How intravitreal anti-vascular endothelial growth factor initial dosing impacts patient outcomes in diabetic macular oedema

Abstract

Background

Intravitreal anti-vascular endothelial growth factor (VEGF) treatment for diabetic macular oedema (DME) may begin with several initial monthly doses. Characteristics, treatment patterns and outcomes were compared for eyes with DME that did and did not receive such initial doses.

Methods

This was a retrospective database study using American Academy of Ophthalmology Intelligent Research in Sight^® Registry data (01/01/15–31/12/20; index period). Eligible adults had documented DME within 2 months of first anti-VEGF treatment (index date), data available for 12 months beforehand, and ≥ 1 visual acuity (VA) recording ≤ 60 days before index date. Eyes must have received intravitreal anti-VEGF injections during the index period, but none in the prior 12 months. Characteristics and outcomes for eyes with initial doses (three injections within 100 days of index date) were compared with those without. Multivariate Cox Proportional Hazards modelling estimated predictors for treatment discontinuation, re-initiation, or switch; Generalized Estimating Equations-adjusted modelling estimated characteristics associated with receiving initial doses. Demographics and characteristics were summarised. Injection frequency and number, and VA were determined annually for ≤ 6 years. Discontinuations, reinitiations and switches were compared.

Results

Included were 217,696 eyes (n = 77,769 initial; n = 139,927 non-initial) from 166,868 patients. Mean (SD) baseline VA was numerically higher for eyes with versus without initial doses (63.0 [18.1] vs. 62.5 [19.8] letters); this remained during follow-up. Based on modelling results, Eyes with initial doses received more injections (mean [standard deviation (SD)] 11.6 [8.9] vs. 6.1 [6.8] injections) more frequently (interval 7.6 [2.8] vs. 12.6 [7.7] weeks) than eyes without. These differences occurred across follow-up years. Discontinuation (45.7% vs. 63.8%), re-initiation (17.2% vs. 25.0%), and switch (24.5% vs. 31.5%) were less common with initial doses. Asian, Black, and patients of other/unknown race were less likely (P < 0.01) to receive initial doses than White patients, as were Medicare/Medicaid-insured patients versus commercially insured patients (P < 0.01).

Conclusions

Various sociodemographic factors associate with initial anti-VEGF doses, including race, ethnicity and insurance. Although eyes with frequent initial doses maintained higher VA than those without, they also receive more injections over time. Further research may elucidate the impact of frequent initial doses versus total injection number on DME outcomes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12886-024-03797-9.

Background

Diabetic macular oedema (DME) is a leading cause of visual impairment in patients with diabetes, estimated to affect 5.5% of those with diabetes worldwide [1]. DME is defined by an abnormal accumulation of fluid in the macula, which causes retinal and macular swelling [2]. Vascular endothelial growth factor (VEGF) plays a pivotal role in its pathogenesis, and is largely responsible for the capillary hyperpermeability, vascular dysfunction and intravascular pressure that precedes macular fluid buildup [2]. Therefore, intravitreal injection of anti-VEGF agents is the first-line treatment for DME and other common macular diseases such as neovascular age-related macular degeneration (nAMD) and macular oedema secondary to retinal vein occlusion [3, 4]. In DME, anti-VEGF injections have been shown to induce morphological and functional improvements [2]; however, achieving the optimal dosing frequency needed to elicit these improvements can be lacking in clinical practice, and real-world outcomes with anti-VEGF injections consistently fall short of those seen in clinical trials [5, 6].

Anti-VEGF treatment can be initiated with three monthly injections, known as loading doses within the field of nAMD, followed by a maintenance phase with either fixed or individualised treatment intervals [7, 8]. Data from patients with nAMD suggest that these early loading doses correlate with better visual acuity (VA) outcomes, leading to the view that a series of monthly doses may be necessary when beginning anti-VEGF injections [9]. Post-hoc analyses from the phase 3 VISTA-DME and VIVID-DME trials of intravitreal aflibercept support a similar treatment schedule for DME, with functional and anatomical improvements seen following a series of intensive, initial monthly doses [10].

Real-world studies are valuable to understand treatment patterns and unmet needs among patients with DME, especially since clinical trial data often have limited applicability to real-world practice due to restrictive inclusion criteria or other inherent trial biases [11]. Large United States (US) real-world studies using data from electronic health records have shown patients with DME receive fewer anti-VEGF injections and have worse VA outcomes than patients in randomised clinical trials [5, 12].

Understanding real-world treatment patterns and outcomes in DME is vital to guide clinical decision-making. To date, few data exist on the characteristics of patients with DME who do and do not receive initial monthly doses of anti-VEGF agents when starting treatment, or describe the differences in clinical outcomes. The objective of this study was to define and compare patient characteristics, treatment patterns and clinical outcomes of treatment-naïve eyes with DME, which do or do not receive a series of frequent initial anti-VEGF doses when starting treatment.

Methods

This was a retrospective database study using data from the American Academy of Ophthalmology IRIS^® Registry (Intelligent Research in Sight), collected between 01 January 2015 and 31 December 2020 (index period). Data on patients with a diagnosis of DME and treated with an anti-VEGF agent were included. The index date was the earliest documented anti-VEGF injection date within the IRIS^® Registry during the index period, with no injections in the previous ≥ 12 months. The length of follow-up was defined as the time between the index date and the data cutoff (31 December 2021) used in the IRIS^® Registry.

