A systematic review of the applications of the real option value of medical technologies in oncology

Abstract

BACKGROUND:

Real option value (ROV) offers an innovative paradigm to evaluate the dynamic value of medical technologies, particularly in cancer, by capturing the value of extending patient survival to access future innovations. Despite its potential, the application of ROV in medical technologies in oncology remains underexplored.

OBJECTIVE:

To synthesize existing evidence on the application of ROV in medical technologies in oncology.

METHODS:

A comprehensive search of PubMed, ScienceDirect, and Web of Science was conducted to identify peer-reviewed studies published in English from January 2000 to May 2024. In the search query, a combination of keywords related to “real option value” and “cancer” was used. Key data extracted included study characteristics, objectives, ROV modeling technique, and primary findings. The Consolidated Health Economic Evaluation Reporting Standards 2022 checklist was used for quality assessment of the studies.

RESULTS:

A total of 13 of 165 studies assessed the ROV of medical therapies, with a primary focus on melanoma, lung cancer, and prostate cancer. ROV was modeled from the ex post and ex ante perspectives. The methodologies employed vary, with common forecasting approaches including the Lee-Carter model to project future decreases in mortality rates, fitting Cox proportional regression models on administrative claims data, or estimating the approval likelihood of early pipeline drugs based on data from early randomized clinical trials.

CONCLUSIONS:

The ROV represents a critical dimension in evaluating medical technologies in oncology, where innovation is rapid. The implications of ROV extend beyond oncology, with the potential to influence funding, pricing, and access decisions in other disease areas as well. However, challenges such as oversimplification of assumptions for forecasting, methodological consistency, and lack of standardized framework remain pervasive. This systematic review underscores the need to integrate ROV into Health Technology Assessment practices to inform resource allocation and policy decisions.

Plain language summary

This study looked at how researchers use a concept called “real option value” (ROV) to measure the long-term benefits of cancer treatments. ROV means the benefit of the therapy extending patients’ lives to live to a better future treatment. We reviewed past studies to see how this idea has been used in cancer research. Our findings show that although the concept is promising, it is still new and not widely used in real-world decisions.

Implications for managed care pharmacy

This review highlights how ROV can support a broader understanding of the long-term benefits of cancer treatments. By recognizing the value of extended survival that enables access to future therapies, managed care decision-makers may consider ROV when assessing oncology drugs. Integrating this perspective into value frameworks and coverage decisions could improve benefit design and align reimbursement with treatments that offer future health gains.

The escalating complexity and expense of oncology therapies require novel frameworks to evaluate their usefulness beyond traditional cost-effectiveness metrics.¹ Cost-effectiveness analyses are limited by their static assumptions that health states and treatment effects remain constant over time. This approach overlooks the impact of future medical innovations and evolving patient needs, hindering the adoption of potentially transformative therapies.¹ In contrast, frameworks like real option value (ROV) offer a dynamic evaluation that incorporates uncertainty and the potential for future advancements.^2,3 ROV recognizes the added value of medical technology by evaluating its ability to prolong a patient’s life to enable access to newer and better medicines.^2,3 This approach is especially pertinent to oncology because the treatment landscape changes rapidly.

ROV, based on financial theory, signifies the necessity of maintaining flexibility in decision-making amid uncertainty.^4,5 In oncology, there are 3 main mechanisms through which ROV of current treatment arises: type 1, when the current treatment extends survival thereby increasing the chance for future innovation; type 2, when it delays disease progression to increase chance for future innovation; and type 3, when it directly modifies the effectiveness, safety, or use of future innovation.³ Type 1 is exemplified by cancer therapies such as epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) and immunotherapies used in the treatment of lung cancer and metastatic melanoma, which extend survival allowing patients to remain alive long enough to benefit from future, more effective treatments.^6–9 Type 2 can be demonstrated by the use of antiandrogens, such as bicalutamide and flutamide in prostate cancer, which offer minimal survival benefit; however, they slow disease progression and deterioration of functional status, ensuring that patients remain eligible for future lines of therapy.¹⁰ Also, type 3 can be observed in certain therapies, such as radiotherapy and surgery, which can influence tumor biology in ways that affect the efficacy or use of future interventions.¹¹ Unlike the traditional cost-effectiveness analysis, ROV seeks to incorporate the dynamic nature of medical advancements and their long-term implications for patient care and reimbursement decisions.¹²

Even though ROV is theoretically appealing, its application and integration in Health Technology Assessments (HTAs) are limited because of methodological complexities and lack of standardization cited as barriers to wider adoption.^2,3 One significant challenge is the uncertainty in forecasting future innovations in medical technology. Accurately predicting how and when new treatments will emerge, along with their potential effects on patient outcomes, poses a considerable challenge for decision-makers. This highlights the need to systematically explore ROV in the context of the assessment of medical technologies for oncology as well as the insights that can be gleaned for future research and policy.³

This systematic review synthesizes evidence regarding ROV’s application to oncology. It discusses the methodologies employed, key findings of the studies, and implications for decision-making. Hence, this review seeks to contribute to understanding ROV as an addition to economic evaluations of health care and to highlight gaps in the literature that warrant further investigation.

