Impact of CFTR Modulators on Longitudinal Cystic Fibrosis Survival and Mortality: Review and Secondary Analysis

Abstract

Introduction

Cystic fibrosis (CF) transmembrane conductance regulator modulators (CFTRm) have transformed CF care, shifting treatment from only managing symptoms to also addressing the underlying defects that cause CF. CFTRm first entered clinical practice in 2012 and was followed by additional CFTRm combinations—including the approval of elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA) in 2019—which treats most CF genotypes.

Methods

We identified peer-reviewed literature for a narrative review (January 1990 to January 2025) describing longitudinal trends in CF survival and age of death and assessing the influence of CFTRm, particularly ELX/TEZ/IVA. To supplement the existing literature, a secondary analysis of historical, longitudinal trends in the United States CF Foundation Patient Registry (U.S. CFFPR, 1990–2023) was conducted using recent available data.

Results

Quantitative data from published studies show that the median age of survival and death increased over time but with varying magnitudes across regions. Most cohort and registry-based studies were conducted in settings where CFTRm were not yet widely available, limiting the evaluation of CFTRm effects on survival trends over time. In the secondary U.S. CFFPR analysis, the median survival age increased from 29.0 years in 1990 to 38.6 years in 2012 prior to the introduction of CFTRm and to 68.0 years in 2023, demonstrating substantial improvement following the introduction of CFTRm. Linear regression analyses showed gains in median survival age increased from 0.48 years per year prior to CFTRm to 4.79 years per year after approval of ELX/TEZ/IVA in 2019.

Conclusions

Study results provide initial evidence of the impact of CFTRm to meaningfully improve survival. Longer-term follow-up data across geographies will provide a deeper understanding of the full impact of CFTRm on predicted CF survival and mortality.

Plain Language Summary

Cystic fibrosis (CF) transmembrane conductance regulator modulator (CFTRm) treatments address the underlying defects that cause CF and were first introduced in the United States (U.S.) in 2012 to treat a small portion of the CF population with specific gene mutations. Additional CFTRm treatments were introduced over time that could treat roughly 90% of people with CF in the U.S. This study synthesized the literature from January 1990 to January 2025 to understand survival trends in CF over time, with the specific aim of evaluating the impact of CFTRm treatments. A secondary analysis of survival data from the United States Cystic Fibrosis Patient Registry was conducted to supplement the available literature. The literature suggests that CF survival has been increasing over the last five decades due to things like newborn screening programs and advancements in symptomatic treatments, with notable regional variations. After the introduction of CFTR modulators in 2012, a faster rate of increased survival was observed. The secondary analysis of US Cystic Fibrosis Foundation Patient Registry outcomes confirms the accelerated increase in survival after the advent of CFTRm, particularly elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA), which obtained US Food and Drug Administration approval in 2019. The rate of increase in the median survival age was tenfold higher after the introduction of ELX/TEZ/IVA compared to the period before the availability of any CFTRm.

Key Summary Points

People with cystic fibrosis (CF) have experienced steady gains in the median ages of survival and death since the first disease description, with a faster rate of increase since the introduction of transmembrane conductance regulator (CFTR) modulators in 2012.

The United States Cystic Fibrosis Foundation Patient Registry (USCFFPR) provides robust longitudinal survival data over the past 30 years; an analysis of the rate of improvement in median survival from the USCFFPR shows that survival has improved as much in the 5 years since the introduction of elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA) as it had over nearly three decades between 1990 and 2018.

As CFTR modulators become available to younger age groups, the median survival age trend observed is expected to continue to increase.

Introduction

Cystic fibrosis (CF) is a progressive, autosomal recessive genetic disease that adversely impacts health and decreases survival. Worldwide, there are roughly 111,000 diagnosed people with CF [1]. Mutations in the CF transmembrane conductance regulator (CFTR) gene cause CF by impairing CFTR protein synthesis. The disease affects multiple organ systems, including the lungs, pancreas, and gastrointestinal tract [2]. For people with CF, progressive lung disease is the leading cause of death [3].

Early CF therapies primarily addressed symptoms and end-organ findings, including poor nutrition, respiratory function, and lung injury and infection. First used in the 1950s, pancreatic enzyme replacement therapy (PERT) represented a major step in improving nutritional status and extending survival in CF, particularly among children [4, 5]. At the same time, aggressive nutritional interventions with high-caloric meals were implemented [4, 5]. Advancements in pulmonary treatments targeting pulmonary symptoms such as airway obstruction, inflammation, and infection followed, including improved airway clearance, inhaled dornase alpha (1993), inhaled tobramycin (1997), azithromycin (2003), hypertonic saline (2006), and inhaled aztreonam (2010) [5, 6]. Further advancements—including centralization of multidisciplinary care, prenatal screening, and newborn screening (NBS) programs—all improved health outcomes and survival among people with CF [3, 6]. Multiple countries spanning North America, Europe, and Oceania report a history of CF survival gains, including the United States (U.S.) [7–15]. National data from the U.S. CF Foundation Patient Registry (CFFPR) demonstrates increases in survival since the registry’s inception in 1986 [11, 16].