Inclusion/exclusion criteria

Eligible patients were aged ≥ 18 years with documented evidence of DME within 2 months of the index date, with data available for 12 months before the index date, and with one or more recordings of VA at or within 60 days before the index date.

Included eyes must have received anti-VEGF injections during the index period, but none in the prior 12 months. Eyes must not have received intravitreal steroids within the 12 months before the index date, nor received brolucizumab at any point during the study period. Eyes with less than 100 days of follow-up were excluded.

Stratification

Outcomes for eyes that received intravitreal anti-VEGF initial doses were compared with those that did not receive such initial doses. The initial dose cohort included any eyes meeting the inclusion and exclusion criteria that received three anti-VEGF injections of the same agent within 100 days of the index date. This definition is consistent with prior real-world studies [13–15] and expert opinion [16], although recommended doses vary by US Food and Drug Administration (FDA) label. The non-initial dose cohort included any eyes meeting the inclusion and exclusion criteria that did not meet the initial dose cohort definition. Patients were added to the cohorts according to their eyes’ individual characteristics, regardless of the patient’s bilateral status.

Outcomes

Baseline demographics and characteristics were summarised for both cohorts, and described by location, ethnicity, insurance model and prescriber specialty (defined as the treating provider at the first anti-VEGF injection). The frequency and number of intravitreal anti-VEGF injections were determined by follow-up year, as were the intervals between injections, for both cohorts. Treatment discontinuation was defined as more than 1 year (365 days) of contribution to the IRIS^® Registry without an additional anti-VEGF injection following the last injection recorded. Treatment re-initiation was defined as a record of treatment with an anti-VEGF agent ≥ 1 year after the last injection. Switch was defined as a record of a different anti-VEGF agent ≥ 1 year after the last injection. Handling of patients who switched < 1 year after the last injection is described in the statistical analysis section.

VA and changes from baseline VA were evaluated annually in the VA cohort for study eyes for up to 6 years of follow-up. Values were calculated as best VA in approximate ETDRS (Early Treatment for Diabetic Retinopathy Study) letters converted from either Snellen (ETDRS = 85 + 50 × log [Snellen fraction]) or Logarithm of the Minimum Angle of Resolution values [17].

Statistical analyses

Means, standard deviations (SD) and medians were calculated for continuous variables and counts and percentages for categorical variables. Differences in baseline characteristics and patient follow up were analysed using Chi-square tests of independence with Yates’s correction for categorical variables, and Welch’s t-tests were used for continuous variables [18, 19].

Various models were used to assess the impact of the receipt of initial doses on the VA at 1, 3 and 5 years after treatment initiation controlling for patients’ clinical and demographic characteristics, including age, race/ethnicity, insurance status, baseline VA, history of glaucoma or cataracts, and initial anti-VEGF agent. These models also controlled for an anti-VEGF agent switch at any point through the models’ follow-up periods.

Multivariate Cox Proportional Hazards modelling was done to estimate predictors for time to treatment discontinuation, re-initiation, or switch.

A Generalized Estimating Equations (GEE)-adjusted model was used to estimate the association of baseline clinical and demographic characteristics with receipt of initial doses. This controlled for age, race/ethnicity, insurance status at treatment initiation, baseline VA, history of glaucoma or cataracts and initial anti-VEGF agent. A GEE-adjusted model was also used to further investigate the factors influencing the likelihood of receiving three initial doses. A patient-eye level analysis was performed with data grouped at the patient level to account for non-independence between eyes of the same patient. The GEE model was chosen over a traditional logistic regression model because it allows for the appropriate handling of correlated data, such as the hierarchical structure of eyes nested within patients [20].

Change in VA from baseline was assessed using a standard linear regression (a generalised linear model with a Gaussian family and identity link function). Bilateral status was always considered for the adjusted analyses.

Survival analysis

Cox proportional hazards models were used to evaluate the time to three distinct treatment-related events: (1) treatment discontinuation, (2) treatment switch (initiation of a different agent than the starting agent ≥ 1 year after the last injection), and (3) re-initiation after discontinuation. The analyses were conducted at the patient-eye level, accounting for the non-independence between eyes in the same patient by clustering the data at the patient level. Loading status was included as a stratification factor to account for potential violations of the proportional hazards assumption related to this variable. This allowed for the control of any differences in starting treatment patterns that may have confounded the hazard ratios (HRs) without estimating the effects of initial/non-initial dose status.

Results

Patient population

In total, 2,017,445 eyes from 1,555,917 patients were identified as having initiated anti-VEGF between 01 January 2015 and 31 December 2020, with no evidence of injections in the preceding 12 months. Of these, 217,696 eyes from 166,868 patients met the eligibility criteria and were included in the study (Table 1).

Table 1.

Cohort attrition

		Patients, n (%)	Eyes, n (%)
1	All eyes with anti-VEGF injections in index period with no anti-VEGF injections 12 months pre-index	1,555,917 (100.0)	2,017,445 (100.0)
2	No intravitreal steroid use in 12-months pre-index	1,548,070 (99.5)	2,005,325 (99.4)
3	At least 12 months of data prior to index date	840,880 (54.0)	1,046,296 (51.9)
4	Documentation of DME within 2 months prior to or on index date	195,787 (12.6)	257,514 (12.8)
5	Age ≥ 18 and known sex at index date	194,397 (12.5)	255,633 (12.7)
6	Record of ≥ 1 VA recording at or within 60 days before index date	179,181 (11.5)	233,702 (11.6)
7	Exclude eyes with brolucizumab usage at any point in the study periodb	179,089 (11.5)	233,572 (11.6)
8	Exclude eyes without ≥ 100 days of follow-up	166,868 (10.7)	217,696 (10.8)
9	Enrolled to:
	Initial-dose cohort (if met the initial dose definition: receipt of ≥ 3 injections within the first 100 days of treatment)	65,867 (39.5)a	77,769 (35.7)
	Non-initial dose cohort (if did not meet the initial dose definition)	113,383 (68.0)a	139,927 (64.3)