Methods

A systematic approach was employed to search for research articles from relevant databases, including PubMed, Web of Science, and ScienceDirect, as seen in Figure 1. In the search query, a combination of medical subject headings (MeSH) and keywords related to “Real Option Value” and “cancer” was used. A search string was employed to find relevant studies: “((((real option value[Title/Abstract]) OR (real option analysis[Title/Abstract])) OR (option value[Title/Abstract])) OR (value of life[Title/Abstract])) AND (cancer[MeSH Terms]).” Studies published from January 2010 to May 2024 were included in the review to ensure relevance and currency. Also, studies that were published in English and that incorporated the use of ROV in the decision-making processes related to cancer treatment were included. Primary research studies reporting outcomes regarding clinical efficacy, cost-effectiveness, or resource allocation using ROV were also included. All databases and sources included in this review were last searched in September 2024.

FIGURE 1.

Flowchart for the Systematic Review of the Application of Real Option Value in Oncology

Studies were selected after 2 independent rounds of screening titles and abstracts for eligibility. Table 1 captures all the eligibility criteria for the systematic search. Full-text articles were further assessed for eligibility. The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 checklist was used to evaluate the quality of 13 studies that met the eligibility criteria. The CHEERS 2022 checklist is a 28-item list recognized by the EQUATOR network as a guide for health economic studies.¹³ This checklist assessed the quality of the title, abstract, introduction, method used, discussion, and other relevant information (Table 2). Scoring criteria were designed to categorize the quality of the studies into good, moderate, and low quality. Relevant data was extracted using a predefined data form. Table 3 shows the extracted data, including study characteristics (objective, authors, and year of publication), methodology, ROV applications, and findings. Evidence synthesis was conducted from the data extracted to provide a comprehensive summary of the applications of ROV in cancer care. Cross-tabulation methods were used to synthesize the extracted data. The overall confidence in the body of evidence was assessed narratively based on study design, consistency of findings, and relevance to the research question. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed for reporting.

TABLE 1.

Inclusion and Exclusion Criteria for a Systematic Review of the Application of ROV in Oncology

	Inclusion criteria	Exclusion criteria
Study design	Quantitative, qualitative, mixed-methods full-text studies in English	Editorials, commentaries, and reviews
Publication date	From January 2000 to May 2024	Before January 2000
Population	Studies focused on patients with cancer of any age	Studies not related to patient care in oncology
Intervention	Anticancer agents: targeted therapy, immunotherapy, chemotherapy, surgery, and radiotherapy	Outside the context of anticancer agents
Outcomes	All outcomes relevant to the application of ROV, including but not limited to incremental cost-effectiveness, health resource utilization, quality of life, survival extension, and disease progression	Studies without specified outcomes related to ROV

ROV = real option value.

TABLE 2.

Quality Assessment of Economic Evaluations Using the Consolidated Health Economic Evaluation Reporting Standards 2022 Checklist¹³

Item	Wong et al, 202414	Lee et al, 20221	Wong et al, 202115	Lee et al, 202116	Li et al, 201917	Li et al, 201918	Snider et al, 201819	Shafrin et al, 201820	Snider et al, 201721	Thijssen et al, 201722	Snider et al, 201223	Sanchez et al, 201224	Grutters et al, 201125
Quality	Good	Good	Good	Good	Good	Good	Good	Good	Good	Moderate	Moderate	Moderate	Moderate
1. Title	–	✓	✓	✓	✓	–	✓	✓	✓	–	–	✓	✓
2. Abstract	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
3. Background and objectives	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
4. Health economic analysis plan	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
5. Study population	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
6. Setting described	✓	–	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
7. Comparators described	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
8. Perspective described	–	✓	–	✓	✓	X	–	✓	–	✓	✓	X	X
9. Time horizon described	–	✓	–	✓	✓	✓	✓	✓	✓	–	✓	–	✓
10. Discount rate	X	✓	X	✓	✓	–	✓	✓	✓	✓	✓	X	✓
11. Selection of outcomes	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
12. Measurement of outcomes	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	–
13. Valuation of outcomes	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	_
14. Rationale and description of model	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
15. Analytics and assumptions	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
16. Characterizing heterogeneity	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
17. Characterizing distributional effects	X	X	X	X	X	X	X	X	X	X	X	X	X
18. Characterizing uncertainty	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
19. Approach to engagement with patients/stakeholders affected by the study	–	X	–	X	–	–	X	X	X	X	X	X	X
20. Study parameters described	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	_
21. Summary of main results	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
22. Effect of uncertainty described	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
23. Effect of engagement with patients and others affected by the study	X	X	X	X	X	✓	X	X	X	X	X	X	X
24. Study findings, limitations, generalizability, and current knowledge	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
25. Source of funding declared	✓	✓	✓	✓	X	✓	X	✓	✓	X	X	✓	✓
26. Conflicts of interest declared	✓	✓	✓	✓	X	✓	✓	✓	✓	X	X	✓	X

The scoring symbols are as follows: (✓) = 2 points: fully reported and transparent; (–) = 1 point: partially reported/unclear; and (X) = 0 points: not reported. The categories are ranked as follows: good quality (≥42), moderate quality (31-41), and low quality (≤30).

TABLE 3.