The introduction of CFTR modulator therapies (CFTRm) to medical practice starting in 2012 reshaped the CF treatment landscape, shifting the focus from only symptom management to also addressing the protein synthesis defects that cause CF. CFTRm are mutation-specific therapies designed to restore and improve CFTR protein synthesis and function [3, 17]. The U.S. Food and Drug Administration (FDA) initially approved ivacaftor (IVA), the first CFTR potentiator, in 2012 for people with CF with G551D mutations (4% of people with CF), and ultimately extended approval for treatment of other gating mutations, covering approximately 6% of people with CF [3, 18, 19]. The FDA subsequently approved four other CFTR corrector combination therapies, expanding treatment to the most common disease-causing mutations. Lumacaftor/ivacaftor (LUM/IVA) combination therapy for homozygous F508del-CFTR mutations received FDA approval in 2015, followed by the 2018 approval of tezacaftor/ivacaftor (TEZ/IVA) for people with CF with two copies of the F508del mutation or at least one responsive mutation, accounting for 45–50% of all people with CF [4, 20, 21]. CF treatment changed again with the 2019 FDA approval of elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA), a combination therapy with superior efficacy that also expanded use to people with CF with at least one F508del mutation and other responsive mutations, comprising up to 90% of people with CF [4, 11, 22–25]. Since its approval, ELX/TEZ/IVA has experienced rapid uptake among people with CF in countries with regulatory approval, with over 80% of people with CF eligible for CFTRm using ELX/TEZ/IVA in the U.S. in 2023 [11]. The newest approved modulator combination, vanzacaftor/tezacaftor/deutivacaftor, is just entering clinical practice in 2025.

CF comorbidities, including pulmonary dysfunction, poor nutritional status related to pancreatic insufficiency, and greater frequency of pulmonary exacerbations, increase mortality among people with CF [8, 26–30]. CFTRm improve pulmonary function and nutritional status and slow disease progression in clinical and real-world studies, resulting in an overall decrease in CF-related comorbidities. Improvements in these clinical outcomes may result in additional survival gains for people with CF treated with CFTRm. Indeed, several real-world observational studies with longer follow-up periods have demonstrated benefits in pulmonary and non-pulmonary outcomes, including observed reductions in mortality [31–37]. A recent U.S. CFFPR study (2010–2019) reported significant clinical benefits among those treated with IVA, including reduced mortality and lower rates of lung transplantation, pulmonary exacerbations, and all-cause hospitalizations relative to a comparison cohort of those with no prior CFTRm treatment [38].

This review critically evaluates the existing literature to better understand worldwide changes in CF survival age and age at death over the past several decades, and to assess the impact of CFTRm on these outcomes. Further, to supplement narrative review findings, a secondary data analysis was conducted to statistically examine longitudinal survival trends (1990–2023) among people with CF before and after the introduction of CFTRm in the U.S.

Methods

CF Survival and Mortality Narrative Literature Review

This comprehensive narrative literature review focused on longitudinal trends in CF survival and mortality using two key indicators: median survival age predicted at birth and median age observed at death. The median survival age is interpreted as the estimated predicted age beyond which 50% of people with CF are expected to live (not accounting for future advancements in care) [7, 10, 39–41]. Also important to consider is the median age at death—measured for individuals who died during the period under evaluation—[7, 10, 39–41], which provides additional context when providers communicate with patients and families. Mean values (as opposed to medians) were excluded, as they may reflect skewed distributions of CF deaths [40].

Peer-reviewed articles on longitudinal CF survival and mortality were identified via PubMed using the search terms “cystic fibrosis” and “survival” or “mortality” or “death” in January 2025. A narrative review of the literature published between January 1, 1990 and January 2, 2025 was conducted to identify historical and current data on median survival age and age at death inclusive of all age groups and geographic regions. Data are reported by cohort years and geography when available, and summarized longitudinally and geographically.

Historical U.S. CF Survival Trend Analysis

To further explore and quantify CF survival trends during the periods before and after the introduction of CFTRm, a secondary analysis was conducted using recent longitudinal data from the U.S. CFFPR. The U.S. CFFPR collects prospective data on consenting people with CF followed at more than 130 accredited U.S. care centers [11]. Median survival age is calculated by the U.S. CFFPR using life-table analysis [39]. Annual data on the median survival age for people with CF (1990–2021) was obtained from a figure presented at the 2022 North American CF Conference [42]. Survival data from the figure was digitized using the Engauge Digitizer to obtain quantitative estimates by year [43]. Separately, the median survival age for 2022 and 2023 was sourced from the U.S. CFFPR Annual Reports from 2022 and 2023 [11, 44]. The annual proportion of U.S. CFFPR participants taking ELZ/TEZ/IVA and other CFTRm was defined as the number with at least one reported prescription for CFTRm out of the total registry population for the given year [11, 44]. The 2023 annual report is the most recent year of data available at the time of this analysis.

