FormalPara Key Points for Decision Makers

From 2005 to 2023, US cancer drug prices increased faster than inflation.

Launch prices and post-launch price increases are not aligned with the clinical benefit a drug offers to patients.

In particular, greater price increases were observed for orphan drugs. Therefore, the Inflation Reduction Act of 2022 should be extended to orphan drugs to ensure that patients with rare diseases can access their treatment.

There is no evidence that brand-brand competition results in drug price reductions.

1 Introduction

From 2020 to 2021, US launch prices for half of all new medicines approved by the US FDA exceeded $150,000 per year [1]. In particular, high prices were observed for oncology drugs, with 95% of new anticancer drugs in the US priced beyond $100,000 per year in 2023 [2]. For cancer patients in the US, who typically bear 20–30% of treatment costs out of pocket (OOP), these high prices are a major cause of financial distress and financial toxicity [3]. High drug prices contribute to catastrophic healthcare expenditure, which ultimately leads to personal bankruptcy [4]. In particular, cancer patients are at 2.7 times greater risk of personal bankruptcy than non-cancer patients in the US [5]. This financial toxicity results in non-adherence to recommended treatment regimens [6] and therefore higher mortality rates [7].

High prescription drug prices are not only caused by high launch prices. Contributing to the economic burden of prescription drug costs are post-launch price changes. Price changes exceeding inflation were identified as a major contributor to rising treatment costs and patients’ OOP, particularly for cancer drugs in the US [8,9,10]. For instance, the annual price for the tyrosine kinase inhibitor (TKI) imatinib, a scientific breakthrough for patients with chronic myeloid leukemia, more than tripled in price from $30,000 to $92,000 in merely 10 years [3]. This price increase occurred despite the introduction of new second-generation TKIs, such as dasatinib and nilotinib. Although these competitors are similar in their mechanism of action and treat similar diseases, they were commercialized for launch prices beyond $110,000 per year.

Previous studies analyzed the association between launch prices and research and development (R&D) costs, competition, drug safety and efficacy, disease incidence and burden, and special FDA review pathways [2, 11,12,13,14,15,16,17,18,19,20,21,22,23]. Further studies investigated the correlation between post-launch price changes and new competitors, new indication approvals, new off-label uses, and safety and efficacy measures [8, 15, 24,25,26,27,28,29,30,31]; however, these studies are limited in their sample size, analyzed time horizon, and statistical analysis. Therefore, in this longitudinal study, we identify and quantify factors associated with launch prices and post-launch price changes of injectable cancer drugs. We evaluate the association between launch/post-launch prices and time-dependent and time-independent variables, characterizing each drug’s innovativeness, efficacy, disease epidemiology, and competition.

2 Data and Methods

2.1 Sample Identification

We identified all new drugs that received FDA approval between 1 January 2000 and 1 January 2022. The sample was then restricted to include only anticancer medicines, excluding those for supportive cancer care, diagnostic agents, and antiemetics. For these anticancer drugs, we identified all original and supplemental anticancer indication approvals until 1 January 2022. In our analyses, we only included drugs covered under Medicare Part B, given that no longitudinal price data are available for drugs covered under Medicare Part D from the Centers for Medicare and Medicaid Services (CMS). In general, Medicare Part B covers injectable cancer drugs (typically drugs that are administered at a hospital or doctor’s office), while Medicare Part D covers oral cancer drugs (typically self-administered drugs). Certain drugs with multiple routes of administration are covered by Medicare Part B and D, for example, everolimus.

2.2 Data Collection

For all identified cancer agents, we collected price data and information characterizing each drug’s characteristics, disease epidemiology, and market dynamics (electronic supplementary material [ESM] Table e1).