Abbreviations: DME diabetic macular oedema, VA visual acuity, VEGF vascular endothelial growth factor

^aThe number of patients enrolled in the initial and non-initial dose cohorts does not equal the total in row 8, because some patients had an eye in each cohort and are thus included in both totals

^bEyes receiving brolucizumab were excluded due to safety signals identified and rapid discontinuation of brolucizumab use in the US, to prevent mischaracterization of responses

The initial dose cohort included 77,769 eyes from 65,867 patients, and the non-initial dose cohort contained 139,927 eyes from 113,383 patients. Some patients had one eye meeting the criteria for the initial dose cohort and another eye that did not; thus, some patients are counted in both groups.

The baseline demographic and clinical characteristics are shown in Table 2 and supplementary Table S1.

Table 2.

Baseline demographics (included eyes)

	Initial (N = 77,769)	Non-initial (N = 139,927)	P value
Age, years
Mean (SD)	65.08 (11.1)	63.43 (11.6)	< 0.0001
Median (range)	66 (18–88)	64 (18–88)
IQR	58–73	56–71
Age group, n (%)
18–39	1,783 (2.3)	4,685 (3.4)	< 0.0001
40–49	4,721 (6.1)	11,135 (8.0)
50–59	15,537 (20.0)	31,632 (22.6)
60–69	27,819 (35.8)	48,933 (35.0)
70–79	20,939 (26.9)	33,180 (23.7)
80–89	6,970 (9.0)	10,362 (7.4)
Sex, n (%)
Male	39,865 (51.3)	70,607 (50.5)	0.0003
Female	37,904 (48.7)	69,320 (49.5)
Race, n (%)
White or Caucasian	52,754 (67.8)	87,147 (62.3)	< 0.0001
Black or African American	10,554 (13.6)	22,350 (16.0)
Asian	2,109 (2.7)	4,130 (3.0)
Other race	2,413 (3.1)	5,333 (3.8)
Unknown	9,126 (11.7)	19,221 (13.7)
Ethnicity, n (%)
Hispanic	11,242 (14.5)	24,447 (17.5)	< 0.0001
Non-Hispanic	55,305 (71.1)	94,917 (67.8)
Unknown	11,222 (14.4)	20,563 (14.7)
US census regiona, n (%)
Midwest	16,424 (21.1)	27,110 (19.4)	< 0.0001
North	12,463 (16.0)	25,802 (18.4)
South	33,180 (42.7)	59,441 (42.5)
West	14,864 (19.1)	25,529 (18.2)
Unknown	838 (1.1)	2,045 (1.5)
Insurance/payer type, n (%)
Medicare	40,916 (52.6)	70,395 (50.3)	< 0.0001
Medicaid	5,244 (6.7)	11,893 (8.5)
Commercial	24,670 (31.7)	45,049 (32.2)
Other	6,939 (8.9)	12,590 (9.0)
Provider specialty, n (%)
Retina Specialist	67,105 (86.3)	120,453 (86.1)	< 0.0001
Non-Retina Specialist	2,300 (3.0)	4,197 (3.0)
General Ophthalmologist	6,963 (9.0)	11,831 (8.5)
Unknown	1,401 (1.8)	3,446 (2.5)
Length of follow up (years) Mean, SD	2.87 (1.61)	2.94 (1.67)	< 0.0001
Patients with follow up, n (%)  Totalb  > 0 and ≤ 1 years  > 1 and ≤ 2 years  > 2 and ≤ 3 years  > 3 and ≤ 4 years  > 4 and ≤ 5 years  > 5 and ≤ 6 years  > 6 and ≤ 7 years  > 7 years	61,246 (100) 7,998 (13.1) 13,358 (21.8) 13,522 (22.1) 11,340 (18.5) 7,982 (13.0) 4,455 (7.3) 2,092 (3.4) 499 (0.8)	105,612 (100) 13,791 (13.1) 22,023 (20.9) 22,454 (21.3) 18,813 (17.8) 14,834 (14.1) 8,525 (8.1) 4,194 (4.0) 978 (978)	< 0.0001

Abbreviations: IQR interquartile range, SD standard deviation, US United States

^aMidwest: ND, SD, NE, KS, MO, MN, IA, WI, IL, IN, OH, MI. North: ME, NH, VT, PA, NY, NJ, MA, RI, CT. South: MD, DE, WV, VA, DC, NC, SC, KY, TN, FL, GA, AL, MS, LA, AR, OK, TX. West: NM, AZ, CO, UT, WY, MT, ID, NV, CA, OR, WA, AK, HI

^bOne eye per patient is reported for follow up. To define the study period for which healthcare resource utilization would be described, the maximum amount of follow-up time was used. For example, for patients with bilateral eyes affected, healthcare utilization would be assessed from the earliest index date to the latest end of follow-up date, regardless of eye. P values were calculated using Chi-square tests of independence with Yates’s correction for categorical variables, and Welch’s t-tests were used for continuous variables

The mean (SD) length of follow-up was differed slightly between cohorts: 2.87 (1.61) years for the initial dose cohort and 2.94 (1.67) years for the non-initial dose cohort.