Extracted Data From Studies Included in the Review

Author and year	Objective	Type of cancer/oncology technology assessed	ROV mechanism	Perspective	ROV approach	Findings
Wong et al, 202414	Evaluated how the AA pathway affects patients on lifelong treatments and their ability to access next line of therapy	Existing first-line therapies vs sequential AA innovations in metastatic melanoma, bladder cancer and EGFR NSCLC	Type 1	Ex post	The effect of AA on survival was quantified using the Flatiron Health electronic health records database. It modeled the counterfactual scenario in which AA did not exist. The study assessed observed survival gains from early access to AA therapies against a hypothetical scenario where these AA treatments were only available at full FDA approval	Patients with access to AA had twice as much increase in survival compared with traditional approval pathway for metastatic melanoma, bladder cancer, and NSCLC
Lee et al, 20221	Evaluated the ex ante ROV for 1L alectinib in NSCLC positive for ALK	1L alectinib vs 1L critizonib; advanced ALK-positive NSCLC	Type 1	Ex ante	A conventional CEA and ROV scenario were constructed. The pipeline approach in which future drug approvals for NSCLC are projected was used to account for the ROV. The pipeline approach was implemented by identifying ongoing phase 2 and 3 trials for ALK-positive NSCLC as of the approval date of 1L alectinib (November 2017). Future drugs considered were those with higher efficacy than brigatinib, based on publicly available ORR data. The likelihood of FDA approval, timing of arrival, and potential uptake were then estimated to forecast their impact on postprogression survival and ROV	Compared with crizotinib, alectinib achieved incremental LYs and QALYs of 1.11 years and 0.92 QALYs, respectively, in the traditional approach. The option value scenario raised the incremental LYs and QALYs by 0.14 years and 0.10 QALYs, respectively, representing a 12.9% and 11.2% increase from the traditional approach
Wong et al, 202115	Evaluated the ROV of treatments in advanced melanoma using real-world data	Advanced unresectable or metastatic melanoma; 4 treatment groups: 1L ipilimumab, 1L chemotherapy, 2L ipilimumab, and 1L and 2L chemotherapy	Type 1	Ex post	ROV was estimated using the population getting the next innovation and its benefit. Those who got additional immunotherapy after ipilimumab were multiplied by the difference in mean survival between those who did and did not. To compute the overall ROV for ipilimumab, the estimate was weighted by the proportion of patients receiving ipilimumab as 1L vs 2L therapy	Patients who started 1L ipilimumab had an average additional 3.7 months of survival compared with those who started 2L ipilimumab, who had an average additional 4.8 months of survival
Lee et al, 202116	Estimated the ROV of patients taking ipilimumab for the treatment of melanoma	Metastatic melanoma;ipilimumab compared with chemotherapy for patients with/without subsequent immunotherapy	Type 1	Ex post	Four scenarios were created. “A: 1L chemotherapy with no further innovations; B: 1L chemotherapy with CIT as a 2L therapy; C: 1L ipilimumab monotherapy with no further innovations; D: 1L ipilimumab monotherapy with CIT as a 2L. The ex post ROV of ipilimumab was calculated as (D−C)−(B−A)”	In comparison with chemotherapy without following CIT, ipilimumab resulted in an incremental QALY gain of 1.74. When CIT became available at 0, 3, 6, 9, 12, and 24 months after the start of 1L treatment, the incremental QALYs of ipilimumab compared with chemotherapy went up by 0.92, 0.60, 0.33, 0.18, 0.10, and 0.02. The ex post ROV ranged from 1% to 54% of the conventional value for patients diagnosed within 2 years before CIT availability
Li et al, 201917	Evaluated how ROV impacts the cost-effectiveness of ipilimumab in metastatic melanoma	Previously untreated unresectable metastatic melanoma; ipilimumab plus dacarbazine vs dacarbazine	Type 1	Ex ante	A conventional CEA and ROV scenario were constructed. Two methods were employed: “one was based on the prediction of historic trends in melanoma-specific mortality (SEER approach), while the other was based on the prediction of probable future drug approvals for metastatic melanoma (pipeline approach)”	In the SEER approach, “incremental QALYs gained and incremental cost increased by 6.2% and 3.8%, respectively, but the ICER fell by 2.3% compared to the traditional scenario.” With the ClinicalTrials.gov strategy, “incremental QALY gained and cost increased by 7.5% and 7.1%, respectively, while ICER dropped by 0.40%”
Li et al, 201918	Assessed the ROV of ipilimumab in metastatic melanoma	Ipilimumab in metastatic melanoma	Type 1	Ex ante	The use of treatment indicated for metastatic melanoma were examined using interrupted time-series. A multinomial logistic regression was then applied “to quantify the impact of ipilimumab’s phase 2 and 3 disclosures on the risk of surgical resection, systemic therapy, or both within 3 months of the first metastatic diagnosis vs to no treatment”	Disclosure of ipilimumab’s phase 2 outcome led to an approximately 2-fold increase in metastatic surgical resection probability compared with no therapy. No substantial effects were observed after announcing the phase 3 outcome
Snider et al, 201819	Constructed a model to quantify the incremental survival that patients with metastatic melanoma gain from a therapy, enabling them to benefit from subsequent advancements	Advanced unresectable or metastatic melanoma; ipilimumab	Type 1	Ex post	Using the SEER cancer registry and clinical trial data, 3 survival curves were estimated: pre-ipilimumab survival, conventional ipilimumab survival, and ipilimumab patients who later received a PD-1 inhibitor	2L patients with metastatic melanoma receiving ipilimumab gained 10.5 months relative to the previous standard of care. PD-1 inhibitors increased life expectancy by 1.6, 2.8, and 5.1 months for patients diagnosed in 2011, 2012, and 2013, respectively
Shafrin et al, 201820	Assessed the impact of 2L nivolumab in NSCLC	Nivolumab vs docetaxel	Type 1	Ex ante	Net monetary benefit framework to assess the value of 2L nivolumab in NSCLC from multiple perspectives. The total benefit was converted to monetary value by multiplying the entire QALY benefit by the dollar value of a QALY. The net monetary benefit was determined by subtracting incremental total costs from incremental total benefits	Adding option value contributed to additional 0.037 QALYs or $5,599 of survival gains
Snider et al, 201721	Constructed a model to examine the ROV of nivolumab in RCC and NSCLC	NSCLC and RCC; nivolumab was compared with docetaxel in NSCLC and with everolimus in renal carcinoma	Type 1	Ex ante	Estimated pre-nivolumab and nivolumab survival curves. Survival improvements were forecasted using the Lee-Carter method. Using the survival curves, life expectancy was calculated under pre-nivolumab and nivolumab. The ROV is determined by comparing life expectancy with nivolumab without further innovation to its projected life expectancy with future innovation	In RCC, the conventional economic value of nivolumab is $775 million, with an option value of $105 million. In nonsquamous NSCLC, the conventional value of nivolumab was $2.6 billion, and in squamous NSCLC it was $2.0 billion. The option values were $203 million and $73 million, respectively
Thijssen et al, 201722	Explored the role of sequential experimentation using ROV to evaluate adoption and trial or abandonment of robot-assisted vs standard laparoscopic prostatectomy	Robot-assisted laparoscopic prostatectomy compared with standard laparoscopic prostatectomy in prostate cancer	Type 2	Ex ante	The optimal trial, adoption, and abandonment decision modeling for robot-assisted laparoscopic proctectomy was applied	Robot-assisted laparoscopic prostatectomy was rapidly adopted when its QALY benefit was significantly higher than the standard technique, with all trials leading to adoption within 1 month. However, when the benefit was uncertain, trials were prolonged indefinitely owing to the low cost of continued evidence collection and the potential high benefits if proven superior later. The expected trial cost was £110,600, with an initial project value of £15.07 billion, reinforcing the economic potential of adopting the robotic procedure if cost-effectiveness was confirmed
Snider et al, 201223	Estimated the ROV of prophylactic tamoxifen preceding the use of adjuvant Herceptin treatment	Prophylactic tamoxifen+Herceptin (with Taxol and anthracycline) vs Herceptin as an adjuvant therapy (with Taxol+anthracycline) without tamoxifen in metastatic breast cancer	Type 2	Ex post	The extent of breast cancer onset delay using prophylactic tamoxifen was determined. “Survival functions from diagnosis to death were estimated for Taxol+anthracyclines and Taxol+anthracyclines+herceptin”	The expected LYs of conventional of tamoxifen ignoring ROV was 1.26 in 1999 and 2003, whereas it increased to 1.57 in 1999 and 1.38 in 2003 after incorporating ROV. The conventional willingness-to-pay for tamoxifen is $200,339. However, considering the option value of tamoxifen, this amount increases to $247,913 for those who started tamoxifen in 1999
Sanchez et al, 201224	To quantify ROV of patients receiving treatment for CML	CML; tyrosine kinase inhibitors treatment (imatinib and dasatinib)	Type 1	Ex ante	A Cox proportional hazards model was employed to estimate CML-specific mortality and other cause mortality. The Lee-Carter method was then used to forecast future improvements in CML	A newly diagnosed patient with CML has a 0.76 LY option value for future medical innovation therapy. This option value is worth $63,000 or 9% of the average survival gains from existing treatment
Grutters et al, 201125	To evaluate whether a proton therapy should be adopted for stage 1 NSCLC immediately or wait for additional evidence	Proton therapy vs stereotactic body radiotherapy in stage 1 NSCLC	Type 2	Ex ante	Three options for adopting proton therapy: “(a) adopt without further research, (b) adopt and undertake a trial, and (c) delay adoption and undertake a trial.” The predicted net gain of each option was computed by subtracting the total cost from projected benefits. The alternative with the largest predicted net gain and optimal trial design was picked if trialing was worthwhile	“Adopt and trial” was preferred over “delay and trial” for sample sizes <950 patients. The optimal sample size was 200, where the highest expected net gain was achieved, balancing reduced uncertainty with manageable costs