2.2.1 Drug Characteristics

First, we characterized each drug’s innovativeness. Two reviewers assessed the novelty of the underlying drug target based on the World Health Organization’s Anatomical Therapeutic Chemical code. Drugs with novel targets were considered first-in-class, whereas those with known targets were considered next-in-class. Second, the University of Alabama’s drug database, ‘Drug Bank’, was accessed to determine each drug’s product type. Drugs were categorized as small molecules and others, which entail biological agents, antibody-drug conjugates, gene therapies, cell therapies, enzymes, and radionuclides. Third, we obtained information on the approval of companion biomarkers from FDA labels. Finally, we obtained data from FDA websites to determine if special FDA designations were associated with each drug, e.g. orphan designation, accelerated approval, fast track, priority review, and breakthrough therapy designation [32].

2.2.2 Clinical Benefit

The clinical benefit of new cancer drugs was measured by their benefit in overall survival (OS), progression-free survival (PFS), and tumor response rates. We accessed FDA labels and ClinicalTrials.gov to collect data on OS and PFS hazard ratios and tumor response rates from randomized controlled trials (RCTs). Furthermore, we extracted median improvements in OS, PFS, and duration of response. The absolute median improvement in OS/PFS was calculated as the difference between median OS/PFS in the treatment relative to the control arm. The percentage improvement in OS/PFS was then calculated as the quotient of the absolute median OS/PFS improvement relative to the absolute median OS/PFS in the control arm. Although multiple analyses evaluated the association between the clinical benefit and launch prices of new drugs [2, 11,12,13,14, 16,17,18, 22], evidence scrutinizing the association between the clinical benefit and post-launch price changes of new drugs remains scarce [8]. Most European countries introduced regulations to limit drug price increases exceeding inflation and even reduce drug prices to control expenditure on new drugs. For instance, drug price increases are re-evaluated in Switzerland every 3 years, controlled by the government in England, and re-evaluated for drugs with new indications in Germany and France [33,34,35]. Given that over the study period there was no value-based pricing policy in the US that regulates launch prices and post-launch price changes, we hypothesized that there is no association between launch/post-launch prices and the drugs’ clinical benefit.

2.2.3 Disease Epidemiology

We obtained epidemiologic data for the US population in 2019 from the Global Burden of Disease study to describe the disease treated by each drug [36]. First, we collected disease incidence rates (per 100,000 US inhabitants) as a measure of disease rarity, and second, we collected disability-adjusted life-years (DALYs) per person as a measure of disease burden. DALYs are calculated as the sum of years lived with disability (YLD) and years of life lost (YLL). Therefore, DALYs not only capture the forgone lifetime but also the reduced quality of life that is caused by diseases. The disease-specific epidemiologic data were matched to each drug according to the treated disease specified in FDA labels.

2.2.4 Market Dynamics

Market dynamics were captured in two variables. We tracked the FDA approval of new supplemental indications for each drug. Given that these supplemental indications are often for non-orphan diseases supported by robust clinical trials with a relatively low clinical benefit (‘low-value indications’), we expect drug prices to decline following the introduction of new indications for the same drug [33, 37, 38].

We then monitored the number of new competitors entering the market for each drug. We used two alternative measures of new competitors. First, we counted the number of new cancer drug indications receiving FDA approval within the same disease during each quarter. This represents a broad measure of competition in the market of anticancer drugs (variable: new competitors [broad]). Second, we counted the number of new anticancer drug indications receiving FDA approval within the same disease, in the same line of therapy, for the same treatment setting, with the same biomarker during each quarter. This represents a narrow measure of competition in the market of anticancer drugs (variable: new competitors [narrow]). The narrow measure of competition might be more reflective of the underlying market dynamics in the cancer drug market, given that each drug and indication often fills a distinct therapeutic niche that is defined by the therapeutic setting (neoadjuvant vs. adjuvant vs. metastatic), line of therapy (first-line vs. second-line vs. advanced-line), and biomarker profile (for example, differentiated by driver mutations for non-small cell lung cancer (NSCLC): KRAS vs. EGFR vs. ALK vs. BRAF vs. MET vs. ROS1 vs. HER2 [ERBB2] vs. NTRK). For both measures of competition, we included the market entry of all new drug indications with FDA approval regardless of insurance states, e.g. we included drug indications covered under Part B and D.