Annual predicted median survival ages were categorized into three periods defined by CFTRm availability: (1) 1990–2011 (“pre-CFTR modulator era”), representing the years when CFTRm were not yet approved; (2) 2012–2018 (“IVA, LUM/IVA, TEZ/IVA era”), representing the period between the introduction of IVA (2012) [18], LUM/IVA (2015) [20], and TEZ/IVA (2018) [21], but prior to ELX/TEZ/IVA; and (3) 2019–2023 (“ELX/TEZ/IVA era”), representing all available U.S. CFFPR data post-FDA approval of ELX/TEZ/IVA in 2019 [22]. The latest CFTRm combination, vanzacaftor/tezacaftor/deutivacaftor, was approved in January of 2025, and was therefore not included in this analysis. Median survival age (available for all study years) was examined rather than rolling 5-year estimates (presented in U.S. CFFPR annual reports) to investigate changes in the ELX/TEZ/IVA era.

Three univariate linear regression models were fit separately by period (“pre-CFTR modulator era,” “IVA, LUM/IVA, TEZ/IVA era,” and “ELX/TEZ/IVA era”) to estimate the rate of change in median survival age per year, using year as the independent variable and median survival age as the dependent variable. Models were separated by period due to the non-linear nature of the relationship between time and median survival age over the entire study period. Linear regression assumptions were verified with diagnostic plots (e.g., residuals versus fitted values, standardized residuals versus fitted, and Q-Q plot). Beta-coefficients and 95% confidence intervals (CI) for each model are presented, representing the rate of change in median survival age per year.

Ethical Approval

This is a comprehensive narrative literature review and peer-reviewed articles were identified using PubMed. Additionally, a secondary analysis was conducted using recent longitudinal data from the U.S. CFFPR. The U.S. CFFPR collects prospective data on consenting people with CF. Internal Review Board (IRB) approval was not required.

Results

Literature Review: Median Survival Age Longitudinal Trends

Longitudinal median survival age among people with CF was reported in studies spanning the U.S. [45–50], Australia [51], Ireland [46], France [52], Canada [45, 53, 54], and Japan [55]; all studies reported increases over time (Table 1). Of these, eight studies presented quantitative longitudinal data for all people with CF in a geographically defined population (e.g., not limited to specific age groups, genotypes, or gender) [46–49, 51–53, 55] (Fig. 1). Studies suggest that geographic differences in CF survival may be driven by differential access to transplantation, increased post-transplant survival, varied healthcare delivery systems, and genotype variation [45, 56].

Table 1.

CF survival and mortality literature review—studies reporting longitudinal median survival age among people with CF, by country (n = 10 studies)

Source	Study years	Population description	Ages	Study no.	Sub-group	Age of survival median (95% CI)	Longitudinal trend direction
United States
FitzSimmons 1994 and FitzSimmons 1993 [47, 48]a, b	1969–1990	U.S. CF Foundation Patient Registry	All ages	7795	1969	14	↑
				17,857	1990	27.6
Kulich 2003 [50]	1985–1999	U.S. CF Foundation Patient Registry	All ages	31,012	1985, Males	34.5	↑
					1999, Males	37.1
					1985, Females	31.5
					1999, Females	34.9
Jackson 2011 [46] a	1986–2008	U.S. CF Foundation Patient Registry patients with death records	All ages	8849	1986	26.7 (25.3–28.3)	↑
					2008	37.4 (35.0–40.1)
Stephenson 2017 [45] c	1990–2013	U.S. CF Foundation Patient Registry	All ages	32,699	2009–2013, Includes post-transplant follow-up	40.6 (39.1–41.8)	↑ (data in figure only)
					2009–2013, Censoring at transplant	44.0
Ostrenga 2023 [49] a, d	1986–2017	U.S. CF Foundation Patient Registry linked to the National Death Index	All ages	30,788	1986–1990	29.7	↑
					2013–2017	45.8
Canada
Corey 1996 [54]	1970–1989	Canadian Patient Data Registry	All ages	3795	1970–1989, Males	29.5	↑
					1970–1974, Males	26.6
					1985–1989, Males	36.7
					1970–1989, Females	21.3
					1970–1974, Females	19.7
					1985–1989, Females	27.8
Stephenson 2015 [53] c	1990–2012	Canadian CF Registry	All ages	5787	1990	31.9 (28.3–35.2)	↑
					2008–2012	49.7 (46.1–52.2)
Stephenson 2017 [45] c	1990–2013	Canadian CF Registry	All ages	4662	2009–2013, Includes post-transplant follow-up	50.9 (50.5–52.2)	↑ (data in figure only)
					2009–2013, Censoring at transplant	57.1
France
Bellis 2007 [52] a	1994–2003	Observatoire National de la Mucoviscidose in mainland France and Reunion Island	All ages	n/s	1994–1996	28.1	↑
					2001–2003	36.4
Ireland
Jackson 2011 [46] a	1986–2008	CF Registry of Ireland patients with death records	All ages	421	1986	20.1 (13.0–34.4)	↑
					2008	35.2 (24.5-N/A)
Australia
Ruseckaite 2022 [51] a	2005–2020	Australian CF Data Registry	All ages	4601	2005–2009	48.9 (44.7–53.5)	↑
					2016–2020	56.3 (53.0–60.4)
Japan
Kozawa 2023 [55] b	1994–2022	National epidemiological survey and Japan CF Registry	All ages	132	2012	18.8 (16.6–21.1)	↑
					2022	25.2 (18.7–31.7)

CF cystic fibrosis, CI confidence interval, N/A not available, N/S not stated, U.S. United States; ↑ = increasing over time; ↓ = decreasing over time

^aData also reported for additional years; the first and last years of data available are presented in this table

^bMedian survival age also reported by gender (males, females)

^cData also reported for additional years in figure only

^dMedian survival age also reported for CFFPR standard reporting with no linkage to NDI

Fig. 1.