1L = first-line; 2L = second-line; AA = accelerated approval; ALK = anaplastic lymphoma kinase; CEA = cost-effectiveness analysis; CML = chronic myeloid leukemia; EGFR = epidermal growth factor receptor; FDA; US Food and Drug Administration; ICER = incremental cost-effectiveness ratio; LY = life-year; NSCLC = non–small cell lung cancer; ORR = overall response rate; QALY = quality-adjusted life-year; RCC = renal cell carcinoma; ROV = real option value; SEER = Surveillance, Epidemiology, and End Results.

Results

A total of 13 studies were included in the final review following the application of the inclusion and exclusion criteria as listed in Table 1. The studies primarily focused on immunotherapies and targeted therapies for various types of cancer, which include non–small cell lung cancer (NSCLC),^1,20,21,25 metastatic melanoma,^14–19 breast cancer,²³ prostate cancer,²² and chronic myeloid leukemia (CML).²⁴ The CHEERS 2022 checklist was used to assess the quality of the studies. Items 14 and 15 of the original CHEERS checklist, which assess how costs were valued, were excluded in assessing these studies so as not to underestimate the quality of the studies. These included studies focused on estimating ROV regarding survival gains and hence did not incorporate cost in the analysis except for Li et al.¹⁷ Consequently, the maximum possible CHEERS score was reduced owing to excluding 2 criteria. To maintain consistent quality categorization, the thresholds were proportionally adjusted based on the new maximum score, as seen in Table 2.