The mean age was slightly higher in the initial dose cohort, and the cohorts had slightly different distributions of males vs. females. Most eyes were from White/Caucasian patients, and these were slightly more numerous in the initial than non-initial dose cohort (67.8% [n = 52,754] initial; 62.3% [n = 87,147] non-initial). Compared with those in the non-initial dose cohort, fewer eyes in the initial dose cohort were from patients of Hispanic ethnicity (14.5% [n = 11,242] initial; 17.5% [n = 24,447] non-initial).

Beginning treatment with or without initial doses was similarly common in all census regions, but overall, the largest number of eyes were enrolled from patients in the South (42.7% [n = 33,180] initial; 42.5% [n = 59,441] non-initial]). Just over half of patients had Medicare insurance, and approximately a further third had commercial insurance, with the proportions slightly different between cohorts. Over 86% of patients were treated by a retina specialist (Table 2). Eyes with proliferative diabetic retinopathy were less commonly treated with initial doses than without initial doses (39.1% [n = 30,382] initial; 50.7% [n = 70,944] non-initial; supplementary Table S1.

Injection numbers and frequency

Over the average follow-up period, the number of intravitreal injections (including initial doses) of anti-VEGF agents was higher in eyes that received initial doses than in eyes that received no initial doses (mean [SD] intravitreal injections: 11.6 [8.9] initial; 6.1 [6.8] non-initial). Most patients had ≤ 3 years of follow up (56.9%, n = 34,878).

At Year 1 of follow-up, the mean (SD) number of intravitreal injections was 6.3 (2.5) for eyes in the initial dose cohort and 2.9 (2.1) (Fig. 1A) for eyes in the non-initial dose cohort. 30% (n = 23,435) of eyes in the initial dose cohort had received only 3–4 injections in Year 1, compared with 22.8% (n = 31,913) of those in the non-initial dose cohort. Similarly, 37.4% (n = 29,059) of eyes in the initial dose cohort received only 5–7 injections in Year 1, compared with 15.8% (n = 22,075) in the non-initial dose cohort.

Fig. 1.

A Non-annualized number of injections and (B) injection interval at Years 1–6, stratified by initial dose. Error bars are standard deviation. I, initial dose cohort; NI, non-initial dose cohort

At Year 6 of follow-up, the mean (SD) number of intravitreal injections was 3.5 (2.5) for eyes in the initial dose cohort and 3.0 (2.1) for eyes in the non-initial dose cohort. In Year 6, 26.6% (n = 980) eyes in the initial dose cohort received 3–4 injections compared with 28.3% (n = 1679) in the non-initial dose cohort. The numbers that received 5–7 injections in Year 6 were lower: 21.6% (n = 797) in the initial dose cohort and 16.5% (n = 979) in the non-initial dose cohort.

Injection intervals

Over the average follow-up period, mean (SD) injection intervals were shorter for the eyes that received initial doses (7.6 [2.8] weeks) compared with those that did not (12.6 [7.7] weeks; Fig. 1B). Most eyes in the initial dose cohort (59.5%; n = 46,279) experienced injections at intervals > 5 weeks to ≤ 9 weeks, compared with 30,599 (44.3%) in the non-initial dose cohort. Intervals of > 13 to ≤ 18 weeks, and of > 18 weeks were far more numerous among eyes in the non-initial dose cohort than in the initial dose cohort (> 13 to ≤ 18 weeks: 4.7% [n = 3656] initial vs. 12.6% [n = 18,233] non-initial; and > 18 weeks: 0.6% [n = 504] initial versus 10.8% [n = 17,384] non-initial; supplementary Table S2.

VA outcomes

Mean (SD) baseline VA was similar between eyes in the initial dose cohort and in the non-initial dose cohort (63.0 [18.1] vs. 62.5 [19.8] ETDRS letters, respectively; Fig. 2; supplementary Table S3. After 100 days, mean (SD) VA was slightly higher for eyes in the initial dose cohort than in the non-initial dose cohort (67.4 [15.9] vs. 65.94 [18.0] ETDRS letters). This trend continued across 1–5 years of follow-up, although the difference in mean VA reduced with each additional year of follow-up (mean [SD] VA in initial vs. non-initial doses, respectively, was 68.0 [16.3] vs. 66.2 [18.4] at Year 1 [P < 0.0001]; 67.6 [16.9] vs. 65.9 [18.7] at Year 2 [P < 0.0001]; 67.2 [17.3] vs. 65.8 [18.9] at Year 3 [P < 0.0001]; 66.5 [17.9] vs. 65.6 [18.9] at Year 4 [P < 0.0001], 66.0[18.1] vs. 65.2 [19.1] at Year 5 [P = 0.0041], and 65.3 [18.2] vs. 65.9 [18.2] at Year 6 [P < 0.0001]).

Fig. 2.

Visual acuity outcomes in initial and non-initial dose cohorts. ETDRS, Early Treatment Diabetic Retinopathy Study; I, initial dose cohort; NI, non-initial dose cohort; VA, visual acuity

15-Letter improvers

At any point in the study, VA improvement of ≥ 15 letters was more common among eyes that received initial doses than those that did not (37.7% [n = 21,204] initial; 31.6% [n = 31,979] non-initial; Fig. 3).

Fig. 3.

