As early as 2004, Dr Deborah Schrag of the Memorial Sloan Kettering Cancer Center, New York, warned of the rising costs of cancer care (1). Dr Schrag pointed out that the median survival time of advanced colorectal cancer patients was notably improved, almost doubled in the last decade, thanks to the advent of new anticancer agents, such as irinotecan, oxaliplatin, bevacizumab and cetuximab. However, the progress was accompanied by a 340-fold increase in drug costs (1).
In 2004, some target-based drugs, such as imatinib, gefitinib and erlotinib were available. In retrospect, however, this was just the end of the beginning. The epidermal growth factor receptor (EGFR)-activating mutation, the true target of EGFR tyrosine kinase inhibitors (TKIs), was discovered that year. Anaplastic lymphoma kinase (ALK) fusion in lung cancer was not reported until 2007. In 2004, we did not have, or even know of, immune-checkpoint inhibitors (ICIs). We did not have antibody-drug conjugates. Nor did we know CAR-T.
Even before we began to use the full armamentarium we now have, in 2011, researchers from Europe and North America issued a statement warning that the skyrocketing cost of cancer care would endanger its affordability, even in high-income countries (2). Please note that this was 3 years before nivolumab was launched. They concluded that ‘the cancer profession and industry should take responsibility and not accept a substandard evidence base and an ethos of very small benefit at whatever cost; rather, we need delivery of fair prices and real value from new technologies’ (2). Has their recommendation been widely heeded? Probably not.
In Japan, medical costs in 2021 exceeded 45 trillion Japanese yen (JPY), a 4.8% increase compared to 2020 and ∼a 35% increase compared to 2004. We used 8.18% of our Gross Domestic Product (GDP) on medical care, which was 7.99% and 6% in 2020 and 2004, respectively (3). Medical costs in Japan have more than tripled since 1986, when one of the authors (H.K.) graduated from medical school, while Japanese GDP has stagnated over the past three decades. When the other author (T.K.) graduated from medical school in 1967, Japan’s medical costs were <4 trillion JPY, ∼3.5% of the GDP. Is such an increase sustainable? Definitely not.
But why is this happening? Medical cost increases are mainly driven by two factors: progress in medicine, such as the advent of new drugs, and the aging population. Nobody is to blame for either of them—and nobody can stop them. However, if left unchecked, we will soon witness a total collapse of our healthcare system. The Japanese public insurance system allows us to spend as much money as necessary to treat our patients. Although the Japanese constitution guarantees the right to maintain ‘the minimum’ standard of living, we have provided ‘optimum’ medical care for everyone. It is highly unlikely that we can go on as we have, due, alas, to a lack of money.
What can we do? To be honest with you, we do not know. Then what should we do? First, we need to understand the situation for what it is. We physicians did not care about money and tried to ignore inconvenient truths. However, understanding ourselves and understanding what we are doing is the first step we need to take. How much money do we spend on the care of our patients? Unfortunately, we are no longer living in a world where every expense can be justified in the name of ‘patient care’. We need to look at how we can curb this excessive use of resources. But first, we have to understand what we are actually doing.
In a series of articles accompanying this Editorial, nine subgroups of the Japan Clinical Oncology Group (JCOG) conducted investigations on first-line treatment selections during 2021–2022 and their associated costs in a total of 17 cancer types (4–12). A total of 15 564 patients (29% of them are aged 75 years old or more) were surveyed in probably one of the largest surveys of this kind in the 2020s and certainly the first ever in Japan. Although there are substantial variations according to each specific cancer, we could observe several common features from the data.
First, the monthly cost is high both in absolute and relative terms. Among the 17 tumor types, a median of 59% of the patients (1QR: 44% to 3QR: 87%) received treatments with monthly costs of 500 000 JPY or more. A total of 17% of the patients received treatments with monthly costs of 1 000 000 JPY or more (Fig. 1). As compared to conventional chemotherapy, which was the standard of care 10–15 years ago, there were 10- to 50-fold cost increases.
Proportions of patients who receive treatments at monthly costs of 500 000 JPY and 1 000 000 JPY, according to tumor type.
Elderly patients received high-cost treatments less frequently, although this was apparently due to concerns over toxicities rather than cost. In fact, high-cost regimens such as nivolumab + ipilimumab in non-small cell lung cancers (7) or nivolumab/pembrolizumab in head and neck cancers (12) are more frequently used in patients of 75 years or older.
Some ‘effective’ drugs such as androgen receptor signaling inhibitors for prostate cancer and CDK4/6 inhibitors for breast cancer are used for extended periods, because of longer progression-free survival, leading to high total (not only monthly) costs (4,11). This kind of phenomenon was previously reported on pertuzumab, an anti-HER2 antibody (13).