Of the 13 studies reviewed, 9 were classified as good quality and 4 as moderate quality based on the CHEERS checklist. The studies, classified as good quality in Table 2, provided comprehensive details on the objective, rationale, description of modeling, valuation of outcomes, uncertainty characterization, and discussion, reflecting a strong emphasis on clarity in their analytical approaches. Studies classified as moderate quality did not clearly report the perspective, discount rate, characterizing distributional effects, stakeholder engagement and its impact, or conflict of interest.^22–25

Among these, ROV was defined as “the benefit a therapy provides patients by enabling them to survive to the next innovation.”^14–21,23 However, Grutters et al²⁵ and Thijssen et al²² referred to ROV as “the ability to delay a decision till a later period, until better information regarding costs and benefits may become available.” The definition of ROV in these studies influenced the methodological framework for ROV analysis. ROV was quantified from 2 viewpoints: ex ante and ex post perspectives. Ex ante ROV uses information readily accessible during the product’s launch to project the chance, approval timing, and survival implications for future innovations.^1,17,21 In comparison, “the ex post ROV is estimated by treating the arrivals of the subsequent innovations with certainty, given that they are known.”¹⁶ Table 3 provides extracted data on the disease condition, perspective, and modeling technique in each study included in this analysis.

Among the studies, the most used ROV mechanism was survival extension, which increases the likelihood of patients to live to a better future. This mechanism has been extensively studied in melanoma and lung cancer.^1,14,17–19 Wong et al¹⁴ investigated how accelerated approval (AA) pathways improved access to cancer drugs that extend survival. They revealed patients who received early approved medications that extend life had a greater chance of surviving than those following a conventional approval of drugs. Li et al¹⁷ and Snider et al¹⁹ also investigated how ipilimumab improved survival in the context of metastatic melanoma, allowing patients to later receive newer treatments not accessible when ipilimumab was initially launched. A smaller subset of studies examined delaying disease progression as the ROV mechanism, particularly in prostate cancer,²² prophylaxis in breast cancer,²³ and proton therapy in NSCLC,²⁵ while biological pathway modifications remain an emerging but underrepresented aspect of ROV analysis.

KEY FINDINGS BY CANCER TYPE

Metastatic Melanoma.

ROV application has been well-studied in metastatic melanoma, including the survival gains of immunotherapy, such as ipilimumab, and PD-1 inhibitors, such as pembrolizumab and nivolumab.^14–17,19 Lee et al¹⁶ evaluated the ROV of ipilimumab using the ex post approach. The Markov model estimated the additional survival and quality-adjusted life-year (QALY) gains with and without future improvements. Lee et al¹⁶ constructed 4 scenarios comparing chemotherapy and ipilimumab as first-line treatments, with or without cancer immunotherapy (CIT) as a later-line therapy. The authors estimated the ex post ROV of ipilimumab by “calculating the differential health gains of ipilimumab patients accessing subsequent CIT vs those not accessing it, adjusting for the incremental benefits attributable to subsequent CIT in patients initially treated with chemotherapy.”¹⁶ Key determinants in estimating the ROV encompassed CIT availability (from immediate access to delays of up to 24 months), CIT use, and additional QALYs acquired.¹⁶ Access to CIT increased the incremental QALYs by 0.92; however, this advantage waned with delayed CIT availability.¹⁶ Ex post ROV accounted for 1% to 54% of the traditional scenario of ipilimumab based on future treatment timing, suggesting an essential role for early access to innovative drugs in achieving health outcomes.

Snider et al¹⁹ estimated ipilimumab option value using the Surveillance, Epidemiology, and End Results (SEER) cancer registry. This study simulated survival in 3 scenarios: pre-ipilimumab survival (standard of care), conventional ipilimumab survival (assuming no further treatment advance), and ipilimumab survival with subsequent access to PD-1 drugs.¹⁹ The option value was defined as “the difference in life expectancy between patients who survived on ipilimumab long enough to access PD-1 inhibitors and patients assumed to have no further treatment innovations (conventional scenario).”¹⁹ Patients diagnosed in 2011, 2012, and 2013 had an increased life expectancy of 1.6, 2.8, and 5.1 months, respectively, until the introduction of PD-1 inhibitors. These gains translate into ROV contributions of 15%, 27%, and 49%, respectively, of the conventional survival benefit of ipilimumab.¹⁹

In a real-world setting, Wong et al¹⁵ leveraged electronic health record data to examine the ROV by analyzing survival benefits and treatment trajectories in patients administered ipilimumab before introducing CITs. The average survival among patients receiving the first-line ipilimumab was 3.7 months, representing 35% of the standard survival benefit. Conversely, patients receiving second-line ipilimumab had an average of 4.8 months of survival, constituting 57% of the standard value.¹⁵ In a recent real-world analysis, Wong et al¹⁴ again examined survival outcomes for bladder cancer, NSCLC, and metastatic melanoma under AA. Their study showed that “the AA pathway contributed to 3.73 additional months of survival per patient among all patients who initiated ipilimumab in metastatic melanoma, 0.85 months per patient among all patients initiating CITs post platinum therapy in advanced bladder cancer, and 2.89 months per patient among all patients initiating first- and second-generation EGFR-TKIs.”

