The value of the accelerated approval pathway: real-world outcomes associated with reducing the time between innovations

Figures & data

Figure 1. Study concept and design.

(A) Depicts how the analysis isolates the impact of the AA pathway on patient survival. All patients start with the initial treatment (a new drug with extended survival) and then surviving patients progress to a subsequent line of therapy. As the AA drug represents the best overall survival, it is assumed that more patients who are alive when the drug is approved will use the AA drug, and the remainder of patients are distributed between the no treatment C and SoC B cohorts based on real-world trends in the real-word data. As the timing of the AA drug availability is delayed for the counterfactual scenario, the percentage of patients alive and eligible for the drug reduces. The analysis compares average patient outcomes across these two possible outcomes to isolate how the timing of the AA drug impacts patient outcomes. (B) Depicts how patients flow through initial and subsequent treatments across the three treatment cohorts (AA drug [A], SoC [B], no treatment [C]). Each cohort is followed up from the time of the initiation of the initial drug through subsequent treatments to estimate overall survival at the patient level before extrapolating patient-level outcomes to population estimates based on real-world market use of AA drugs during the AA period based on shares from publicly available epidemiology data.

*Next-line SoC treatments for surviving patients in cohorts A and B may occur in pre-AA period.

AA: Accelerated approval; SoC: Standard of care.

Table 1. Included drugs and key approval dates.

	Metastatic melanoma	Advanced bladder cancer	EGFR+ aNSCLC
New drug/indication	1L ipilimumab	2L+ post platinum immunotherapy^§	EGFR+ erlotinib, gefitinib or afatinib
Traditional approval date	28 March 2011	18 May 2016 (atezolizumab)^‡	14 May 2013 (erlotinib)^‡
Next innovation (AA product[s])	Immunotherapy (pembrolizumab and nivolumab)	Enfortumab vedotin-ejfv	Osimertinib
AA date	4 September 2014 (pembrolizumab)^‡	18 December 2019	13 November 2015
Counterfactual AA dates
Date of data presentation	19 April 2015^¶	12 February 2021^¶	6 December 2016^¶
Date of actual full approval	18 December 2015 (pembrolizumab)^‡	9 July 2021	30 March 2017
Average oncology conversion time^†	4 March 2017	17 June 2022	13 May 2018
Assumed delay in access to AA drug, months
Date of data presentation	7.6	14.1	13.0
Date of conversion to full approval	15.7	19.0	16.8
Average oncology conversion time	30.4	30.4	30.4

† Average time for an oncology product approved through the AA pathway to convert to traditional approval, in months, based on oncology products with completed confirmatory trials per Center for Drug Evaluation and Research data on accelerated approvals; available at: www.fda.gov/drugs/nda-and-bla-approvals/accelerated-approvals.

‡ For instances where there is more than one drug, we used the drug with the earliest approval dates.

§ Includes atezolizumab, nivolumab, durvalumab, avelumab and pembrolizumab.

¶ Pembrolizumab data presented at AACR (metastatic melanoma), enforumab vedotin-ejfv data presented at ASCO GU (advanced bladder cancer) and osimertinib data published in The New England Journal of Medicine and presented at World Conference on Lung Cancer.

1L: First-line; 2L+: Second-line and beyond; AA: Accelerated approval; AACR: American Association for Cancer Research; aNSCLC: Advanced non-small-cell lung cancer; ASCO: American Society of Clinical Oncology; EGFR+: Epidermal growth factor receptor positive; GU: Genitourinary.

Table 2. Real-world outcomes: proportion of patients receiving AA drugs and mean survival gains for patients.

