HFA-ICOS cardio-oncology baseline risk assessment for VEGF inhibitors

Clinical background

VEGF inhibitors, including tyrosine kinase inhibitors such as sunitinib, sorafenib, pazopanib and lenvatinib as well as monoclonal antibodies such as bevacizumab, are effective against a range of solid tumours but carry a broad profile of cardiovascular toxicity. Hypertension occurs in up to 80% of treated patients, left ventricular dysfunction and heart failure in about 15 to 20%, and arterial as well as venous thromboembolic events, QTc prolongation and arrhythmias occur at varying rates [2]. Toxicity appears early: in a prospective study of 78 patients starting a VEGF inhibitor, 93% of the cases of cancer therapy-related cardiac dysfunction that occurred were already manifest at 4 weeks [3].

The decision the instrument serves is how intensively the patient's heart should be monitored during treatment, and whether referral to cardio-oncology is needed before starting. Without a structured baseline assessment, monitoring risks becoming arbitrary, either too intensive in patients at low risk or insufficient in patients at high risk, in whom early detection of dysfunction is decisive.

Applying the HFA-ICOS baseline cardiovascular risk assessment

The instrument was produced by the Cardio-Oncology Study Group of the Heart Failure Association of the ESC in collaboration with the International Cardio-Oncology Society and published in 2020 as a position statement [1]. It is expert consensus-based, not statistically derived from a specific cohort. The VEGF inhibitor proforma is one of seven therapy-specific risk assessments presented in the same document; the others cover anthracyclines, HER2-targeted therapies, BCR-ABL inhibitors, myeloma therapies, RAF/MEK inhibitors and androgen deprivation therapy.

The classification is hierarchical and rests on three levels of risk factors:

Very high risk factors (each overrides the total score to very high risk):

• Existing heart failure or cardiomyopathy
• Arterial vascular disease (ischaemic heart disease, previous PCI/CABG, stable angina, TIA, stroke or peripheral arterial disease)
• Venous thrombosis (DVT or pulmonary embolism)
• LVEF <50% at baseline
• QTc ≥480 ms
• Age ≥75 years
• Previous anthracycline exposure

High risk factors (each classifies the patient as high risk if no very high risk factor is present; weighted 2 points in the medium-risk score):

• Borderline LVEF 50 to 54%
• QTc 450 to 480 ms (men) or 460 to 480 ms (women)
• Arrhythmia

Medium risk factors (weighted 1 point each):

• Hypertension
• Raised troponin at baseline
• Raised BNP or NT-proBNP at baseline
• Age 65 to 74 years
• Diabetes mellitus
• Hyperlipidaemia
• Chronic kidney disease
• Proteinuria
• Previous radiotherapy to the left chest or mediastinum
• Current smoker or significant smoking history
• Obesity (BMI >30 kg/m²)

The overall classification follows this logic:

$$\text{Risk} = \begin{cases} \text{Very high} & \text{if any very high risk factor is present} \ \text{High} & \text{if any high risk factor is present or the medium-risk score is} \geq 5 \ \text{Moderate} & \text{if the medium-risk score} = 2\text{--}4 \ \text{Low} & \text{if the medium-risk score} = 0\text{--}1 \end{cases}$$

The underlying basis is therefore not a statistical model but an expert-driven risk matrix. The factors were chosen from published evidence that they contribute to cardiovascular risk in patients about to be treated with a VEGF inhibitor, and the weighting (1 or 2 points) reflects expert judgement of their relative importance [1].

Interpretation in practice

A central principle of the position statement is that risk classification should not lead to effective cancer treatment being withheld, except at high or very high risk and only after multidisciplinary discussion [1]. The purpose is to identify patients who need enhanced monitoring and cardioprotective intervention, not to act as an exclusion criterion.

Validation and performance

The instrument is expert-based and has no original derivation cohort in the statistical sense. External validation has been published for other therapy classes within the same HFA-ICOS framework, but for VEGF inhibitors specifically the data are limited.

The most relevant prospective study for VEGF inhibitors to date included 78 patients (mean age 63 years, 68% men) with renal cell carcinoma, hepatocellular carcinoma, sarcoma or thyroid cancer who started a VEGF inhibitor, alone or in combination with immunotherapy [3]. HFA-ICOS was used for baseline classification: 45% of patients fell into the high or very high risk categories. During 24 weeks of follow-up, 19% developed cancer therapy-related cardiac dysfunction (an LVEF fall of ≥10 percentage points to <50%) and 77% developed hypertension. Toxicity appeared early, at 4 weeks in 93% of cases. The study was not powered to calculate a c-statistic for HFA-ICOS, and no association between baseline risk category and outcome was reported separately. Cardiac MRI did, however, show that the mechanism behind the fall in LVEF included microvascular dysfunction and reduced myocardial perfusion, partly independent of the rise in blood pressure [3].