In a different scenario, Li et al¹⁸ conducted an interrupted time-series analysis and multinomial logistic regression using Flatiron claims data to examine the use of current treatment among patients with metastatic melanoma following the phase 2 and phase 3 results disclosure of the then-investigational drug ipilimumab. The rationale of their study was to examine whether cancer treatments offer option value such that the expectation of a new experimental drug influences the use of existing treatments. Li et al¹⁸ discovered that “the announcement of ipilimumab’s phase 2 results correlated with a nearly 2-fold increase in the likelihood of undergoing surgical resection of metastases compared to no therapy.” However, for the phase 3 result announcement, no notable impacts were observed. This finding illustrates real-world evidence of how the anticipation of an investigational new drug can affect the use of existing treatments. Li et al¹⁸ revealed that, following the phase 2 announcement of ipilimumab, the option value of surgical resection is likely to increase because of its ability to prolong survival, reduce tumor burden, and reverse immunosuppression, thereby positioning patients to benefit from future therapies. In contrast, systemic therapies at the time offered limited survival benefit and high toxicity, contributing less option value.

Li et al¹⁷ proceeded to examine the ROV of ipilimumab use in metastatic melanoma from an ex ante viewpoint by predicting mortality trends and treatment approvals. Two distinct methods were employed: the SEER-based approach, which forecasted historical mortality reductions, and the pipeline-based approach, which assessed the efficacy probability of approval of prospective pharmaceuticals identified on ClinicalTrials.gov. The results revealed that the ROV approach increased QALYs gained by 6.2% and 7.5% for the SEER approach and pipeline approach, respectively. Also, the incremental cost increased by 3.8% and 7.1% for the trend approach and pipeline approach, respectively. Li et al¹⁷ found a decrease in conventional incremental cost-effectiveness ratio (ICER) by 2.3%, from $229,175/QALY to $223,944/QALY for the SEER approach, whereas for the pipeline approach the conventional ICER decreased by 0.4% to $228,258/QALY.

NSCLC and Renal Cell Carcinoma.

ROV has been analyzed using survival forecasting and economic models in advanced NSCLC and renal cell carcinoma (RCC). Snider et al²¹ developed a framework to quantify the ex ante option value of immunotherapies among patients with NSCLC and RCC using SEER-based survival trends. The survival outcomes were compared in 2 scenarios: patients receiving nivolumab with no future innovation and those who will benefit from projected survival gains owing to anticipated therapeutic advances.

The Lee-Carter model was also used to forecast health gains among each treatment group. The Lee-Carter method models age-specific mortality rates by considering a long-term trend in overall mortality improvements alongside an age-specific response to this trend.^21,26 This technique serves as a benchmark for long-term mortality projections, employed by the US Census Bureau and the Social Security Administration.^21,26 The Lee-Carter model projects declining mortality rates in the future, reflecting anticipated improvements in population survival. As mortality rates decrease, the probability of surviving into future years increases. For patients receiving life-extending therapies, this translates into a greater likelihood of surviving longer. The option value captures the additional LYs from surviving to the use of these future treatments. The ROV accounted for 18% of conventional survival value in RCC, 5% in squamous NSCLC, and 10% in nonsquamous NSCLC.²¹

Lee et al¹ assessed the ex ante ROV of first-line alectinib vs crizotinib in advanced NSCLC harboring anaplastic lymphoma kinase. A Markov model was employed to project survival gains in scenarios with and without future innovations. Following the incorporation of ROV, incremental LYs and QALYs rose by 12.9% and 11.2%, respectively, suggesting that life-extending therapies may provide additional treatment options.¹

Grutters et al²⁵ used real options analysis to assess proton treatment for the early stages of NSCLC. The study considered the trade-off between immediate proton therapy adoption with uncertain benefits and postponing adoption for further research. The real options analysis provided a framed framework for weighing the benefits of early adoption against the costs of reducing uncertainty through additional evidence generation. Analysis suggested an “adopt and trial” approach with 200 patients as the optimal sample size, highlighting proton therapy’s potential value despite uncertainty and high upfront costs.²⁵

Shafrin et al²⁰ investigated the influence of societal perspective on cost-effectiveness concerning nivolumab in patients with NSCLC. Shafrin et al²⁰ calculated the net monetary benefit to compare the incremental value of nivolumab vs docetaxel in NSCLC. The QALYs gained were multiplied by the assumed value of a QALY ($150,000 CAD) to estimate the benefit, whereas costs included drug costs, medical costs, and other health care expenses. Under the broad societal perspective, the calculation was expanded by incorporating value of hope, option value, and insurance value, inflating QALYs beyond the traditional payer perspective.

Insurance value was estimated as the value of risk protection provided to healthy individuals potentially at risk for lung cancer. Shafrin et al²⁰ used an indirect approach based on a prior study in multiple sclerosis that found that the societal value to healthy individuals at risk for multiple sclerosis was approximately 80% of the value estimated for patients already diagnosed. Applying this ratio, Shafrin et al²⁰ scaled the traditionally estimated value of nivolumab for patients to approximate its insurance value for the healthy population. However, because this estimate was not empirically derived from a lung cancer cohort, it should be interpreted cautiously, as it assumes that societal preferences and risk perceptions are comparable across disease areas. The value of hope was quantified by estimating the certainty equivalent of survival outcomes based on risk preferences of patients with NSCLC, adding an extra 0.04 QALYs beyond the expected gain of 0.66 QALYs. Option value was estimated using a registry-based study that projected future survival gains among patients with NSCLC and RCC. The final net monetary benefit was derived as the difference between the overall monetary benefit and the entire incremental costs. The addition of ROV contributed to additional 0.037 QALYs or $5,599 of survival gains.

