The heart is already at risk before the first dose. The treatment adds to it.

What cardio-oncology is

Cardio-oncology is the cardiovascular care of patients with cancer: before treatment, during it, and for as long afterwards as the risk lasts. The difficulty is that the risk arrives from two directions, and only one of them is the treatment.

Cancer and cardiovascular disease are usually taught as separate problems, and they are not. They share most of their risk factors, from smoking and obesity to diabetes, and chronic inflammation runs through the development of both (Koene, Circulation 2016; Wilcox, Nat Rev Cardiol 2024). The overlap is specific enough to measure. Among 20,305 adults followed for fifteen years, the score used to estimate someone's ten-year risk of atherosclerotic cardiovascular disease also predicted their risk of developing cancer (Lau, JACC CardioOncol 2021). And at least one lesion drives both: clonal hematopoiesis, an expanded blood-cell clone that raises the risk of blood cancer, roughly doubles the risk of coronary disease as well (Jaiswal, N Engl J Med 2017). A patient arriving in a cardio-oncology clinic has often been accumulating cardiovascular risk for years before any oncologist saw them.

The treatment then adds to it, and cancer therapy has improved enough to change how much that matters. Five-year survival across all cancers in the United States has risen from about half in the mid-1970s to about seven in ten today (Siegel, CA Cancer J Clin 2026), and 18.6 million Americans are now living with a history of cancer (Schapira, CA Cancer J Clin 2025). Those survivors carry whatever risk they began with, plus whatever the treatment added, for years and often decades after the cancer itself is gone. The two contributions are not even independent: how much cardiac risk a patient already carries shapes how much the treatment will cost them (Koene, Circulation 2016).

The field is younger than many of its patients' diagnoses. Its professional society was founded in 2009, its first dedicated journal launched in 2019, Europe's first practice guideline appeared only in 2022 (Lyon, Eur Heart J 2022), and there is still no board certification in it. Much of what a cardio-oncologist should do has had to be written down for the first time by the people doing it.

What cancer survivors die of

Follow enough people for long enough after a cancer diagnosis and a second cause of death emerges from behind the first.

Among 3.2 million patients followed through the United States national cancer registry, 11.3% died of cardiovascular disease rather than of their cancer (Sturgeon, Eur Heart J 2019). The average hides the shape of it. For cancers of the bladder, larynx, prostate and endometrium the figure runs between one in six and one in five, and by 2012 cardiovascular disease had become the leading cause of death among people diagnosed with prostate, thyroid or testicular cancer and with Hodgkin lymphoma. Because breast and prostate cancer are so common, survivors of those two alone account for about half of every cardiovascular death after cancer.

The crossover has a timescale. In a large English cohort, cardiovascular death overtook cancer death about 12.7 years after a breast cancer diagnosis in women aged 60 to 79, and about 7 years after an endometrial one (Strongman, JACC CardioOncol 2022). Among patients diagnosed after 80 it overtook for every cancer studied. Below 60 it never did, except after endometrial cancer.

None of this is by itself an indictment of the treatment, and reading it that way is the commonest mistake made with these numbers. A registry records what people died of, not what they were exposed to, and given how much biology the two diseases share, a good deal of that burden would have arrived without any oncologist's help. What the figures establish is where a large share of a survivor's remaining risk sits. Separating the part the treatment caused from the part that was already there takes a different kind of study, and it is why this program runs alongside the lab's work on inflammation rather than apart from it.

Three injuries, three different problems

Cancer therapies damage the heart by mechanisms with almost nothing in common. No single test, and no single schedule, can catch them all.

Immune checkpoint inhibitors take the brakes off T cells, and in about one patient in a hundred those T cells attack the heart muscle (Mahmood, J Am Coll Cardiol 2018). It is the most lethal of the immune-related toxicities, and it is fast: the typical patient presents around five weeks after the first dose, and some after a single one (Salem, Lancet Oncol 2018). Almost nothing about who will get it can be known beforehand, beyond the raised risk from combining two checkpoint drugs and from a diseased thymus.

Anthracyclines injure the muscle directly and cumulatively, and here the risk genuinely can be anticipated, because it tracks the total dose given. At a cumulative dose of 550 mg/m² of doxorubicin about one patient in four develops heart failure, roughly four times the estimate the field had worked from until someone went back and counted (Swain, Cancer 2003). The timing is also tighter than it was long taught to be. Among 2,625 patients followed prospectively, 9% developed a fall in heart function, and 98% of that appeared inside the first year (Cardinale, Circulation 2015). HER2-directed therapy injures differently again: not dose-related, and usually reversible once the drug is stopped.