2.2.5 Drug Prices

Drug prices were calculated according to a methodology that has been described in prior articles [2, 16, 28, 39]. First, we accessed the CMS' quarterly average sales price (ASP) data files to obtain drug pricing data from 2005 to 2023. For each drug, we then calculated monthly treatment costs based on the dosing regimen defined in FDA labels for the average US patient with a body weight of 70 kg and a body surface area of 1.7 m2 [2, 14, 16, 17, 39]. As a result, these treatment costs only include the drug price and do not consider any additional charges for doctor’s fees, delivery expenses, administrative fees, or supportive care that may be necessary for the treatment of cancer patients.

2.3 Statistical Analysis

We used descriptive statistics to describe the sample’s baseline characteristics, and then conducted random-effects regression models to evaluate the association between post-launch price changes and collected variables. Random-effects regressions were performed to examine the association between time-varying and time-invariant variables on drug prices. The use of random-effects rather than fixed-effects models was confirmed by performing the Hausman test (χ2 = 12.47, p = 0.0861) and the Lagrange Multiplier (LM) test (\(\overline{\chi }^{2}\) = 42,858.22, p < 0.001). All models account for drug-level clustered standard errors to adjust for heteroscedasticity and autocorrelation. Drug prices, disease incidence, and disease prevalence were transformed with the natural logarithm to account for their skewed distribution.

For all models, the dependent variable (\({y}_{dt}\)) is the inflation-adjusted log-price for each drug (\(d\)). First, we evaluated the association between each independent variable and launch as well as post-launch drug price changes in a series of separate univariate regression analyses. In these models, each independent time-invariant variable (\({x}_{d}\)) was included alongside an interaction term between the time-invariant variable and the time since launch (\({q}_{dt}\)) (Eq. 1). We defined drug launch as the first time a drug’s price was listed in CMS files. Coefficients of the independent variable (\({\beta }_{d}\)) can be interpreted as the association between the independent variable of interest and launch prices. The coefficient of the interaction term (\({\beta }_{dq}\)) can be interpreted as the association between the independent variable of interest and post-launch price changes. Product type, innovativeness, companion biomarkers, special FDA designations, disease incidence, and DALYs per person were included as time-independent variables. Across all models, \({v}_{d}\) represents the drug-specific error and \({\varepsilon }_{dt}\) represents the idiosyncratic error.

$${y}_{dt}=\alpha +{q}_{dt}{\beta }_{0}+{x}_{d}{\beta }_{1}+{q}_{dt}{x}_{d}{\beta }_{q1}+{v}_{d}+{\varepsilon }_{dt}$$
(1)

Among models with time-varying variables, the variable of interest (\({x}_{dt}\)) was the only independent variable and its coefficient can be interpreted as the post-launch price change (Eq. 2). The number of new competitors and new supplemental indications were included as time-varying variables.

$${y}_{dt}=\alpha +{q}_{dt}{\beta }_{0}+{x}_{dt}{\beta }_{1}+{v}_{d}+{\varepsilon }_{dt}$$
(2)

Thereafter, a multivariate regression model was conducted (Eq. 3). Special FDA designations, except for the orphan designation, were excluded as they are often granted concurrently.

$${y}_{dt}=\alpha +{q}_{dt}{\beta }_{0}+\sum_{i=1}^{I}{(x}_{di}{\beta }_{i}+{q}_{dti}{x}_{di}{\beta }_{qi})+\sum_{j=1}^{J}({x}_{dtj}{\beta }_{j})+{v}_{d}+{\varepsilon }_{dt}$$
(3)

Sensitivity analysis was conducted using two-step fixed-effects, rather than random-effects, regression models. First, we constructed a fixed-effects panel regression including all time-varying variables. Based on this model, we predicted log-prices at launch. Thereafter, we conducted an ordinary least squares (OLS) regression including all time-invariant variables on predicted log-prices at launch.