Eyes with 15-letter improvement at (A) any point during the study, and (B) at the end of the study, stratified by receipt of initial dose. I, initial dose cohort; NI, non-initial dose cohort; VA, visual acuity

By the end of the study, 20.3% (n = 11,395) of eyes in the initial dose cohort, and 18.2% (n = 18,408) in the non-initial dose cohort had achieved 15-letter improvements. Conversely, 10.3% (n = 5766) and 12.1% (n = 12,221) of eyes in the initial and non-initial dose cohorts, respectively, had experienced 15-letter declines by the end of the study. In both cohorts, VA improvement was most prevalent among eyes that initiated anti-VEGF treatment with worse than 20/40 vision, while 15-letter declines did not correlate with baseline VA.

Treatment switching and discontinuations

Switching treatments was less common among eyes that received initial doses than eyes that did not receive initial doses. Overall, one quarter (n = 17,028) of eyes in the initial dose cohort and just under a third (n = 22,481) of eyes in the non-initial dose cohort underwent a treatment switch. The median time to the first switch was 39 and 38 weeks, respectively (Fig. 4).

Fig. 4.

Treatment switches among eyes with ≥ 4 injections in initial and non-initial dose cohorts (A). Change in VA from baseline at time of switch (B). I, initial dose cohort; NI, non-initial dose cohort

In both groups, patients who discontinued did so after experiencing improvements in VA. The mean (SD) improvement from baseline to discontinuation was + 3.87 (13.9) letters in the initial-dose cohort and + 1.14 (10.9) letters in the non-initial dose cohort. Those who re-initiated did so after VA had worsened, by a mean (SD) 6.73 (17.6) and 4.79 (19.0) letters in the initial and non-initial dose cohorts, respectively, from the time of discontinuation, and by a mean (SD) 2.83 (19.4) and 3.62 (19.9) letters, respectively, from baseline (Fig. 5).

Fig. 5.

Treatment discontinuation, re-initiation, and switching in initial and non-initial dose cohorts. Mean (SD) VA change, baseline to discontinuation: I: +3.87 (13.9) letters, NI: +1.14 (10.9) letters. Mean (SD) VA change, baseline to re-initiation: I: −2.83 (19.4) letters, NI: −3.62 (19.9) letters. Discontinuation to re-initiation: I: −6.73 (17.6) letters, NI: −4.79 (19.0) letters. Discontinuation defined as ≥ 365 days of contribution to the IRIS^® Registry without an anti-VEGF injection. The mean (SD) number of injections received from index to discontinuation was 6.83 (4.8) among patients in the initial dose cohort and 2.82 (3.2) among patients in the non-initial dose cohort. I, initial dose cohort; NI, non-initial dose cohort; SD, standard deviation; VA, visual acuity; VEGF, vascular endothelial growth factor

Switching away from bevacizumab to FDA-approved agent(s) was more common than the other way around for both cohorts. In the initial and non-initial dose cohorts, respectively, 18.4% (n = 12,762) and 22.9% (n = 16,310) switched away from bevacizumab, while 1.8% (n = 1251) and 4.2% (n = 3003) switched to bevacizumab.

Both treatment discontinuation and subsequent re-initiation were less common among eyes in the initial dose cohort. Less than half (45.7%; n = 35,558) of the eyes in the initial dose cohort discontinued treatment, compared with nearly two thirds (63.8%; n = 89,285) of those in the non-initial dose cohort. A similar trend was evident with eyes re-initiating treatment: 17.2% (corresponding to 37.7% of the patients discontinuing, n = 13,392) and 25.0% (corresponding to 39.2% of patients discontinuing, n = 35,015) of eyes in the initial and non-initial dose cohorts re-initiated treatment, after a median treatment-free interval of 81.7 (range: 52.3–354.1) and 84.0 (range: 52.3–384.9) weeks (Fig. 5).

Models and adjusted analyses

After controlling for baseline characteristics, receiving initial dosing was associated with a slight increase in VA from baseline to Year 1 (+ 1.23 letters), Year 3 (+ 1.47 letters) and Year 5 (+ 1.23 letters) compared with eyes that did not receive the initial dosing regimen. Similarly, initial doses were associated numerically with VA improvements from Year 1 to Year 3 (+ 0.51 letters) and to Year 5 (+ 0.53 letters). All baseline VA statuses were significantly associated with vision gains following anti-VEGF initiation (P < 0.01). Across all models, having Medicaid insurance was significantly associated with vision loss following anti-VEGF initiation (P < 0.01), as was starting bevacizumab as a first agent and receiving multiple anti-VEGF agents during the follow-up period (P < 0.05; Table 3).

Table 3.

Linear model: change in visual acuity from baseline to years 1 and 3, stratified by demographics