Cellular therapies produce the last pattern. CAR T-cells can set off a body-wide cytokine storm, and the cardiac injury follows the storm rather than any dose: in one cohort every cardiovascular event occurred in a patient with at least moderate cytokine release syndrome, at a median of three weeks (Alvi, J Am Coll Cardiol 2019), and heart function falls within days of infusion and recovers in most of those affected (Ganatra, Circulation 2020).

Read together, those are three separate research problems. Rare, fast and unpredictable needs surveillance with a sensitive test. Cumulative and dose-driven needs risk stratification before the first dose. Cytokine-driven needs monitoring through the acute illness. Very little that is built for one transfers to the others.

Getting the patient through the treatment

The job is not to rule on who may be treated. It is to get the heart ready for what is coming, keep it going through, and look after it once the oncologist has finished.

Stem cell transplantation cures diseases that nothing else will, and it is also one of the most demanding things medicine does to a body. The cardiologist's task is to work out what that heart can take, treat what is treatable first, and get the patient to the point of being able to have it. Too often the assessment stops at a single number instead. Of 26 transplant centers surveyed in the United Kingdom, 81% used an ejection fraction threshold to exclude patients, and the threshold ranged from 50% at one center to 30% at another, with no justification offered for the difference (Gent, EJHaem 2022).

That range is the problem in miniature: no guideline sets the number, and where centers have gone back and compared patients transplanted with a reduced ejection fraction against matched patients with a normal one, complication rates and early mortality came out much the same (Qazilbash, Biol Blood Marrow Transplant 2009). The 2024 American Heart Association statement on this question, which this lab led, holds that refusing transplantation for cardiovascular reasons should be reserved for the occasional patient with severe untreatable disease or a life expectancy under a year (Hayek, Circulation 2024).

The evidence is not all one way and should not be made to sound like it: a recent cohort did find a reduced ejection fraction predicted cardiovascular events in the first hundred days (Aghel, JACC Adv 2025). But a risk worth knowing about is a reason to manage it, not a reason to withdraw the treatment, and that is what a risk score is for: one derived in 2,435 transplant recipients and validated in 919 more uses age, transplant type, prior anthracycline dose and vascular history, and no ejection fraction at all (Vasbinder, J Am Heart Assoc 2024). Knowing which patients will struggle is what makes it possible to watch them closely, treat them early, and carry them through.

Watching survivors, decades later

The risk does not end when the treatment does, and neither does the uncertainty about what to do with that fact.

Radiation gives the clearest picture, because the exposure can be counted in units. Among women irradiated for breast cancer, the rate of major coronary events rose 7.4% for every additional gray delivered to the heart, beginning within five years and still climbing past twenty, with no dose below which the effect disappeared (Darby, N Engl J Med 2013). Chemotherapy leaves a longer, quieter tail. When adults treated for cancer in childhood were examined decades afterwards, a substantial minority were found to have cardiomyopathy or valve disease on the study's own echocardiogram, and almost none of them had symptoms (Mulrooney, Ann Intern Med 2016).

What should follow from that is far less settled than the risk itself. Guidelines do recommend surveillance and are candid that they rest on expert consensus: the American Society of Clinical Oncology states plainly that beyond a single echocardiogram in the year after treatment, the evidence is insufficient to recommend any particular frequency or duration for a survivor without symptoms (Armenian, J Clin Oncol 2017). No trial has shown that scheduled surveillance changes how survivors actually fare. And what is recommended is largely not done, with roughly four in ten of the highest-risk survivors of childhood cancer receiving the cardiac surveillance intended for them (Yan, J Clin Oncol 2020).

This is where a blood test would help most and is weakest. Measuring a biomarker during treatment has a real evidence base behind it; measuring one in a survivor ten years later mostly does not. The American Heart Association statement on biomarkers in cancer survivors, which this lab helped write, separates those two uses deliberately, because they are not the same claim (Zaha, Circulation 2021). Establishing what a marker means in someone whose baseline is not the baseline of a cardiology clinic is the work, and it is unfinished.

Building the field

Some of the work is not a study. It is deciding what the practice should be, and training the people who will carry it out.

When centers were first surveyed about this in 2014, about a quarter had a cardio-oncology service. Five years later half did, but of the fifty-three programs then in existence only nine offered any training in it (Hayek, J Am Coll Cardiol 2019). That gap, between the number of patients who need this care and the number of people qualified to give it, is the field's rate-limiting step, and it is why a training curriculum had to be written from nothing (Alvarez-Cardona, J Am Coll Cardiol 2020).

For this lab that has meant chairing the American Heart Association statement on cardiovascular care through stem cell transplantation (Hayek, Circulation 2024), serving as vice chair of its statement on biomarkers in cancer survivors (Zaha, Circulation 2021), and making the case for the training pathway itself (Hayek, J Am Coll Cardiol 2019). Work of that kind produces no result of its own. It produces the conditions under which everyone else's results can be acted on.