Coherent with previous studies, we examined the cancer drug market within drug classes based on the correlation of prices [24, 29,30,31, 40]. The relationship between nine drug classes with a total of 25 injectable cancer agents was analyzed and visualized using a Pearson correlation matrix. Stigler and Sherwin (1985) suggest that price movements between two products can be used to define the extent of a market [41]. A positive correlation coefficient close to 1 indicates that two products are competing in the same market, whereas a low correlation coefficient suggests that the two products are competing in separate markets [42]. We tested for causality between each drug pair’s logarithmic first difference of prices using the Granger causality test [43].

Data were stored in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) and were analyzed using Stata software, version 14.2 (StataCorp LLC, College Station, TX, USA). Two-tailed p-values <0.05 were considered significant. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines where applicable [44].

3 Results

3.1 Sample Overview

We identified a total of 720 new drugs that received FDA approval between 2000 and 2022. Among these, we identified 170 anticancer agents with FDA approval for a total of 455 indications; 104 of these drugs were covered under Medicare Part D and were therefore excluded from our analysis. The final sample consisted of 66 cancer drugs with quarterly price data from 2005 to 2023 (Fig. 1).

Fig. 1
figure 1

Flow chart of FDA-approved injectable cancer drugs included in the analysis, 2000–2023. CAR chimeric antigen receptor, FDA US Food and Drug Administration

Full size image

For these 66 drugs, mean prices amounted to $27,688 per month in 2023. Higher prices were observed for on-patent drugs than generic drugs (mean $33,988 vs. $7529, Student’s t-test p = 0.008). Figure 2 illustrates the post-launch trajectory of cancer drugs in our sample. For on-patent drugs, prices increased by an average of 94% from 2005 to 2023. For drugs that lost their exclusivity, prices declined by an average of −94% (ESM Fig. e1).

Fig. 2
figure 2

Prices of injectable cancer drugs from 2005 to 2023. The graph shows individual and median prices for injectable cancer drugs (covered under Medicare Part B) from 2005 to 2023. Prices were calculated for the average patient’s monthly treatment cost for the first indication with FDA approval. The median monthly drug price amounted to US$2809 in 2005 and rose to US$14,950 in 2023. We only included on-patent periods, while excluding periods after patent expiry. The outlier triptorelin pamoate was excluded for visualization. All prices are presented in US$. US$ US dollars. FDA US Food and Drug Administration

Full size image

Of the 66 included drugs, 42% were first-in-class agents and 41% were small molecules (Table 1). Twelve drugs (18%) were approved with a companion biomarker. Cancer drugs frequently received the orphan designation (62%), accelerated approval (50%), fast track (44%), priority review (80%), and breakthrough therapy designation (63%). Median disease incidence per 100,000 US inhabitants was 8.3 (interquartile range [IQR] 3.2–69.3) and median DALYs per person was 7.7 (IQR 5.5–12.9). A median of 1 (IQR 0–3) supplemental indication was approved by the FDA for each drug. On a broad level, a median of 14 (IQR 6–23) new competitors entered the market, and, on a narrow level, a median of 3 (IQR 1–6) new competitors entered the market during the study period.

Table 1 Sample overview
Full size table

3.2 Univariate Regression Analysis

Results of the univariate regression analyses are exhibited in Table 2. OS hazard ratios were not significantly correlated to launch prices (ß = 1.12, p = 0.441) or post-launch price changes (ß = −0.0097, p = 0.651). Accordingly, there was no meaningful association between PFS HRs and launch (ß = 0.11, p = 0.945) or post-launch prices (ß = 0.0071, p = 0.729).

Table 2 Univariate random-effects regression analyses of collected variables on prices for FDA-approved injectable cancer drugs from 2005 to 2023
Full size table

In the univariate random-effects models, there was a non-significant trend that launch/post-launch prices are 50% (p = 0.220)/2% per year (p = 0.183) higher for first-in-class, and 55% (p = 0.191) and 2% (p = 0.283) higher, respectively, for biologic agents. There was no relevant association between launch/post-launch prices and a drug’s biomarker status. The price elasticity between launch prices and disease incidence was −0.26 (p = 0.006), and −0.0065 (p = 0.028) between annual post-launch price changes and disease incidence. Accordingly, 73% (p = 0.108) higher launch and 2% (p = 0.037) higher price increases per year post-launch were observed for orphan drugs compared with non-orphan drugs. The approval of new supplemental indications (ß = −0.0082, p = 0.097) and the market entry of new competitors were only marginally and non-significantly associated with post-launch price changes (ß = −0.0064, p = 0.076).