The ‘benefits’ of the high-cost treatments were highly variable. It could be substantial (such as EGFR-TKIs in EGFR-mutated non-small cell lung cancer), marginal [such as nivolumab to gastric cancer (8)], or nonexistent [such as bevacizumab to glioblastoma (11)]. In most of the cases, lack of adequate benefit does not affect the drug price, since it is the result of an indication-extension trial, and the price had already been determined at the original indication approval.
Oftentimes, more than one drugs of the same class are available. This includes androgen receptor signaling inhibitors for prostate cancer (4), ICIs and TKIs for renal cell carcinoma (4), ICIs for hepatocellular carcinoma (6), ICIs, TKIs and anti-vascular-endothelial growth factor antibodies for lung cancers (7), CDK4/6 inhibitors for breast cancer (10,14) and anti-EGFR and anti-VEGR antibodies in colon cancer (9). Drug prices can vary greatly among the same class, despite exhibiting the same efficacy and toxicities. This again is mainly a result of an indication-extension trial and the price having already been determined at the time of the original indication approval. For example, both ramucirumab and bevacizumab are available for second-line treatments for colorectal cancers, with almost identical efficacy and toxicities, but ramucirumab is many times more expensive than bevacizumab (15–17). The use of ramucirumab in colorectal cancers is strongly discouraged in the literature (15,16), with the speculation that the pharmaceutical company wants to maintain the high price for the lead indication of gastric cancer (17).
We all sense that cancer care is unsustainable and soon to become unaffordable, but there is no easy way out. We are very reluctant to discuss cost-effectiveness. For example, suppose you find a report that the incremental cost-effectiveness ratio for the treatment you are going to give to your patient is 500 000 US dollars per quality-adjusted life year, far exceeding the willing-to-pay threshold of any country, including Japan. But could you ‘abandon’ the patient, just because it is too expensive? Could you explain this to your patient? Human life is ‘priceless’, isn’t it? How can we balance our duties to patients versus society? (18,19) In addition, cost-effectiveness analyses reports are not reliable. Industry-sponsored studies are far more likely to give ‘cost-effective’ reports (20). What can we believe?
Still, there should be something we could do, or at least something we could start with. For example, we could move away from the idea of ‘care at any cost’ and stop using expensive drugs which provide no clear additional benefit. This can be done without any discussion on cost-effectiveness, since there is no (additional) ‘effectiveness’. These drugs include bevacizumab for glioblastoma [as one of the JCOG reports pointed out (11)], or ramucirumab for colorectal cancer (15–17). Ramucirumab can be substituted with bevacizumab, or even better, by bevacizumab biosimilar, which is a much less expensive alternative, without any compromise in efficacy.
As a next step, perhaps we could initiate some discussions. Panitumumab and cetuximab, both anti-EGFR antibodies, are used in Rat sarcoma virus (RAS)-wild colorectal cancer, with similar efficacies (9). Panitumumab is double the price of cetuximab. The only difference appears to be that panitumumab is administered bi-weekly, whereas cetuximab is given weekly. Does this difference, which amounts to a matter of convenience for the patients, justify the huge increase in cost? Or simply, how about giving cetuximab bi-weekly? We have talked about efficacy, toxicity and convenience. It is time to add cost to our list of discussion points.
Last but not the least, we should perform more and more research on treatment optimization. The dosage of modern cancer drugs might be suboptimally determined, by adopting classic dose-increase studies with cytotoxic agents (21). Food and Drug Administration (FDA) has launched an initiative, Project Optimus, to reform the dose optimization and dose selection paradigm (https://www.fda.gov/about-fda/oncology-center-excellence/project-optimus). In addition to dose reduction, many trials evaluate de-escalation strategies, including shortened duration, longer intervals and more (21–26). Some studies have been launched in Japan, including those conducted at JCOG (27).
Life is priceless. But our resources are limited. We have to use the limited resources wisely to treat our patients who are all mortal. We used to believe that it was not our responsibility to think about cost, resources, affordability and sustainability. In fact, it is. If we are to avoid the collapse of the Japanese Health Insurance System, we need to start thinking seriously about medical costs. We hope you can agree on this.
Conflict of interest
None declared.
Funding
None declared.
Disclaimer
Dr Kunitoh is the Editor-in-Chief, and Dr Kakizoe is the Emeritus Editor-in-Chief of JJCO.
However, the opinions expressed in the Editorials are their personal opinions and do not necessarily reflect the views of the journal or publisher.