CF survival literature review – longitudinal median survival age (years) among people with CF (n = 8 studies). CF cystic fibrosis, JCF Journal of Cystic Fibrosis, U.S. United States. Note: Studies (n = 1) only reporting a comparison of two overlapping study periods (i.e., 1990–2013 vs. 2009–2013 presented in Stephenson et al. [45]) were excluded from this figure. Studies (n = 2) reporting data separately for males and females only were excluded [50, 54]. For studies reporting median survival age over a 3- [52] or 5-year [49, 51, 53] period, data in the above figure are reported as the most recent year in the window (e.g., for 2005–2009, reported as 2009)

Increases in median survival were observed in North American countries, including the U.S. and Canada. The U.S. CFFPR reported survival gains with median survival age increasing from 14.0 years in 1969 to 27.6 years in 1990 [47, 48]. From 1986 to 2008, median survival rose in the U.S. from 26.7 years (95% CI 25.3–28.3) to 37.4 years (95% CI 35.0–40.1) [46]. Additional U.S. CFFPR data linked to National Death Index records (1986–2017) reported an increase in median survival age from 29.7 years in 1986–1990 to 45.8 years in 2013–2017 [49]. While this analysis largely focused on quantifying mortality among individuals lost to follow-up in the U.S. CFFPR, the authors noted survival gains over the 30-year study period were attributed to expanded, interdisciplinary CF care (1980s), use of inhaled antibiotics (mid-1990s to 2000s), and the introduction of CFTRm (2012) [49]. Over a two-decade period in Canada, the median survival age also increased significantly from 31.9 years (95% CI 28.3–35.2) in 1990 to 49.7 years (95% CI 46.1–52.2) in 2012 [53].

Parallel increases in median survival age were observed in Australia. In a 2022 study from the Australian CF Data Registry, median survival age increased over a 15-year period from 48.9 years (95% CI 44.7–53.5) for people with CF born in 2005–2009 to 56.3 years (95% CI 51.2–60.4) for those born 2016–2020 [51]. These improvements in median survival age in Australia likely reflect several factors, including diagnostic advancements, NBS, nutritional interventions, expanded disease surveillance, access to novel therapies, and enhanced management and care of people with CF [51]. In Australia, NBS started in 1981 and became universal in 2001, which may explain recently observed survival increases [51]. The authors caution it is too soon to see the impact of CFTRm on survival, given IVA was first introduced in Australia in December 2014 for a small number of patients and only later expanded to include additional CFTRm that were effective for a wider group of people with CF [51]. Given these advancements in treatment, the median survival age is expected to continue improving [51].

Recent data from Japan found increases in the median survival age of people with CF from 18.8 years (95% CI 16.6–21.1) in 2012 to 25.2 years (95% CI 18.7–31.7) in 2022; reported survival is still low relative to CF populations in North America and Europe, though the sample size is small (n = 132) and CF is likely underdiagnosed in Japan [55].

Older data extending to the 2000s (before the introduction of CFTRm) were also reported in Europe. In Ireland, median survival age for people with CF ranged from 20.1 years (95% CI 13.0–34.4) in 1986 to 35.2 years (95% CI 24.5–N/A) in 2008 [46]. Similar gains were observed in France, with median survival improving substantially from 28.1 years in 1994–1996 to 36.4 years in 2001–2003 [52].

Literature Review: Median Age at Death Longitudinal Trends

Median age at death depends on the population age distribution at the time of measurement [40]. Median survival age for a given period is often higher than median age at death, as median survival age is projected for people with CF born during the current period with access to the latest efficacious treatments. However, median age at death is computed for a cohort of people with CF that died during the same period, many of whom lived through an earlier era with fewer efficacious treatments available [40]. Further, differential access to CF diagnostic testing (which can lead to missed CF cases) and cause of death misclassification may also result in biased estimates, particularly among retrospective death certificate studies. This is exemplified in a death record review conducted in Brazil; although a striking increase in the median age at death was reported between 1999 and 2017 (7.5 years; IQR: 0.5–24.5 to 56.5 years; IQR: 18.5–74.5), authors noted that results could be skewed by misdiagnosis or misreporting [57].

Similar to median survival age, median age at death among people with CF who died during the study period varied widely across geographies and time. Progressive increases in median age at death were reported across multiple countries including the U.S., Canada, England, France, Germany, Ireland, Italy, Spain, Sweden, Wales, and Brazil [45, 52, 53, 57–67]; a second Brazilian study did not report an increase, though ranges were wide due to the small number of deaths per year [68] (Table 2). Eleven studies presented quantitative longitudinal data for people with CF of all ages across non-overlapping periods [52, 53, 57–61, 64, 66–68]; findings are summarized in Fig. 2.