3.3 Multivariate Regression Analysis

In the multivariate random-effects model, special FDA pathways (e.g. fast track, priority review, accelerated approval, breakthrough therapy) other than the orphan designation were excluded given that these are often granted concurrently to medicines with substantial benefits in treating serious conditions with significant unmet needs, e.g. orphan conditions [23, 45]. We conducted three different multivariate random-effects models (Table 3). The first model included all variables, the second model excluded the orphan designation, and the third model excluded disease incidence and DALYs per person, given that these three variables are collinear. The second model shows that launch and post-launch price changes were negatively associated with disease incidence but not disease burden. The elasticity between launch/post-launch price changes and disease incidence was −0.25 (p = 0.008)/−0.0086 (p = 0.032). The third model highlights that launch prices were 120% (p = 0.051) higher for orphan drugs than non-orphan drugs, with 3% (p = 0.008) greater annual post-launch price increases.

Table 3 Multivariate random-effects regression analyses of collected variables on prices for FDA-approved injectable cancer drugs from 2005 to 2022
Full size table

In all three models, post-launch prices declined by up to −2%, as the FDA approved new supplemental indications for the same drug. Prices did not significantly decline as new competitors for the same disease entered the market.

3.4 Sensitivity Analysis

Results remain robust under sensitivity analysis. Regression coefficients and significance levels were robust when using a two-step fixed-effects approach to evaluate the association between collected variables and launch/post-launch price changes (ESM Table e2). Furthermore, there was also no association between post-launch price changes and the narrow measure of competition (ESM Table e3).

3.5 Pairwise Pearson Correlation Coefficients

Figure 3 shows pairwise Pearson correlation coefficients of inflation-adjusted prices for drugs within a class. Price changes for PD-1/PD-L1 inhibitors were closely aligned with Pearson correlation coefficients of between 0.8 and 1.0 (only coefficients for dostarlimab, which was only recently approved by the FDA, were lower). The Granger causality test suggests that for most PD-1/PD-L1 inhibitor drug pair prices (at least) univariate causality exists (ESM Fig. e2). Similarly, coefficients were approaching 1.0, suggesting a very strong positive correlation between prices of drugs within the same class, for CD20 antibodies (r = 0.68 [no causality], 0.93 [bidirectional causality], and 0.97 [unidirectional causality]), CD38 antibodies (r = 0.91, no causality), HER2 antibodies (r = 0.88, unidirectional causality), and mTOR inhibitors (r = 0.98, bidirectional causality). In contrast, correlation coefficients were negative for VEGFR antibodies (r = −0.24, unidirectional causality), proteasome inhibitors (r = −0.83, no causality), and HDAC inhibitors (r = −0.31, no causality). For EGFR antibodies, we observed positive (r = 0.20–0.72) and negative (r = −0.45) correlation coefficients with mixed causality test results.

Fig. 3
figure 3

Pairwise Pearson correlation coefficients of prices for injectable cancer drugs. This matrix presents pairwise correlation coefficients for prices of nine injectable cancer classes entailing 25 distinct drugs. We included cancer drugs with FDA approval between 2000 and 2022, and further included rituximab and trastuzumab to accurately analyze the competitive dynamics of the HER2 and CD20 antibodies. We only included price data for drugs before patent expiry. CD20 cluster of differentiation 20, CD38 cluster of differentiation 38, EGFR epidermal growth factor receptor, HDAC histone deacetylase, HER2 human epidermal growth factor receptor, mTOR mammalian target of rapamycin, NE no estimates, PD-1 programmed death 1, PD-L1 programmed death-ligand 1, VEGFR vascular endothelial growth factor receptor

Full size image

Figure e3 shows that the number of overlapping FDA-approved indications was positively associated with the measured correlation coefficient for each drug pair. For example, the CD38 antibodies daratumumab and isatuximab were both only approved to treat multiple myeloma (100% overlapping indications). Their price correlation coefficient was 0.91. In contrast, the price correlation coefficient was −0.45 for panitumumab and necitumumab, EGFR antibodies that were separately approved for colorectal cancer and NSCLC, respectively (0% overlapping indications).