For anthracyclines, a large external validation has been performed in the CARDIOTOX registry of 1,066 patients [4]. It showed that HFA-ICOS produced a stepwise increase in the incidence of symptomatic or moderate to severe asymptomatic cardiac dysfunction across risk categories, with a hazard ratio of 28.7 for very high risk compared with low risk. For trastuzumab, a retrospective cohort of 931 patients showed that the rate of cardiotoxicity rose from 14.0% at low risk to 30.3% at high/very high risk (p = 0.002) [5]. These validations concern other drug classes and cannot be extrapolated directly to VEGF inhibitors, but they support the predictive validity of the instrument's overall structure.

A summarising review from 2025 confirms that the 2022 ESC cardio-oncology guidelines recommend HFA-ICOS for baseline risk assessment and that validation has so far been carried out for anthracyclines, HER2-targeted therapies and BCR-ABL inhibitors, while VEGF inhibitors still lack formal predictive validation [6].

Limitations

The instrument is not statistically derived, meaning that the scoring and thresholds rest on expert consensus rather than on calibrated risk models. It therefore lacks c-statistics and calibration data for VEGF inhibitors specifically.

The following patient groups and situations are not covered:

• Patients with manifest heart failure or an LVEF <50% are automatically classified as very high risk. The instrument gives no finer differentiation within this group, where the decision is instead about optimising heart failure treatment and choosing the cancer therapy.
• Combination treatment with immunotherapy (checkpoint inhibitors) is not part of the proforma's original design. In the prospective Glasgow study, 49% of patients were treated with a VEGF inhibitor combined with immunotherapy, but the instrument was not validated separately in this group [3].
• Paediatric patients are not covered, since the age criteria and risk factor profile are designed for adults.

Common misuses:

• Interpreting the risk category as an absolute contraindication tool. It is intended to guide monitoring intensity and referral pathways, not to justify withholding cancer treatment except after multidisciplinary discussion at high or very high risk [1].
• Using the VEGF inhibitor proforma for patients due to receive other cardiotoxic drugs. Each therapy class has its own proforma with different risk factor weightings, and the results are not interchangeable.
• Failing to reassess risk during treatment. Hypertension that develops during VEGF inhibitor treatment is itself a risk factor for further toxicity and should trigger an updated risk assessment, not merely antihypertensive treatment.

Implementation in practice

Cardio-oncology services are typically concentrated in regional cancer centres and university hospitals. The HFA-ICOS proformas have not been incorporated into national guidelines in most countries, but the 2022 ESC cardio-oncology guidelines, which recommend HFA-ICOS, are applied in clinical practice at cardio-oncology units. Access to echocardiography and cardiac MRI varies between hospitals, which may affect how closely patients in the high and very high risk categories can in practice be monitored.

References

1. Lyon AR, Dent S, Stanway S, et al. Baseline cardiovascular risk assessment in cancer patients scheduled to receive cardiotoxic cancer therapies: a position statement and new risk assessment tools from the Cardio-Oncology Study Group of the Heart Failure Association of the ESC in collaboration with the International Cardio-Oncology Society. Eur J Heart Fail. 2020;22(11):1945-1960. PMID: 32463967
2. Dobbin SJH, Petrie MC, Myles RC, et al. Cardiotoxic effects of angiogenesis inhibitors. Clin Sci (Lond). 2021;135(1):71-100. PMID: 33404052
3. Dobbin SJH, Mangion K, Berry C, et al. Vascular endothelial growth factor inhibitor-induced cardiotoxicity: prospective multimodality assessment incorporating cardiovascular magnetic resonance imaging. Heart. 2025;111(19). PMID: 40180444
4. Rivero-Santana B, Saldaña-García J, Caro-Codón J, et al. Anthracycline-induced cardiovascular toxicity: validation of the Heart Failure Association and International Cardio-Oncology Society risk score. Eur Heart J. 2025;46(3):273-284. PMID: 39106857
5. Battisti NML, Andres MS, Lee KA, et al. Incidence of cardiotoxicity and validation of the Heart Failure Association-International Cardio-Oncology Society risk stratification tool in patients treated with trastuzumab for HER2-positive early breast cancer. Breast Cancer Res Treat. 2021;188(1):149-163. PMID: 33818652
6. Tong J, Senechal I, Ramalingam S, et al. Risk Assessment Prior to Cardiotoxic Anticancer Therapies in 7 Steps. Br J Hosp Med (Lond). 2025;86(1):1-21. PMID: 39862029