Table 2.

CF survival and mortality literature review—studies reporting longitudinal median age at death among people with CF, by country (n = 15 studies)

Source	Study years	Population description	Ages	Study no.	Sub-group	Age at death Median (IQR)	Longitudinal trend direction
United States
Halliburton 1996 [67] b	1979–1991	CF deaths identified via the National Center for Health Statistics	All ages	6500	1979	15	↑
					1991	23
Hurley 2014 [62] b	1972–2009	CF deaths identified via CDC WONDER	All ages	13,239	2009	23.4 (16.4–27.6)	↑ (data in figure only)
Stephenson 2017 [45] b	1990–2013	U.S. CF Foundation Patient Registry	All ages	45,448	1990–2013	24.9 (range 0.0–81.4)	↑
				32,699	2009–2013	26.9 (range 0.3–76.9)
Singh 2023 [58] a, c	1999–2020	Death certificate review of CDC WONDER	All ages	11,068	1999	24 (18–33)	↑
					2020	37 (25–55)
Canada
Stephenson 2015 [53] b	1990–2012	Canadian CF Registry	All ages	5787	1990	21.7	↑
					2012	32
Stephenson 2017 [45] b	1990–2013	Canadian CF Registry	All ages	5941	1990–2013	26.6 (range 0.0–79.3)	↑
				4662	2009–2013	31.9 (range 0.3–79.3)
Mexico
Bustamante 2021 [60] d	2000–2020	People with CF registered and followed at the CF Center at University Hospital, Monterrey	All ages	205	2000–2020	12.5 (range 0.1–43.7)	↑
					2000–2005	9.6 (range 0.1–30.6)
					2016–2020	15.5 (Range: 0.4–39.3)
Brazil
Santo 2021 [57] a, e	1999–2017	CF deaths identified in the Brazilian National Ministry of Health Mortality Database	All ages	2854	1999–2017, underlying cause	23.5 (7.5–65.5)	↑
					1999–2017, non-underlying cause	15.5 (0.3–69.5)
				68	1999	7.5 (0.5–24.5)
				289	2017	56.5 (18.5–74.5)
de Azevedo 2023 [68] a	2009–2018	Brazilian CF Register	All ages	1044	2009	21.4 (range 6.3–80.0)	↓
				3359	2018	18.4 (range 0.3–40.6)
England and Wales
Hurley 2014 [62] b	1972–2009	CF deaths identified via the Office of National Statistics for England and Wales	All ages	4622	2009	22.1 (16.6–32.0)	↑ (data in figure only)
France
Bellis 2007 [52] a	1994–2003	Observatoire National de la Mucoviscidose in mainland France and Reunion Island	All ages	2168	1994	14.5	↑
				4104	2003	22.0
Germany
Stern 2008 [64]	1995–2005	German CF Quality Assessment project	All ages	6835	1995	18.5	↑
					2005	23.7
Italy
Alicandro 2015 [61] a, e	1970–2011	CF deaths identified in the Italian National Institute of Statistics	< 65 years	338	1970–1975	0.5 (0.1–4.0)	↑
				293	2006–2011	29.0 (21.0–38.0)
Spain
Ramalle-Gomara 2008 [63]	1981–2004	CF deaths for patients < 30 years identified from the Instituto Nacional de Estadıstica	< 30 years	528	1999–2004, Males	24 (range 0.0–28.9)	↑
					1999–2004, Females	19 (range 0.0–29.3)
					1981, Males	4.4
					2004, Males	20.1
					1981, Females	3.8
					2004, Females	17.7
Sweden
Lannefors 2002 [65] b	1971–1999	People with CF across Sweden identified via clinical and mortality record review	All ages	628	1991–1998	26 (range 0–72)	↑ (data in figure only)
South Africa
Zampoli 2022 [59] a	1974–2019	Children with CF diagnosed and managed at the Red Cross War Memorial Children's Hospital	Pediatric	288	Before 1990 (1974–1990)	5.5 (0.5–7.6)	↑
					After 2010 (2010–2019)	22.5 (19.7–25.2)
Multi-Country (11 countries)
Fogarty 2000 [66] b, f	1974–1994	CF deaths identified via national death records	n/s	18,182	1974	8	↑
					1994	21

CDC U.S. Centers for Disease Control and Prevention, CF cystic fibrosis, IQR interquartile range, N/S not stated, U.S. United States; ↑ = increasing over time; ↓ = decreasing over time.

^aData also reported for additional years; the first and last years of data available are presented in this table

^bData also reported for additional years in figure only

^cMedian age at death also reported by gender (males, females), race (white, others), and ethnicity (non-Hispanic, Hispanic)

^dMedian age of death also reported by gender (males, females), SES (high, low), pancreatic insufficiency (yes, no), chronic Pseudomonas aeruginosa infection at 6 years (yes, no), and genotype (homozygous F508del, heterozygous F508del, other genotype, missing)

^eMedian age at death also reported by gender (males, females)

^fCountries include Australia, Belgium, England and Wales, France, Germany, the Netherlands, Ireland, New Zealand, Scotland, Sweden, and the U.S.

Fig. 2.