4 Discussion

This longitudinal study analyzed factors associated with launch prices and post-launch price changes of injectable cancer drugs in the US. Over the study period from 2005 to 2023, prices for on-patent cancer drugs increased by an average of 94%. We found that launch prices were non-significantly higher for innovative first-in-class drugs (Table 3). Post-launch price changes were positively associated with the orphan designation and negatively associated with disease incidence and the approval of new supplemental indications (Table 3). We found no consistent association between launch/post-launch prices and the drugs’ clinical benefit (Table 2). The market entry of new competitors was not associated with price reductions. Twenty-eight of 33 drug pairs within the same class had positive correlation coefficients.

4.1 Clinical Benefit

We found that post-launch price changes were not significantly associated with the clinical benefit of new cancer drugs in terms of OS, PFS, or tumor response benefit. Similarly, Vokinger et al. could not identify any association between post-launch price changes and the drugs’ clinical benefit, as measured by the American Society of Clinical Oncology Value Framework (ASCO-VF) and the European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS) [8]. We observed a positive association between launch prices and the drugs’ PFS, but not OS and tumor response. This positive association between launch prices could be explained by the orphan drugs’ greater PFS benefit that is measured in non-robust clinical trials [23, 46]. These findings are consistent with prior studies that could not confirm a consistent link between the benefits and launch prices of the drugs [2, 11,12,13,14, 16,17,18, 22]. In summary, there is little evidence that launch and post-launch prices are aligned with the clinical benefit a drug offers to patients in the US.

US policymakers could implement value-based pricing policies that regularly re-examine drug prices following initial market entry to better align launch prices and post-launch price changes with the drugs’ clinical benefits. Thereby, new drug prices could better reflect each drug’s clinical benefit as new evidence is generated. This is particularly relevant for cancer drugs [47]. Most cancer drugs are initially approved based on small, non-robust, single-arm trials testing the new drug for rare diseases in a heavily pretreated patient population and reporting surrogate endpoints, e.g. PFS or tumor response [38]. Over time, more robust post-approval trials are conducted for the first-line setting evaluating clinical endpoints, e.g. OS. Evidence from European countries highlights that the resulting lower benefit is associated with price reductions [33].

4.2 Cancer Epidemiology

Greater post-launch price increases were observed for orphan drugs. Accordingly, post-launch price changes were negatively associated with disease incidence. A 1% decline in disease incidence was associated with a 0.2511% (p = 0.008) increase in launch prices and a 0.0086% annual increase in post-launch prices. These findings are coherent with previous studies highlighting that sponsors of specialty drugs, in particular those for rare diseases, demand a launch price premium and further increase prices at more than double the pace of non-orphan drugs [2, 19,20,21, 23], Orphan drugs often offer significant therapeutic advances for patients [23, 46]. However, their high launch and rising post-launch prices pose a barrier for patients to access these medicines [48]. Insured patients do not only have to pay their insurance premium but also have to bear the insurance plans’ deductible and co-payment. Given that co-payments are calculated as a percentage of a drug’s list price, rising post-launch list prices result in rising OOP expenditure for patients. The Orphan Drug Act of 1983 effectively grants drugs for rare diseases 7 years of market exclusivity. Sponsors often use this monopoly position to raise prices [23], which results in excess profits and returns for pharmaceutical firms [49, 50]. The recently introduced Inflation Reduction Act of 2022 (IRA) contains an OOP cost cap of $2000 per year for Medicare Part D drugs. A similar provision would be necessary to limit patients’ OOP expenditures of Medicare Part B drugs, particularly those with rare diseases.