	Baseline to Year 1		Baseline to Year 3
	Coefficient (95% CI)	P value	Coefficient (95% CI)	P value
Intercept	−2.10 (− 2.71, − 1.48)	0.000	−2.97 (− 3.83, − 2.12)	0.000
Initial	1.23 (1.03, 1.44)	0.000	1.47 (1.19, 1.75)	0.000
Age (from 18)	−0.08 (− 0.09, − 0.08)	0.000	−0.11 (− 0.12, − 0.10)	0.000
Sex (reference: female)
Male	0.45 (0.27, 0.63)	0.000	0.39 (0.13, 0.66)	0.004
Race (reference: White)
Asian	−0.50 (− 1.05, 0.04)	0.072	0.47 (− 0.32, 1.26)	0.242
Black	−0.58 (− 0.85, − 0.31)	0.000	−0.42 (− 0.82, − 0.01)	0.044
Native American and Alaska Native	0.44 (− 0.68, 1.57)	0.440	−1.08 (− 2.73, 0.56)	0.197
Native Hawaiian and Other Pacific Islander	0.75 (− 0.88, 2.38)	0.366	0.18 (− 2.05, 2.40)	0.876
Other	0.15 (− 0.44, 0.74)	0.623	0.01 (− 0.85, 0.87)	0.984
Unknown	0.01 (− 0.32, 0.34)	0.937	0.33 (− 0.18, 0.85)	0.199
Ethnicity (reference not Hispanic or Latino)
Hispanic or Latino	−0.73 (− 1.03, − 0.42)	0.000	−0.62 (− 1.06, − 0.18)	0.006
Unknown	−0.02 (− 0.32, 0.28)	0.876	−0.21 (− 0.69, 0.26)	0.379
Payer (reference: commercial)
Medicare	0.43 (0.22, 0.65)	0.000	0.52 (0.20, 0.83)	0.001
Medicaid	−1.07 (− 1.46, − 0.68)	0.000	−2.11 (− 2.68, − 1.54)	0.000
Other	0.19 (− 0.14, 0.53)	0.257	−0.27 (− 0.78, 0.25)	0.314
Comorbidity
Glaucoma	−1.20 (− 1.43, − 0.97)	0.000	−1.91 (− 2.24, − 1.57)	0.000
Cataract	0.72 (0.54, 0.91)	0.000	1.25 (0.98, 1.52)	0.000
Anti-VEGF agent
Aflibercept (reference)	–	–	–	–
Bevacizumab	−0.51 (− 0.74, − 0.27)	0.000	−0.70 (− 1.06, − 0.35)	0.000
Ranibizumab	0.13 (− 0.19, 0.44)	0.438	0.56 (0.09, 1.03)	0.019
Multiple agents (80% threshold)	−0.76 (− 1.02, − 0.50)	0.000	−1.20 (− 1.61, − 0.78)	0.000
Anti-VEGF injections through Year 1	0.27 (0.22, 0.31)	0.000	–	–
Anti-VEGF injections through Year 3	–	–	0.08 (0.06, 0.10)	0.000
Baseline visual acuity (reference: 20/25 or better)
Worse than 20/25 − 20/40	3.59 (3.33, 3.85)	0.000	4.17 (3.79, 4.54)	0.000
Worse than 20/40 − 20/80	7.76 (7.49, 8.03)	0.000	8.84 (8.45, 9.23)	0.000
Worse than 20/80 - better than 20/200	12.22 (11.81, 12.62)	0.000	14.37 (13.76, 14.98)	0.000
20/200 or worse	26.50 (26.15, 26.85)	0.000	30.35 (29.82, 30.88)	0.000
Switch before Year 1	0.03 (− 0.23 0.30)	0.805	–	–
Switch before Year 3	–	–	0.26 (− 0.12, 0.65)	0.185

Abbreviations: CI confidence interval, VEGF vascular endothelial growth factor

GEE modelling revealed that a patient’s race/ethnicity and insurance status significantly impacted the likelihood of receiving initial doses. Asian (− 0.17), Black (− 0.23), and patients of other (− 0.11) and unknown race (− 0.20) had a significantly lower expected likelihood (P < 0.01) of receiving initial doses than White patients, adjusted for other baseline demographic and clinical factors. Similarly, patients of Hispanic ethnicity were significantly less likely to receive initial doses than their non-Hispanic counterparts (− 0.17; P < 0.01). Furthermore, patients insured with Medicare (− 0.05) or Medicaid (− 0.12) were significantly less likely (P < 0.01) to receive initial doses than patients with commercial insurance. Compared with aflibercept, initial doses were significantly less likely when bevacizumab was given as the first treatment (− 0.20; P < 0.01), and significantly more likely when ranibizumab was given as the first treatment (+ 0.13; P < 0.01). Patients with baseline VA of worse than 20/25 to better than 20/200 were also more likely to receive initial doses than those starting treatment with vision 20/25 or better (Table 4).

Table 4.

GEE modelling of initial dose as a response variable

	GEE – Initial dose
	Coefficient (95% confidence interval)	P value
Age category
18–39	–	–
40–59	0.11 (0.03, 0.18)	0.00
50–59	0.23 (0.17, 0.30)	0.00
60–69	0.37 (0.30, 0.43)	0.00
70–79	0.47 (0.40, 0.53)	0.00
80–89	0.53 (0.45, 0.60)	0.00
Sex
Female	–	–
Male	0.03 (0.01, 0.05)	0.01
Race
White	–	–
Black or African American	−0.23 (− 0.26, − 0.20)	0.00
Asian	−0.17 (− 0.23, − 0.11)	0.00
Other	−0.11 (− 0.16, − 0.05)	0.00
Unknown	−0.20 (− 0.23, − 0.16)	0.00
Ethnicity
Not Hispanic or Latino	–	–
Hispanic or Latino	−0.17 (− 0.20, − 0.14)	0.00
Unknown	0.03 (0.00, 0.06)	0.08
Payer
Commercial	–	–
Medicaid	−0.12 (− 0.16, − 0.08)	0.00
Medicare	−0.05 (− 0.07, − 0.03)	0.00
Other	0.01 (− 0.03, 0.04)	0.69
Cataract
No cataracts	–	–
Cataracts	0.08 (0.06, 0.10)	0.00
Glaucoma
No glaucoma	–	–
Glaucoma	−0.05 (− 0.07, − 0.02)	0.00
Index agent
Aflibercept	–	–
Bevacizumab	−0.20 (− 0.22, − 0.17)	0.00
Ranibizumab	0.13 (0.09, 0.16)	0.00
Baseline visual acuity
20/25 or better	–	–
Worse than 20/25 − 20/40	0.09 0.06, 0.12)	0.00
Worse than 20/40 − 20/80	0.16 (0.13, 0.19)	0.00
Worse than 20/80 - better than 20/200	0.14 (0.10, 0.18)	0.00
20/200 or worse	0.00 (− 0.03, 0.04)	0.79