	Metastatic melanoma	Advanced bladder cancer	EGFR+ aNSCLC
Proportion surviving and receiving AA drug
Total number of initial drug-treated patients	353	888	734
Proportion of patients surviving to AA date for future drug, %	49	29	64
Number of patients receiving future AA drug	79	65	158
Proportion of patients surviving and treated with future AA drug among all initial drug patients, %	22	7	22
Survival associated with AA drug
Mean survival of initial drug followed by AA drug, months	56.2	37.6	57.1
Mean survival of initial drug followed by SoC/no treatment, months^†	24.9	21.3	32.8
Difference in survival, months^‡	31.3	16.3	24.3
Average survival gain among all new drug patients, months^§	7.0	1.2	5.2

† Estimate reflects the weighted average survival between no treatment (cohort C) and SoC (cohort B), based on observed utilization patterns in Flatiron Health.

‡ Estimate is the net difference in survival between initial drug patients who survived to move onto a sequential therapy, expressed as total survival for patients who received the future AA drug (cohort A) minus the weighted average of patients who did not receive the AA drug (cohorts B and C).

§ Estimate reflects the probability of receipt for the future AA drug based on the survival curve for the initial drug and is therefore weighted (downwards) by the likelihood of surviving to the future AA drug.

AA: Accelerated approval; aNSCLC: Advanced non-small-cell lung cancer; EGFR+: Epidermal growth factor receptor positive; SoC: Standard of care.

Table 3. Population outcomes for AA pathway compared with counterfactual scenarios.

	Metastatic melanoma	Advanced bladder cancer	EGFR+ aNSCLC
Proportion surviving and receiving AA drug
Total number of initial drug-treated patients	4394	21,488	20,317
Proportion of patients surviving to AA date for future drug, %	40	31	62
Number of patients receiving future AA drug	745	1698	4233
Proportion of patients surviving and treated with future AA drug among all initial drug patients, %	17	8	21
Average survival gain among all new drug patients, months	5.3	1.3	5.1
Primary analysis: traditional approval = date of AA conversion
Number of additional patients receiving future AA drug owing to AA pathway	523	1247	2430
Average survival gain among all initial drug patients due to AA pathway, months^†	3.73	0.94	2.89
Percentage of survival gain among all initial drug patients attributable to AA pathway, %	70	73	57
Ratio of survival gain with AA pathway to traditional pathway	3.4	3.8	2.3
Population LYs gained with AA pathway	1367	1702	4930
Sensitivity analysis 1: shorter time to traditional approval
Number of additional patients receiving future AA drug owing to AA pathway	355	1034	1942
Average survival gain among all initial drug patients due to AA pathway, months^†	2.53	0.78	2.31
Percentage of survival gain among all initial drug patients attributable to AA pathway, %	48	61	46
Ratio of survival gain with AA pathway to traditional pathway	1.9	2.6	1.8
Population LYs gained with AA pathway	927	1411	3934
Sensitivity analysis 2: longer time to traditional approval
Number of additional patients receiving future AA drug owing to AA pathway	678	1571	3473
Average survival gain among all initial drug patients due to AA pathway, months^†	4.84	1.19	4.15
Percentage of survival gain among all initial drug patients attributable to AA pathway, %	91	92	82
Ratio of survival gain with AA pathway to traditional pathway	11.1	13.3	5.6
Population LYs gained with AA pathway	1772	2147	7075

† Estimate reflects the probability of receipt for the future AA drug based on the survival curve for the initial drug, and is therefore weighted (downwards) by the likelihood of surviving to the future AA drug.

AA: Accelerated approval; aNSCLC: Advanced non-small-cell lung cancer; EGFR+: Epidermal growth factor receptor positive; LY: Life-year.

Figure 2. Accelerated approval pathway outcomes.

(A) Average additional survival (in months) per patient with and without AA pathway by proximity to the AA date; (B) number of patients receiving AA drug with and without AA pathway by proximity to the AA date; and (C) LYs gained for initial drug patients with AA pathway by proximity to the AA date.

The calendar year prior to AA reflects the year of initial drug start relative to the date of the AA drug.

AA: Accelerated approval; EGFR+: Epidermal growth factor receptor positive; LYs: Life years; N/A: Not applicable; NSCLC: Non-small-cell lung cancer.