4.3 Competition

In our study, the market entry of new competitors for the same disease was not associated with post-launch price declines. These findings are coherent with previous literature finding no or weak evidence for price competition. Sarpatwari et al. systematically reviewed ten studies and found little evidence and no evidence that launch and post-launch prices are affected by competition, respectively [25]. Howard et al. found launch prices for anticancer drugs to be negatively associated with the number of competitors [15]. Bennette et al. analyzed the market for oral anticancer drugs from 2007 to 2013 using pharmacy claims data and found that the market entrance of new competitors resulted in a 2% price reduction [51]. In contrast, Gordon et al. did not observe any change in anticancer drugs' post-launch prices as new competitors entered the market between 2005 and 2012 [28]. In conclusion, there is little to no evidence of brand-brand price competition in the drug market.

Several factors could help to explain the special competitive dynamics in branded pharmaceutical product markets. First, higher prices could signal greater safety and efficacy, e.g., high-quality products. Acting as a Veblen good, highly-priced drugs could be viewed as superior and thereby induce demand, especially when safety and efficacy data are not available or not accessible (which is often the case for newly approved cancer drugs) [52, 53]. Second, anticancer drugs may often fill niche market segments within a disease. Given only a subset of biomarker-positive patients are eligible to receive targeted agents, from the pool of all available medicines for a disease, only a few products can be perceived as close substitutes. Moreover, the market for anticancer drugs is further convoluted by the use of drugs in different lines of therapy and a complementary manner [51]. Next-in-class drugs may enter market segments for the advanced-line, but not first-line, therapy within the same disease, and thereby not affect the pricing of the first-in-class agent. Nonetheless, our narrow measure of competition, which adjusts for disease type, line of therapy, therapeutic setting, and biomarker profile, did not show any significant price reductions following the entry of new competitors.

Prior studies inappropriately evaluated the competitive dynamics of the pharmaceutical market based on a price correlation analysis [24, 29,30,31, 40]. They assumed that drugs of the same class compete within the same market. Although cancer drugs within the same class are typically not approved for the same indications, authors assumed that high correlation coefficients (>0.80) can be interpreted as a lack of within-class competition, given that drug-pairs’ prices rise at the same pace. However, correlation coefficients approaching 1 may also indicate competitive pressure if drug-pairs’ prices simultaneously decline. In economics, price correlations are used to define and differentiate markets [41, 42, 54]. A positive correlation indicates that two products compete in the same market, whereas a low correlation suggests that they belong to two separate markets. This study observed positive correlation coefficients beyond 0.80 for 25 of 33 evaluated drug pairs (without PD-1/PD-L1 inhibitors: 6 of 12). Excluding PD-1/PD-L1 inhibitors, we found the percentage of overlapping indications to be positively associated with drug-pairs’ correlation coefficients. This result suggests that the drug class, which represents a biochemical classification of a drug’s target, may not be the sole adequate measure to define the competitive market for cancer drugs. Instead, a clinical, patient-centered approach for a cancer drug’s competitive market should also entail its FDA-approved indications, which legally define the eligible patient population. In other words, physicians can only interchangeably prescribe new drugs of the same class to cancer patients (and thereby create a competitive market environment) if both drugs receive approval for the same diseases and line of therapy. Therefore, future studies analyzing competition among cancer drugs must adequately define the competitive market for each drug based on the eligible patient population, and employ appropriate economic methodology to analyze competition.

4.4 Supplemental Indications

Results show a marginal reduction in prices (up to −2%) as new supplemental indications were approved for the same drug. In contrast, Gordon et al. did not observe any association between price changes and new supplemental indications or off-label uses [28], while Bennette et al. found a positive correlation between price changes and new supplemental indications [51]. However, the analysis by Gordon et al. is based on multivariate OLS models and is limited to 24 injectable anticancer drugs with price data from 2005 to 2017. In our analysis, we performed random-effects panel regressions entailing price data from 66 drugs from 2005 to 2023. Furthermore, Bennette et al. examined the market for oral cancer drugs from 2007 to 2013, while we analyzed the market for injectable cancer drugs. The dynamics for these two markets appear to differ. Furthermore, the result that new supplemental indication approvals are associated with marginal price declines was expected given that pharmaceutical companies were shown to first approve cancer drugs for orphan indications and then extend FDA approval to non-orphan diseases with a greater patient population for which the drug often offers a lower clinical benefit [33, 37, 38]. Pharmaceutical companies seem to account for the FDA label extension to more patients with a lower benefit by marginally reducing drug prices.