Abbreviation: GEE Generalized Estimating Equations

In multivariate analyses, the strongest predictor of treatment discontinuation was starting with bevacizumab (HR: 1.7; 95% confidence interval [CI]: 1.67–1.73; P < 0.005). Interestingly, receiving < 3 initial doses was the strongest predictor of treatment continuation (HR: 0.51; 95% CI: 0.51–0.52; P < 0.005), followed by receiving multiple agents (HR: 0.62; 95% CI: 0.6–0.63; P < 0.005). Treatment re-initiation was less likely for patients aged 70–79 years (HR: 0.88; 95% CI: 0.82–0.94; P < 0.005), or 80–89 years (HR: 0.79; 95% CI: 0.73–0.85; P < 0.005), and those with baseline VA worse than 20/80 but better than 20/200 (HR: 0.92; 95% CI: 0.88–0.96; P < 0.005), or 20/200 or worse (HR: 0.79; 95% CI: 0.76–0.82; P < 0.005), compared with patients in other age or VA categories. Ages 60–69 and 50–59 at baseline were also strong predictors of switch (for ages 60–69, HR: 1.23; 95% CI: 1.15–1.33; P < 0.005, and ages 50–59, HR: 1.21; 95% CI: 1.12–1.3; P < 0.005). Receiving < 3 initial doses also predicted treatment switch (HR: 1.17; 95% CI: 1.15–1.2; P < 0.005; Fig. 6), although this may correlate with starting bevacizumab treatment.

Fig. 6.

Multivariate Cox proportional-hazard modelling of likelihood of treatment (A) discontinuation, (B) re-initiation, or (C) switch. VA, visual acuity

Discussion

This was a retrospective analysis of data from the US-wide IRIS^® Registry, involving data from over 200,000 eyes treated in the first line with intravitreal anti-VEGF injections for DME. The study sought to understand differences in outcomes between patients who receive frequent initial doses when starting anti-VEGF treatment and those who do not.

Generally, patients who received initial doses experienced better clinical outcomes than patients who did not, including a significantly increased likelihood of VA improvement through 6 years of treatment (although not exceeding an increase in 1.5 letters from baseline) after controlling for differences in baseline variables. Further, patients who received initial doses were more likely to experience 15-letter improvement. These findings are consistent with data from a small prospective study in patients with nAMD, which found that patients who received three 6-weekly loading doses had better VA and improved visual stability than those who did not, although the associations were weak, with only the visual stability outcome reaching statistical significance [9].

Importantly, patients who received initial doses as part of their DME treatment ultimately had more anti-VEGF injections and at more frequent intervals throughout their follow-up period than patients who did not receive initial doses. This presents the question of whether the observed VA improvements in patients who received frequent initial doses are attributable to the initial doses themselves or the higher overall number of injections over time. Data from the real-world LUMINOUS study of ranibizumab in patients with DME found that VA gains from baseline were greater in patients who received ≥ 5 injections in the year-long study, but particularly in those who had received ≥ 3 injections within the first 100 days [21]. Further research may be warranted to elucidate and clarify the contribution of frequent initial doses to improved treatment outcomes.

Patients who received loading doses were less likely to switch agents, potentially due to the VA gains observed. Among the switches that did occur, the most frequent was a switch from off-label bevacizumab to on-label agents. This could be due to a perceived lack of efficacy of bevacizumab, potentially resulting in physicians wanting to try a therapy with a confirmed efficacy profile. However, research published after this study’s index period found that patients who began therapy with VA worse than 20/50 achieved similar VA outcomes after 2 years whether they started treatment with bevacizumab and then switched to aflibercept, or received aflibercept from the beginning [22]. Such data are filling knowledge gaps around the use of anti-VEGF agents, and it may be interesting to see whether practices have changed since our study was conducted.

Fewer eyes from Black/African American and Hispanic patients received initial doses than their White and non-Hispanic counterparts, as did patients with Medicaid or Medicare insurance compared with commercial insurance. These factors were associated with a significantly reduced likelihood of receiving initial doses, although differences were minimal. Previous studies have shown various socioeconomic characteristics to be associated with visual health, treatment utilisation and outcomes among patients with DME. For example, in another retrospective study from the IRIS^® Registry, those with Medicare or private insurance presented with higher baseline VA than patients on Medicaid, while Black and Hispanic patients presented with a worse baseline VA than White and non-Hispanic patients [23]. Another large database study had similar findings and showed race, ethnicity and insurance category to be significant predictors of the use of anti-VEGF injections and of longitudinal VA outcomes [24].

Together, these findings contribute to the wider picture of healthcare utilisation according to socioeconomic variables and race. Other studies have noted how factors such as systemic racism, socioeconomic status, poor access to care and nutrition, health literacy and education may contribute to worse VA in Black/African American patients upon initiation of treatment [23], as well as perhaps differences in risk factors for diabetes itself, and diabetic retinopathy [25]. In this study, having lower VA at baseline and being non-White or of Hispanic ethnicity was associated with a reduced likelihood of receiving initial doses, providing further evidence to suggest that people in minority groups may be at a particular disadvantage.

