Sound too far-fetched? What if I tell you that we now have evidence to back this new approach?

Trying Different Drugs 'On'

Published recently, the TriMaster study, by Shields and colleagues, is said to be the first where patients on metformin ± a sulfonylurea with uncontrolled diabetes were invited to try three different drugs in succession, in a randomized, double-blinded fashion.

Each drug — pioglitazone 30 mg (thiazolidinedione), sitagliptin 100 mg (DPP-4 inhibitor), and canagliflozin 100 mg (SGLT2 inhibitor) — was trialed for 16 weeks followed by a switch to the next drug in a predetermined sequence, until the individual had tried all three medications. By using this three-way crossover method, researchers in this trial were able to power the study to answer the two primary hypotheses with only 525 patients as opposed to the over 4500 patients required for a parallel-arm randomized controlled trial.

The study found that compared with sitagliptin, pioglitazone reduced glucose more effectively among patients with body mass index (BMI) > 30 while sitagliptin fared better in those with BMI < 30. However, pioglitazone led to significant weight gain, so the improved efficacy came at a high cost to individuals with preexisting obesity or overweight. For its second primary objective, the study found that canagliflozin reduced glucose more effectively among patients with eGFR > 90 while sitagliptin fared better in those with eGFR 60-90. Overall, the three drugs fared similarly in their ability to reduce A1c.

Particularly important for this discussion is the consideration of prespecified outcome of this study; patients were asked to state which drug they preferred the most (without them knowing their A1c reduction). It's not surprising that patients chose the drug with the fewest side effects or on which they felt better .Of note, this was also the medication with the most glycemic-lowering efficacy for each of them.

Each of the three drugs was picked by different patients as their preferred option based on how they felt while taking the medication. About 38% chose canagliflozin, 35% chose sitagliptin, and about 25% chose pioglitazone as their preferred drug after trying out all three. Allocating therapy on the basis of individually preferred drugs (rather than allocating all patients the overall most preferred drug in the study — canagliflozin) was shown to result in more patients achieving the lowest A1c for them (70% vs 30%) along with the fewest side effects (50% vs 67%).

Similar Studies on Patient Choice

Different versions of this approach have been attempted previously with a good measure of success. In Ontario, Canada, patients with hypertension filling prescriptions for new antihypertension medications (calcium channel blocker, beta-blocker, or ACE inhibitor) were first given only a 7-day supply of the medication. A pharmacist would follow up with the patient before the trial period ended, and if the individual was tolerating the medication well, the remainder of the prescription would then be filled. Not only did a majority (82%) of the patients find this approach helpful, but there was also considerable direct savings of healthcare resources by reducing drug wastage.

In diabetes, there have been previous trials that performed a cross-over trial to assess patient preference, but these were limited because they were conducted as open-label studies (Meguro et al; Lüdemann et al). The authors of the TriMaster study report that theirs is the first study done in a double-blind manner, making it unique.

Opinions About the TriMaster Study

My personal take is that even though we don't use thiazolidinediones too often in North America, this study opens the door to precision medicine, unlike anything we've seen before. A three-way crossover trial is an elegant way of assessing response to different classes of medications to understand which patients benefit the most from a medication class.

The TriMaster study is limited because 95% of patients identified as White, and researchers looked at BMI and kidney function. In the future, my hope is that precision medicine studies will look at different ethnicities along with various age groups.

Patient preference is a major factor in precision medicine, and this was essentially driven by A1c lowering and lack of side effects. Coming specifically to the options addressed in TriMaster, I feel that although pioglitazone was noted to lower A1c better in those with a higher BMI, this medication paradoxically leads to more weight gain and may not be the best overall option.

I would have loved to see a comparison between SGLT2 inhibitor and GLP-1RA agents, the two classes of medications most recommended by guidelines across the world. Similarly, although SGLT2 inhibitors did not show better A1c lowering at eGFR levels in the 60-90 range in this study, one should note that these agents are still very crucial in chronic kidney disease because of the immense nephroprotection they offer.

Overall, the TriMaster study shows that precision medicine randomized controlled trials can offer exciting insights into initiation of pharmacotherapy.

Incorporating Into Practice

Perhaps when we see patients requiring the addition of a new medication, it would be best to give them samples of two to three appropriate medications for an adequate duration, have them try out each of these, and they would then fill the prescription of the one they felt was the "best fit."

Doing this will help reduce the risk for side effects, improve drug adherence, and also improve glycemic control. Of course, there are some limitations to this approach. An adequate supply of samples may not always be available for every patient. If utilizing a medication that requires uptitration, such as GLP-1 therapy, the trial period might be longer as it will need to factor in the time it takes to go up to the full intended dose, as per monograph instructions. Also, if doing a trial of many medications, patients might forget exactly how good or bad the first one was if the trial lasted a long time. In such cases, keeping a logbook of symptoms/sugar levels to compare would be helpful

I would love to hear your thoughts about how you choose a medication class after your patients with type 2 diabetes do not respond to metformin ± sulfonylurea therapy. Would you consider using the approach suggested by the "Try" (Tri) Master study?

Akshay B. Jain, MD, is a clinical endocrinologist who has practiced in three countries, focusing on mitigating the complications of diabetes and obesity. He is fluent in six languages and has spoken at more than 500 programs internationally.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.