4.5 Inflation Reduction Act of 2022

US Congress recently passed the IRA [55], which contains three key elements to reduce the financial burden of prescription drug prices for patients and the healthcare system. Regarding drugs covered under Medicare Part B, there are two important IRA provisions. First, the CMS is now, for the first time in US history, permitted to directly negotiate prices of the 10/20 top-grossing prescription drugs with manufacturers, beginning in 2026/2029 [47]. For Medicare Part B drugs, these negotiations may start in 2028, excluding orphan drugs (with only a single approved indication). Given the significantly higher prices of orphan drugs than non-orphan drugs that generate substantial revenues for pharmaceutical companies [23, 56], top-grossing orphan drugs should be included in price negotiations. Second, the CMS sets discounts on post-launch drug price increases exceeding inflation beginning in 2023. Although this provision will limit post-launch net price increases for patients and insurers, pharmaceutical companies will likely continue to raise list prices as certain countries, e.g. Canada, South Korea, or Japan, include US drug prices in their basket of countries to calculate their national price (external reference pricing) [57]. For these countries and for patients’ OOP expenses, limiting the post-launch list, instead of net, price increases in the US would be the more effective pharmaceutical policy.

4.6 Limitations

There are several limitations inherent to our analysis. First, we only assessed prices for drugs covered under Medicare Part B, which covers only injectable drugs that are typically administered at a hospital or doctor’s office. Therefore, our sample is restricted to 66 of 170 drugs (39%) with FDA approval between 2000 and 2022. The distinction between injectable (Part B) and oral (Part D) drugs is particularly important given that these two markets could be subject to different pricing dynamics [51]. These different dynamics may be caused by separate price regulations for Medicare Part B and D drugs. For Part B drugs, the Medicare payment limit is defined as the lesser of 106% of the ASP or 106% of the wholesale acquisition cost for the drug. While this provision effectively limits the reimbursement of Part B drugs, the reimbursement and post-launch price changes for Part D drugs remained largely unregulated until the introduction of the IRA. Furthermore, our analysis of Pearson correlation coefficients is also limited to Part B drugs. Future research should conduct a similar analysis with Part B and D drug prices to fully capture the market dynamics of new competitors. Second, we analyzed list prices for patients covered under Medicare. Net prices and net price changes may vary, particularly for patients covered by private insurers whose plans may offer a distinct set of co-payments, deductibles, and discounts. Third, our analysis was conducted before the IRA’s provision to limit price increases exceeding inflation became effective in 2023. This provision may distort the competitive dynamics of branded pharmaceutical products, thereby limiting the generalizability of our findings for the future. Fourth, the pairwise correlation analysis could be subject to omitted variable bias and spurious correlation [42]. Furthermore, our analysis is limited to price data. Price changes caused by differential changes in the number of units sold for a drug pair could yield too-low correlation coefficients. Moreover, the price correlation analysis does not capture the drugs’ differential quality. Finally, given that our findings rely on cancer drug prices, results and policy implications should be confirmed for other therapeutic areas.

5 Conclusion

Using Medicare and Medicaid data, we observed substantial increases in the post-launch prices of injectable cancer drugs from 2005 to 2023. Launch prices and post-launch price changes were not aligned with the clinical benefit a drug offers to patients. Greater launch prices and post-launch price changes were observed for orphan drugs, while the introduction of new supplemental indications was associated with a −2% price reduction. We show that the competitive market for each drug is defined by the eligible patient population (e.g. FDA-approved indications). In our analysis, the market entry of new competitors was not associated with price declines. Similar to European countries, US policymakers should not only negotiate drug prices at launch but also reassess their initial negotiations several years post-launch to limit the rising cost of cancer drugs for health insurers and patients.