How FAME data may assist in DME treatment decisions

An analysis of data from the FAME trials was undertaken to try to understand the final three-year visual and anatomic outcomes of patients treated with the fluocinolone acetonide intravitreal implant (Iluvien, Alimera Sciences) who exhibited suboptimal early responses, as measured at six weeks after the initial injection.

A unique characteristic about patients in the FAME trials is that they were able to receive additional therapies throughout and still be included in the final data analysis, said Daniel F. Kiernan, MD, of Ophthalmic Consultants of Long Island, New York.

Therefore, the data may reflect more of a real-life experience in situations such as deciding when to re-treat after seeing persistent fluid or a decrease in vision. This analysis may assist in treatment decisions, if positive conversion is seen over time.

Analysis
For this study, two groups of patients were identified from the FAME study data as having suboptimal response at six weeks. These patients had received the 2.0 microgram implant. The first group was chosen based on anatomic response, including patients whose reduction in central subfield thickness was less than 50 microns at six weeks, and the second group was chosen based on visual response, including those with an improvement in best-corrected visual acuity (BCVA) of less than 5 letters at six weeks.

About a third of the patients in the group receiving the FA implant had a suboptimal anatomic response observed at week six. The baseline data for responders and suboptimal responders were about the same, there was no statistically significant difference between baseline visual acuity (responders = 51.8 BCVA, suboptimal responders = 55.7 BCVA), but the suboptimal responders did have a thickness (395.4 µm) that that was a little bit less than the responders (506.6 µm).

In spite of suboptimal anatomic change at week six, 50 percent of patients went on to improve one or more lines at 36 months and 25 percent remained stable. Only about 22 percent lost vision.

These visual outcomes did not appear to be driven by additional treatments (intravitreal injections or laser), as there is little difference in the number of additional treatments in the two groups. In fact, more additional treatments were administered to the responder group than the suboptimal group.

Examining the group of suboptimal responders based on visual acuity outcomes further shows that those with the 36-month visual acuity gain (about 2/3 of these patients) also show moderate OCT central thickness reduction over time, compared to the suboptimal responders who lost visual acuity. Most of these visual and anatomic improvements became apparent at weeks 12 to 14, rather than week 6, but continued through month 36.

Looking at the visual acuity outcomes, about 40 percent of patients receiving the FA implant had a suboptimal response of less than a 5-letter gain at week six. Baseline vision and OCT thicknesses were about the same in this group.

In spite of the suboptimal visual response seen in week six, 50% of patients went on to have visual improvement, an additional 30% were stable, and only about 20 percent experienced visual worsening.

As before, these visual outcomes do not appear to be driven by additional treatments, as there is little difference in the number of subsequent injections or focal laser treatments administered to the two groups, with responders once again receiving slightly more such treatments.

Stratifying this group of suboptimal responders based on visual acuity outcomes further shows that patients with the 36-month visual acuity gain (about 70% of patients) also showed moderate OCT central thickness reduction over time, compared to the other six weeks suboptimal responders without visual acuity gain.

Again, as seen with patients with suboptimal anatomic responses, but subsequent visual improvement, these visual and anatomic improvements became much more evident around weeks 12 through 14, not at week 6, and continued through month 36.

Case Study
The case study depicted in the image (from the practice of Dr. Daniel Kiernan, not from the FAME data) is an example of an early suboptimal responder. After receiving the FA implant treatment, this patient had worsening of vision and OCT thickness on follow up at week six.

Because the pharmacokinetics allow for a sustained micro-dose effect over 36 months, Dr. Kiernan decided to hold off on giving additional treatments. Subsequently the patient did experience improvements of anatomy and vision on later exams including at week 14. At recent check up she had maintained excellent vision. She did require cataract surgery, but her vision is now in the 20/30 range. She has not had any recurrent fluid or received additional therapies after her initial FA injection.

Conclusion
For patients with a suboptimal anatomic or visual response at week six following FA injection, 50% may go on to have an improvement in vision and 25% experience stable vision. Those early suboptimal responders who do convert to responders show moderate reduction in edema compared with early suboptimal responders who remain unchanged.

It does not appear that this conversion is due to an imbalance in the amount of rescue therapies given to the week six suboptimal responders versus the week six responders.  The conversion appears to be related to the continuous micro-dosing of fluocinolone associated with the treatment.

So by just following patients who are suboptimal responders six weeks after FA injection, most of them will maintain, or have improvements in, OCT thickness and visual acuity. Although experience using anti-VEGF treatments has created a mindset of expecting monthly visits for patients and injections when any fluid is present, using corticosteroids can be different. Even if there is no rapid improvement after a month, or two, or even three, it may be best to wait and watch, rather than add therapies.