The current data, and those more widely, show that the effectiveness of intravitreal anti-VEGF injections is variable and sensitive to various factors, including initial dose receipt, baseline factors and clinical characteristics [5, 6, 12]. Achieving the optimal dosing frequency needed to maximise visual outcomes is difficult in clinical practice, even among adherent patients [6]. Newer agents and drug delivery systems are being investigated to extend the durability of existing anti-VEGF therapy, and to inhibit additional therapeutic targets [26]. The study index period was such that treatment with newer, more durable agents for DME management (e.g., aflibercept 8 mg and faricimab) was not captured. It is hoped that such progress may address the limitations of existing treatments and help patients to achieve better clinical outcomes.

As with all retrospective analyses, the present study has inherent limitations relating to selection bias and confounding. Further limitations arise from the nature of reporting data from routine clinical practice, as information is not necessarily complete or standardised between providers and settings and may include recall and reporting errors. Indeed, due to this variation, agent doses may have varied between patients. Relevant data may not have been recorded in an appropriate window around key time points, as VA records in the IRIS registry are scheduled based on patient needs and physician preference, which may not match time points of interest. The majority of patients had follow up of ≤ 3 years, thus data extracted at later timepoints are based on a smaller sample and thus may not be representative of the wider population. Further, it is difficult to interpret discontinuation data, as discontinuation may be due to treatment success as well as failure, and difficult to separate from patients lost to follow-up. Recognizing this problem, a recent IRIS^® Registry study involving patients with nAMD aimed to separate and quantify patients who truly discontinue treatment from those lost to follow-up. In a sample of over 156,000 patients, researchers estimated that 1 in 9 was lost to follow-up, while a larger proportion—1 in 7—had discontinued treatment. It is likely that similar behaviours were present in this study. The lack of baseline anatomical characteristics (e.g., retinal thickness), and therefore the ability to stratify for such factors, is also a limitation.

The requirement for previous observation period of ≥ 12 months with no anti-VEGF treatment likely excluded many patients with shorter follow up who had presented at clinic needing treatment. We chose the ≥ 12 month cut off to ensure prior anti-VEGF treatment experience was not confounding treatment responses; moreover, a shorter time interval may result in patients experiencing a drug holiday or treatment gaps due to insurance issues being excluded. Therefore, changing the ≥ 12 month cutoff would likely increase sample size but would confound subsequent analyses.

The study’s large sample size (over 200,000 eyes) was a key strength and may mitigate the effects of such limitations to provide a snapshot of real-world clinical practice in the US. This study cannot rule out that patients could have been previously treated in other settings prior to data collection for this analysis, and that this data set therefore includes patients who were not treatment naïve. However, this is unlikely or at least very limited, since the look-back period for treatment was a full year before the index date.

Conclusions

In summary, this large retrospective study of eyes from patients with treatment-naïve DME found several sociodemographic factors to be associated with receiving initial doses of intravitreal anti-VEGF, notably relating to race, ethnicity and patient insurance model. Future study will help to understand whether health inequity is solely responsible for the observed differences or if other unreported factors are involved. Further, data suggest that receiving a series of frequent initial doses of anti-VEGF may improve clinical outcomes for patients with DME, although it is unclear to what extent initial doses contribute to these effects versus an increased overall number of injections over time.

Abbreviations

Ang

Angiopoietin

CI

Confidence interval

DME

Diabetic macular oedema

ETDRS

Early Treatment for Diabetic Retinopathy Study

FDA

US Food and Drug Administration

GEE

Generalized Estimating Equations

HR

Hazard ratio

IRIS

Intelligent Research in Sight

nAMD

Neovascular age-related macular degeneration

SD

Standard deviation

US

United States

VA

Visual acuity

VEGF

Vascular endothelial growth factor

Authors’ contributions

RPS, DT, BLK, ALP, TL, EK, MH, VG: Contributed to study design and conduct, results interpretation and critically revised the manuscript.

Funding

This study was funded by Genentech, Inc., a member of the Roche Group, and this sponsor had a role in study design, study conduct, data management, analysis, interpretation and presentation, manuscript review and approval. This research was funded in part by Research to Prevent Blindness (TL) and NIH grant P30-EY026877 (TL).

Data availability

Due to the sensitivity and privacy around the IRIS® Database, raw data must remain confidential and cannot be shared. Aggregated data are available upon reasonable request from the corresponding author (email: drrishisingh@gmail.com).

Declarations

Ethics approval and consent to participate

No human participants were included in this study. This study adheres to the Declaration of Helsinki. The study was reviewed by WCG Institutional Review Board (IRB) (Puyallup, WA) and was considered exempt from ethical approval due to using anonymized, de-identified data. Furthermore, WCG IRB waived the need for patient consent as Veran Health certified the study was conducted on anonymized, deidentified data for research purposes.

Consent for publication

Not applicable.

Competing interests

AL and MH: Employee salary; stock or stock options – Verana Health. BLK: No conflicts of interest. DT, EK, and VG: Employee salary; Stock – Genentech, Inc. RPS: Personal fees – Genentech/Roche, Alcon, Novartis/Gyroscope, Apellis, Zeiss, Bausch + Lomb, Regeneron Pharmaceuticals, Inc. TL: Grants/Contracts – Astellas; Consulting fees – Alcon, Aptitude Medical, Astellas, Boehringer Ingelheim, Iridex, Nanoscope, Regeneron, Roche/Genentech, Verana Health, Virtual Field.