Part XV · Supportive care, survivorship, and patient experience · Chapter 84

Treatment toxicity and supportive care

Organised by modality, because that is how the risk arrives and how the monitoring has to be built.

1 · Acute and chronic cytotoxic toxicity

Myelosuppression sets the schedule, and the drug classes that produce it are described in Cytotoxic chemotherapy. Febrile neutropenia risk is a property of the regimen rather than of the patient. That is why primary growth factor prophylaxis is decided at the point of regimen selection rather than after the first episode. Dose intensity and its trade-offs are in Dose intensity, dose density, and dose modification.

Nausea and vomiting are now largely preventable and the failure mode is undertreatment of a regimen whose emetic risk was misclassified. Prophylaxis is selected from the emetic risk category of the regimen, and an anthracycline with cyclophosphamide sits in the highly emetogenic group1. Delayed emesis is undertreated more often than acute emesis because it happens after the patient has gone home.

Alopecia has a prevention with randomised evidence. One trial enrolled women with stage I or II disease receiving a taxane, an anthracycline, or both. Scalp cooling made hair loss of less than 50% after the fourth cycle significantly more likely than no cooling2. In a parallel study of non-anthracycline adjuvant regimens, scalp cooling reduced hair loss measured 4 weeks after the last dose. Feeling less physically attractive was reported by 27.3% of cooled patients against 56.3% of controls3. Further cohorts found scalp cooling effective with anthracycline and taxane regimens and faster hair recovery when it failed4,5. A systematic review found no increase in scalp metastases with cooling6, and a practice review addressed its logistics and insurance7.

Peripheral neuropathy is the toxicity most likely to persist. In a meta-analysis of 31 studies and 4,179 patients across tumour types, prevalence was 68.1% within the first month of chemotherapy, 60.0% at 3 months, and 30.0% at 6 months or beyond8. No agent is recommended for prevention, and duloxetine is the only agent with reasonable evidence for treating established painful neuropathy9. Physical methods are the exception. Frozen gloves and socks reduced objective and subjective neuropathy from paclitaxel in a self-controlled trial10, and a review of cooling found benefit from greater degrees of cooling11. In the POLAR randomised trial, hand cooling and compression significantly reduced high-grade taxane-induced neuropathy12. Dose reduction remains the intervention that actually works, and it has to be made before the deficit becomes fixed.

Fluoropyrimidine toxicity has a pharmacogenetic prevention. Genotype-guided dose reduction in DPYD variant carriers before treatment reduced severe toxicity in a prospective safety analysis13. Testing happens before the first dose or it is pointless.

2 · Endocrine therapy symptom burden

Endocrine therapy is the best tolerated effective systemic treatment in this disease by any objective grading scale, and it is the treatment patients most often stop. Those two statements are both true and the gap between them is the subject of this section.

The symptom set is predictable. Vasomotor symptoms, arthralgia and the aromatase inhibitor musculoskeletal syndrome, vaginal dryness and dyspareunia, reduced libido, sleep disturbance, and mood change. Tamoxifen adds thromboembolic and endometrial risk. Aromatase inhibitors add accelerated bone loss, which is Bone, metabolic, and endocrine late effects.

The consequence is measured in discontinuation. In a cohort of 8,769 women with early-stage disease, 32% had discontinued adjuvant hormonal therapy by 4.5 years, and of those still taking it only 72% were fully adherent. Overall, 49% completed the full duration at the intended schedule. Women under 40 years had the highest risk of discontinuation, with a hazard ratio of 1.51 and a 95% confidence interval of 1.23 to 1.8514.

Caution

Grade 1 and grade 2 are the categories in which this whole problem lives. A symptom that never reaches grade 3 is invisible to a toxicity table and is the commonest reason a curative treatment is abandoned. Reading tolerability off a trial safety table understates the burden of endocrine therapy more than that of any other class in this book.

For vasomotor symptoms, venlafaxine reduced hot flushes in survivors in a randomised trial, with side effects to set against the benefit15. Clonidine was slower than venlafaxine but gave lower scores by 12 weeks16. An evidence review rated venlafaxine and gabapentin likely to be effective and noted a large placebo effect17. A randomised crossover trial in breast cancer survivors compared venlafaxine and gabapentin. Both reduced hot flash scores by a similar amount. Of 56 patients expressing a preference, 68% preferred venlafaxine and 32% gabapentin18. Selective serotonin reuptake inhibitors that strongly inhibit CYP2D6 are conventionally avoided alongside tamoxifen on pharmacological grounds. The genotype evidence behind that caution is weaker than the caution implies, since CYP2D6 genotype was not associated with recurrence in a tamoxifen-treated cohort19. The drug interaction was measured before the genotype question was settled. Coadministered CYP2D6 inhibitors lowered plasma endoxifen20, and the interactions between antidepressants and tamoxifen were reviewed for psychiatrists21. A population study found no increase in recurrence with concurrent CYP2D6-inhibiting medications22.