Other Cancer Types.

Sanchez et al²⁴ analyzed historical and projected survival trends to evaluate the ROV of TKIs in CML. Survival benefits associated with TKI therapy were assessed using SEER data, and future survival gains were forecasted employing the Lee-Carter approach. The research indicated that the option value added 0.76 LYs, equivalent to $63,000, representing 9% of the survival benefits of current medicines. Significantly, 96% of the option value stemmed from prospective therapy advancements.²⁶

Snider et al²³ assessed the LYs saved from preventative tamoxifen in breast cancer. Using a traditional approach, survival benefits were evaluated using Taxol combined with anthracyclines as the first treatment upon disease onset. The ROV approach evaluated the likelihood that tamoxifen adequately slowed down disease development for patients to access Herceptin (trastuzumab). The results showed that the ROV increased the expected economic benefit of tamoxifen by 24%, and the willingness-to-pay changed from $200,339 to $247,913, thus emphasizing the significance of including future medical advancements in economic analysis.

Finally, Thijssen et al²² employed a Bayesian decision model to assess the adoption of robot-assisted vs standard laparoscopic prostatectomy. The Bayesian decision model facilitates the integration of new evidence over time to revise prior views regarding an intervention’s cost-effectiveness, thus enabling dynamic decision-making amid unpredictability. The results demonstrated that if robot-assisted surgery proved cost-effective, trials would facilitate an immediate adoption decision within approximately 1 month; conversely, if the technique lacked cost-effectiveness, trials might theoretically continue indefinitely because of little costs and possible future benefits.

FORECASTING APPROACH

All studies assessed the value of the medical therapy, including the ROV, by calculating the difference in survival or QALYs between treatments while also factoring in any additional health benefits that patients may receive from future advancements. An exception to this is the study by Li et al,¹⁸ which quantified the use of existing treatment in metastatic melanoma following phase 2 and 3 disclosures of ipilimumab results. Nevertheless, the data and modeling approach employed in assessing the option value differed based on the perspective. Studies that adopted the ex post perspective used administrative data to compute survival curves, accounting for technological advancements’ influence on patient survival at the time of their introduction.^16,21,23 However, these studies may not always isolate the impact of the treatment innovation alone. Studies that fail to adjust for confounding factors like improved supportive care, diagnostics, or broader health system changes risk overestimating survival gains attributable to the new treatment. As such, the calculated ROV may reflect broader trends unrelated to the specific innovation. On the other hand, ex ante studies used methodologies to predict the impact of forthcoming medical advancements on survival.^{1,14,17–20,22,23,25}

Currently, 2 primary forecasting systems have become significant. Sanchez et al²⁴ and Snider et al²¹ employed the forecasting approach initially introduced by Lee and Carter. The Lee-Carter model is a time-series model originally developed for demographic studies, such as projecting decreases in mortality rates over an extended period.²⁶ Also, Li et al¹⁷ revealed 2 approaches: “fitting Cox proportional hazard models on registry data (trend approach); and estimating the timing, approval of pipeline treatments based on early clinical trials (pipeline approach).” The second approach involved the assumption that disease-specific mortality would decline solely upon introducing a new treatment. Li et al¹⁷ reviewed phase 3 trials registered on ClinicalTrials.gov at the time of the launch of the current treatment (in 2011 for ipilimumab) to project treatments in the pipeline. Sanchez et al²⁴ and Li et al¹⁷ calculated the ROV by accounting for advancements in disease-specific mortality in addition to overall mortality. The inclusion of both general and disease-specific mortality trends was justified by Li et al¹⁷ to demonstrate that the observed decrease in all-cause mortality among patients with unresectable or metastatic melanoma was largely driven by reductions in melanoma-specific mortality. Snider et al²¹ estimated ROV by first generating pre-nivolumab survival curves from registry data, applying trial-based hazard ratios, and forecasting future gains using the Lee-Carter method; the option value was then calculated as the difference in life expectancy with nivolumab under scenarios with and without future innovation.

Discussion

ROV demonstrates the additional benefit patients gain by enduring long enough to access advanced medicines. ROV is particularly relevant to cancer, where recent advances in immunotherapy, targeted therapy, and precision medicine have transformed the prognosis of several tumors. This systematic review synthesized the current knowledge on the application of ROV in oncology, highlighting its influence on multiple cancer types.

Across diverse cancer types, such as metastatic melanoma, NSCLC, RCC, and CML, ROV has consistently demonstrated superior survival benefits, augmented QALYs, and improved cost-effectiveness outcomes. Metastatic melanoma emerged as a significant focus in ROV research, underscoring its unique treatment landscape. Its predominance in studies evaluating ROV is mainly because of the rapid and transformative advancements in its treatment options, particularly with the introduction of immune checkpoint inhibitors such as ipilimumab and PD-1 inhibitors. Lee et al¹⁶ asserted that “Ipilimumab was chosen as a case due to its substantial significance for ROV, owing to the rapid progress in metastatic melanoma.” These novel therapies have significantly improved survival rates, offering clear possibilities to assess the additional value patients gain from extended life—a crucial aspect of ROV. The ROV framework is currently in development, and “metastatic melanoma provides an ideal setting to explore the use of an ROV approach to assess how innovations that prolong survival can enable patients to benefit from the next novel treatment.”¹⁵