CF mortality literature review—longitudinal median age at death (years) among people with CF (n = 11 studies). CF cystic fibrosis, U.S. United States. Note: Studies (n = 2) with longitudinal data reported in figures only were excluded from this figure [62, 65]. Studies (n = 1) only reporting a comparison of two overlapping study periods (i.e., 1990–2013 vs. 2009–2013 presented in Stephenson et al. [45]) were excluded from this figure. Studies (n = 1) limited to pediatric populations only were excluded [63]. For studies reporting median age at death over a period [59–61], data in the above figure are reported as the most recent year in the window (e.g., for 2000–2005, reported as 2005)

Two U.S.-based studies examining longitudinal trends were conducted in populations with access to CFTRm, [45, 58] though one study ending in 2013 did not allow sufficient follow-up time post-IVA approval in 2012 [45]. An analysis of U.S. data from the Centers for Disease Control and Prevention (CDC) WONDER death records found an increase in median age at death from 24 years (IQR 18–33) in 1999 to 26 years (IQR 21–37) in 2010, 30 years (IQR 22–42) in 2015, and 37 years (IQR 25–55) in 2020 [58]. Notably, the median age at death increased at a faster rate in more recent years following the approval of CFTRm in the U.S., suggestive of the impact of CFTRm on mortality [58]. However, more recent data that reflect the availability of ELZ/TEZ/IVA, as well as a formal comparison of treated versus non-treated populations are needed to fully support the relationship [58].

Historical U.S. CF Survival Trend Analysis

Longitudinal survival gains were further demonstrated in a secondary analysis of U.S. CFFPR data. From 1990 to 2023, the median survival age for people with CF increased from 29.0 to 68.0 years (39.0-year increase; Fig. 3). In the pre-CFTRm era (1990–2011), the median survival age increased steadily from 29.0 years for people with CF born in 1990 to 38.0 years in 2011 (9.0-year increase). During the IVA, LUM/IVA, TEZ/IVA era (2012–2018), the median survival age increased from 38.6 years in 2012 to 47.5 years in 2018 (8.9-year increase). Following the FDA approval of ELX/TEZ/IVA in 2019, median survival age increased substantially from 48.5 to 68.0 years (19.5-year increase) over the 5-year period. Simultaneously, by 2023, approximately 70% of the total registry participants were using ELX/TEZ/IVA [11].

Fig. 3.

Historical U.S. survival trend analysis—median survival age by year (1990–2023) and CFTR modulator usage among people with CF in the U.S. CF cystic fibrosis, ELX elexacaftor, LUM lumacaftor, IVA ivacaftor, TEZ tezacaftor, U.S. United States. Note: Data on median survival age (years) and CFTR modulator usage (available for 2014–2023) were sourced from the U.S. CFFPR [11, 42, 44]. Depicted trend lines are based on output from the linear regression analysis

Univariate linear regression analyses indicate that the annual rate of change in median survival age increased from 0.48 per year (95% CI 0.43, 0.53) prior to CFTRm (1990–2011) to 1.72 per year (95% CI 1.10, 2.35) in the IVA, LUM/IVA, and TEZ/IVA era (2012–2018). In the ELX/TEZ/IVA (2019–2023) era, the annual rate of change in median survival age increased substantially to 4.79 per year (95% CI 2.29, 7.29). The rate of increase tripled from the pre CFTRm era to the IVA, LUM/IVA, and TEZ/IVA era, and increased tenfold in the ELX/TEZ/IVA era.

Discussion

Evidence from peer-reviewed literature and a secondary analysis of recent U.S. CFFPR data show survival gains among people with CF over the last five decades. Improving survival accelerated since the introduction of CFTRm in 2012, increasing substantially again after the U.S. FDA approval of ELX/TEZ/IVA in 2019. While these findings alone do not confirm a causal relationship, the timing of accelerating survival—which align with the availability of CFTRm to treat the underlying cause of CF—suggest that CFTRm (particularly ELX/TEZ/IVA) extend life.

The U.S.-based historical trend analysis of CFFPR data demonstrated steady improvements in the median survival age prior to the introduction of CFTRm in 2012. This is consistent with literature reporting increasing median survival age [45–49, 51–55] and median age at death [45, 52, 53, 57–67], as well as decreasing mortality rates [47, 48, 52, 53, 58, 65, 69, 70] across countries spanning North America, South America, Europe, Asia, and Oceania over the last several decades. Variability between countries can be attributed to differences in treatment recommendations, diagnosis timing, and healthcare infrastructure [9, 10].

The advent of CFTRm builds upon a long history of monumental therapeutic advances since the discovery of CF in 1938, a time when children with CF died in infancy [5]. Early survival gains were driven by the introduction of PERT to address nutritional deficiencies in the 1950s [4, 5]. This transformed CF from a life-limiting gastrointestinal disease causing failure-to-thrive at a young age to a life-limiting lung disease, and was followed by new approaches to pulmonary treatment, including antibiotics for lung infections and addressing airway obstruction [3, 5, 7, 71–73]. The simultaneous shift towards standardized care at centralized treatment centers and improvements in genetic diagnostic testing (i.e., prenatal and newborn screening) was also associated with better outcomes and survival [6, 74, 75].