Ex post and ex ante modeling approaches applied to assess ROV highlight the adaptability of these methods to elucidate various aspects of value in oncology. Ex post models, exemplified by the works of Lee et al,¹⁶ Wong et al,¹⁵ and Snider et al,¹⁹ focus on realized benefits by retrospectively evaluating the value of therapies following the introduction of subsequent innovations. The ex post method is significant for real-world data analyses, enabling researchers to measure the actual survival benefits seen in patients who have used additional lines of therapy. This approach assists in illustrating the real-world impact of these novel therapies.¹⁵ Conversely, ex ante models provide a forward-looking perspective by estimating the potential future benefits of therapies based on anticipated advancements.^{1,17,21,24,25} Ex ante models offer decision-makers a thorough understanding of a therapy’s broader impact by incorporating the prospective benefits of the novel therapy. For prospective reimbursement decision-making, “it is imperative to estimate the ROV of treatments through an ex ante approach as opposed to an ex post approach.”^1,3

The inclusion of ROV in economic evaluations raises important methodological considerations, particularly in relation to discounting. Whereas discounting is a standard HTA practice used to convert future costs and health benefits into present values, reflecting time preference and opportunity cost, ROV captures an entirely distinct concept. Discounting adjusts for the time at which benefits are realized, whereas ROV recognizes that surviving longer may open access to additional therapeutic options in the future, which generate incremental health gains not included in the standard cost-effectiveness framework. In this sense, discounting and ROV are complementary rather than redundant.

From the review, the integration of ROV into cost-effectiveness analyses (CEAs) may potentially facilitate decision-making regarding access to innovative therapies. Li et al²⁷ emphasize that “ROV may be compared to a benchmark and presented with other product attributes to facilitate a deliberative voting process regarding the product’s value. For inclusion in economic assessments, it is hypothesized that ex ante estimation of the future treatment values can help decision-makers in reimbursement decision-making due to its modeling nature and assumptions made.”²

LIMITATIONS

ROV is increasingly recognized in oncology; however, there remain gaps in its application in oncology. Its application to prevalent high-burden cancers including breast, colorectal, and prostate cancers stands out as a gap. The current literature is dominated by metastatic melanoma and NSCLC, neglecting cancers that represent a substantial public health and economic burden.

Current ROV assessments focus heavily on type 1 ROV—survival gains enabling access to future treatments—while often neglecting type 2 (reducing complications/improving quality of life) and type 3 (influencing future treatment effectiveness). These are critical in cancers like prostate, ovarian, and colorectal, in which benefits may not extend survival. Patient-reported outcomes can help capture type 2 ROV. For type 3, real-world data on tumor evolution and resistance, such as T790M from EGFR-TKIs²⁸ or T-cell depletion after immunotherapy,²⁹ are key. Broader integration of these mechanisms is essential to fully reflect treatment value in oncology.

Another limitation involves the consistency of methodologies and assumptions across studies. Variability in modeling approaches, assumptions about innovation timelines, and survival probabilities may result in discrepancies between ROV estimates. Wong et al¹⁵ and Snider et al¹⁹ used different assumptions regarding the timing of subsequent immunotherapies, resulting in varying magnitudes of ROV for similar treatment scenarios. These inconsistencies require standardized frameworks/guidelines to ensure comparability/reliability among ROV analyses.

Furthermore, insufficient emphasis has been placed on incorporating future innovation costs and their effects on ROV evaluation.³⁰ Several studies, including Lee et al,¹⁶ did not include future innovative drug costs in ROV estimation because they aimed to measure ROV as health gains. In a decision-making setting in which care costs are crucial, “it is essential to consider estimating the ROV that incorporates both the clinical and the economic impact of future innovations.”⁹ Further research is required to develop and implement methodologies for forecasting future costs.

The current review highlights the absence of standardized reporting frameworks for ROV studies. This lack of uniformity in methodology and reporting makes it difficult to compare results and draw consistent conclusions systematically. Recently, Li and Garrison³⁰ developed a 15-item checklist to guide ROV assessment reporting. A standardized reporting framework must be adopted to improve the consistency of reporting study results for decision-making.

An additional concern surrounding ROV integration in value assessment is the risk of double-counting benefits. Many HTA agencies and decision-makers already incorporate elements of flexibility through scenario and sensitivity analyses; therefore, there is the impression that “ROV cannot be estimated without significant risk of attributing benefits that are already captured.”³¹ Double-counting risk is an inherent challenge in CEA methodology. Traditional CEAs often face similar risks, particularly when incorporating indirect benefits.³¹ It is crucial to establish a transparent framework that distinguishes the unique contributions of ROV. Analysts must carefully define costs and benefits, ensuring well-structured relationships between them. Additionally, selecting appropriate effectiveness measures is critical; maintaining consistency between the numerator and denominator in cost-effectiveness calcu-lations helps prevent overlapping valuations and enhances analytical accuracy.³¹

Conclusions

This systematic review demonstrates the growing importance of ROV in oncology; significant opportunities remain to expand and refine its application. One critical area for future research is developing and standardizing ROV valuation methods. The methodological diversity observed in the reviewed studies—ranging from ex ante to ex post models—reflects the flexibility of ROV frameworks but also underscores the need for consistent approaches. A standardized framework for ROV estimation would facilitate the comparability of findings across studies and therapeutic areas and increase the reliability of analyses to inform health care decision-making.

Disclosures

The authors received no funding or sponsorship for the conduct of this systematic review.

All authors declare that they have no financial or nonfinancial competing interests that could have influenced the findings of this systematic review.

Data Availability

This review did not generate any new data sets or code. All data used were extracted from publicly available published studies.