In the U.S., CF care centers were accredited beginning in the 1960s, bolstering the availability of standardized, multidisciplinary care [6]. Starting in 2002, the CF Foundation launched a U.S.-based healthcare delivery quality improvement initiative to accelerate advancements in care, including through implementation of clinical practice guidelines, leadership development, and increased patient engagement [6]. The period following these initiatives saw consistent increased use of nutritional interventions and chronic therapies—including dornase alpha, azithromycin, aztreonam, and hypertonic saline—among eligible people with CF, accompanied by improved nutritional and pulmonary outcomes [6]. A national mandatory NBS program for CF was expanded to all 50 states by 2010, originally established in 1982 in Colorado and completed for the last 45 states between 2005 and 2010 [76, 77]. National NBS was associated with improved nutritional status, a more rapid increase in lung function, and delayed chronic P. aeruginosa infection [78]. Although no studies were identified that directly assessed the impact of the national mandatory U.S. NBS program on survival, several studies in other populations found associations between NBS programs and survival improvements [79–82]. Two additional studies found no significant survival differences, though study periods were older with relatively small samples [83, 84].

Survival gains in the period following the introduction of CFTRm in the historical analysis are consistent with results from a real-world, post-approval observational safety study comparing the risk of death among IVA-treated versus untreated matched comparator people with CF in the U.S. CFFPR and U.K. CF Registry [85]. IVA-treated people with CF had a significant 59% reduction in the risk of death in the U.S and a non-significant 48% reduction in the U.K. [85]. Results were consistent in the follow-up publication with up to 5 years of additional data [86]. More recent U.S. CFFPR data (2010–2019) with up to 7.9 years of follow-up demonstrated IVA improved survival (HR 0.22; 95% CI 0.09–0.45) and nutritional status, preserved lung function, and reduced pulmonary exacerbations and all-cause hospitalizations relative to an untreated comparison cohort [38].

Similar to studies in IVA, observational studies in patients treated with ELX/TEZ/IVA conducted in the U.S. support the population trends identified in this research. An observational, post-approval safety study of ELX/TEZ/IVA within the U.S. CFFPR found the annualized rate of death among those initiating therapy between 2019 and 2020 was 72% lower compared to a 2019 historic CFFPR cohort (annualized rate of death: 0.47% for ELX/TEZ/IVA vs 1.65% for historical controls) [34]. Additional real-world observational studies demonstrated benefits in other pulmonary and non-pulmonary outcomes [17, 31–33, 35–37, 87], including reduced numbers of pulmonary exacerbations, which predict improved survival [29, 30]. Moreover, a recent study presented at a CF congress strengthens findings from previous studies, demonstrating that people with CF in the U.S. treated with ELX/TEZ/IVA had a 79% (95% CI 56–88%) lower risk of death versus a contemporaneous group of people with CF with genotypes not eligible for CFTRm [88]. This finding indicates the benefits of ELX/TEZ/IVA treatment on mortality are robust to potential confounding from protective effects of COVID-19 [88].

Further, modeling studies—valuable in situations where long-term follow-up data are limited—project that treatment with CFTRm will increase median survival in the U.S., Canada, and U.K. populations [89–95]. A U.S.-based study comparing ELX/TEZ/IVA combination therapy plus best supportive care versus best supportive care alone predicted an increase in median survival of 29.7 years for F/minimal function (MF) genotypes (70.4 vs. 40.8 years) [90]. In a related analysis based on the U.K. population, the median projected survival for homozygous F508del-CFTR people with CF treated with ELX/TEZ/IVA combination therapy was 23.2 years longer than TEZ/IVA (71.6 vs. 48.5 years), 26.2 years higher than LUM/IVA (71.6 vs. 45.4 years), and 33.5 years longer than basic standard care alone (71.6 vs. 38.1 years) [91]. Further, a Canadian-based modeling study projected median survival age by 2030 would be 9.2 years higher, assuming all eligible patients started ELX/TEZ/IVA combination therapy in 2021 compared to a scenario with no new therapies available (67.5 vs. 58.4 years) [92]. Survival gains were also projected in modeling studies examining other CFTRm in the U.S., including LUM/IVA combination therapy [89, 94] and IVA alone [93].

The current analysis demonstrates that the annual rate of improvement in median survival age following the U.S. FDA approval of ELX/TEZ/IVA increased sizably compared to the period when only IVA, LUM/IVA, and TEZ/IVA were available (4.79 years vs. 1.72 years, respectively). The analysis suggests that survival has improved as much in the 5 years since the introduction of ELX/TEZ/IVA as it had over nearly three decades between 1990 and 2018. Although years of data are limited, this large survival increase likely reflects the superior efficacy of ELX/TEZ/IVA, as well as the expanded eligibility to people with CF with at least one F508del mutation and other responsive mutations, representing up to 90% of people with CF [4, 22–24, 44].

Increases in survival in the US CFFPR are consistent with the rapid uptake of ELX/TEZ/IVA observed in the registry (over 80% of CFTRm eligible patients in 2023), and also the reported longitudinal improvements in lung function and reduced use of IV antibiotics to treat pulmonary exacerbations [11], both known predictors of mortality [29, 30]. During the last years of the study period (2019–2023), ELX/TEZ/IVA treatment was predominantly restricted to people with CF aged 12 years and older, with availability for children aged 6–11 years in the second half of 2021 [22] and children aged 2–5 years in April 2023 [11]. The median survival age trend observed in the ELX/TEZ/IVA era is expected to continue to increase as ELX/TEZ/IVA is available to treat younger people with CF.

Limitations

With respect to the comprehensive CF survival and mortality literature review, several limitations should be considered when comparing findings across studies. Differences in study populations (e.g., inclusion vs. exclusion of transplant patients), time periods, and geographies, including differences by clinical phenotype, healthcare systems, and access to transplantation and other therapies, impact comparability of estimates [40, 45]. Moreover, differences in the calculation methods of key survival metrics, particularly median survival age (e.g., period approach, birth cohort approach), present challenges in the standardization and harmonization of data [10, 40, 41]. Further, given median age at death is calculated based on a cohort of people with CF without access to the latest treatments during their lifetime, median age at death tends to lag overall changes in survival, requiring more time to observe the impact of more recent treatments on these metrics [40].

Limitations should also be noted for the U.S.-based historical trend analysis. First, demonstrating a direct association between CFTRm and survival improvements is difficult given the reliance on population-level data and limited long-term follow-up. Second, while the analysis presents evidence during the periods of CFTRm availability in the U.S., it does not account for confounding by other treatment improvements that may impact survival. However, the period during which CFTRm experienced rapid, widespread uptake aligns with observed survival improvements; as such, any residual confounding is unlikely to fully explain survival differences over time. Third, this analysis uses annual measurements of median survival age to characterize three periods by CFTRm availability. While measurement of median survival age over 5-year periods may be preferrable to smooth out year-to-year fluctuations due to low death counts per year, the results still demonstrate an increase in median survival age when comparing 2018–2022 vs. 2019–2023 (U.S. CFFPR) [11] and decline in mortality from 2020 to 2021 (U.S. CDC WONDER) [58], consistent with the arrival of ELX/TEZ/IVA. Fourth, digitizing software was used to extract estimates of survival from US CFFPR reports. The accuracy of the data extraction is dependent on the quality of the original graph (e.g., resolution and axis labels). While the data extraction may have impacted the numerical estimates of survival for each year, it likely had minimal impact on the overall trend due to the high-resolution of the image and the clarity of the axis labels. Fifth, estimates for the impact of ELX/TEZ/IVA on survival may be conservative, as the definition of the ELX/TEZ/IVA era included a 10-month period before widespread availability.

Conclusions

Over the past decade, CFTRm—most notably ELX/TEZ/IVA—have advanced CF treatment, with potential to improve morbidity, mortality, and survival. Causality cannot be proven with the available data, but this study demonstrates that the timing of more recent, faster improvements in survival coincide with the introduction of CFTRm (particularly ELX/TEZ/IVA), suggesting CFTRm are the most plausible source of observed improvements in survival and mortality. Advancements in the development and approval of CFTRm therapies that further improve CFTR function above ELX/TEZ/IVA reinforce the importance of studying survival trends in CF.

Author Contributions

Jaime L. Rubin, Craig McKinnon, and Gabriel Ghizzi Pedra were involved in the design of the study. Gabriel Ghizzi Pedra, Devon A. Morgan and Kimberly Zweig collected or generated the data. Jaime L. Rubin, Craig McKinnon, Gabriel Ghizzi Pedra, Devon A. Morgan, Kimberly Zweig, and Theodore G. Liou analyzed and interpreted the data. All authors participated in the development of this manuscript and its critical review with important intellectual contributions. All authors had full access to the data and gave approval of the final manuscript before submission. All authors agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

This study was funded by Vertex Pharmaceuticals Incorporated. The journal’s Rapid Service Fee will be paid by this company.

Data Availability

Peer-reviewed articles available on PubMed.gov. U.S. CFFPR data from published reports are available from https://www.cff.org.

Declarations

Conflict of Interest

Jaime L. Rubin, Craig McKinnon and Gabriel Ghizzi Pedra are employees of Vertex Pharmaceuticals Incorporated and may own stocks/stock options in the company. Devon A. Morgan and Kimberly Zweig have received consultancy fees from Vertex Pharmaceuticals Incorporated. Theodore G. Liou is an employee of the University of Utah which received funding from Vertex for this project. Theodore G. Liou is supported by the NIH/NHLBI (R01 HL125520), the Cystic Fibrosis Foundation, Bethesda, MD (grants CC-132-16AD, LIOU13A0, LIOU14Y0, and LIOU14Y4), the Ben B. and Iris M. Margolis Family Foundation of Utah, and the Claudia Ruth Goodrich Stevens Endowment Fund.

Ethical Approval

This is a comprehensive narrative literature review and peer-reviewed articles were identified using PubMed. Additionally, a secondary analysis was conducted using recent longitudinal data from the U.S. CFFPR. The U.S. CFFPR collects prospective data on consenting people with CF. Internal Review Board (IRB) approval was not required.