Switching within or between classes is a legitimate intervention rather than an admission of failure. A patient intolerant of one aromatase inhibitor may tolerate another, and the alternative to a switch is frequently no treatment at all. Adherence and its determinants are developed in Adherence, symptom burden, and the gap between prescribed and received.

3 · Targeted agent toxicity by class

Targeted agent toxicity is mostly on-target toxicity in normal tissue. This is worth stating plainly because it predicts which toxicities are unavoidable and which are idiosyncratic.

Cell cycle inhibitors suppress proliferating marrow. Neutropenia is a class effect and its severity differs by agent, and abemaciclib produces diarrhoea as its dominant toxicity while the other two agents in the class produce more myelosuppression. Ribociclib prolongs the QT interval and requires electrocardiographic and electrolyte monitoring. Interstitial lung disease is rare across the class and is a recognised event.

PI3K and AKT inhibition disrupts insulin signalling in normal tissue, so hyperglycaemia is the mechanism working as designed outside the tumour. In SOLAR-1, grade 3 or 4 hyperglycaemia occurred in 36.6% of patients receiving alpelisib with fulvestrant against 0.7% receiving placebo with fulvestrant, and grade 3 or 4 rash in 9.9% against 0.3%. Discontinuation for adverse events occurred in 25.0% against 4.2%23. In CAPItello-291, grade 3 or higher rash occurred in 12.1% of patients receiving capivasertib with fulvestrant against 0.3% with placebo. Grade 3 or higher diarrhoea occurred in 9.3% against 0.3%. Discontinuation for adverse events occurred in 13.0% against 2.3%24.

mTOR inhibition produces stomatitis and non-infectious pneumonitis, and the stomatitis is an aphthous ulceration distinct from cytotoxic mucositis25. PARP inhibition produces anaemia, fatigue, and nausea26,27, and carries a small long-term risk of myeloid neoplasm taken up in Treatment-related second malignancies and clonal hematopoiesis. HER2-directed tyrosine kinase inhibitors produce diarrhoea and hand-foot reactions, and tucatinib adds transaminase elevation28.

The combination problem is that these toxicities overlap. Two agents that each suppress the marrow, or each cause diarrhoea, do not simply add their published rates. Overlapping toxicity as a constraint on combination design is in Combination strategies and overlapping toxicity, and dose optimisation as a regulatory question is in Dose optimization and the move away from maximum tolerated dose.

4 · Conjugate-specific toxicity and monitoring

The mechanism of conjugate toxicity belongs to Toxicity mechanisms, interstitial lung disease, ocular, and mucosal. This section is the monitoring algorithm, and the two should not be confused. Understanding why a payload reaches lung epithelium does not tell a clinician when to order a scan.

Interstitial lung disease is the event that changes management. A pooled analysis of nine trastuzumab deruxtecan monotherapy studies covering 1,150 patients found adjudicated drug-related interstitial lung disease or pneumonitis in 15.4%, grade 5 in 2.2%. Among those with an event, 77.4% had only grade 1 or grade 2 disease and 87.0% had their first event within 12 months of the first dose. Median time to first event was 5.4 months. Independent adjudication placed onset earlier than the investigator had for 53.2% of events, with a median difference of 43 days29. In the randomised comparison against trastuzumab emtansine, the adjudicated rate was 10.5% against 1.9%30.

In practice

Screen by symptom at every visit. Ask specifically about new or changed cough, breathlessness, and exercise tolerance, and treat any positive answer as an indication for computed tomography.

Hold the drug while investigating. A held cycle costs little. A grade 3 event costs the drug permanently.

Do not wait for hypoxia or for a grade to declare itself. The adjudication gap says the radiological change precedes the clinical impression by about six weeks.

Treat the first 12 months as the high-risk window, since 87.0% of patients who had an event had it inside that year. Vigilance should not decay at six months29.

For grade 1 events, hold and treat with corticosteroids, and resume only after full radiological resolution. For grade 2 or above, discontinue permanently.

For datopotamab deruxtecan, stomatitis and ocular surface events are managed prophylactically rather than reactively, with mouth care and ocular lubrication started at the first cycle31.

For sacituzumab govitecan, expect neutropenia and diarrhoea rather than pneumonitis. Grade 3 or higher neutropenia occurred in 51% of patients against 33% on single-agent chemotherapy, and grade 3 or higher diarrhoea in 10% against under 1%32.

For trastuzumab emtansine, monitor platelets and transaminases, which are its characteristic toxicities rather than the lung33.

Interplay

The monitoring schedules above are not interchangeable between conjugates, and the reason is in Part VIII. Toxicity tracks the payload and linker rather than the antigen, so two agents against the same target need different surveillance and two agents against different targets may need the same. A clinician who groups these drugs by their antibody will monitor the wrong organ.

See Antibody-drug conjugates and targeted delivery

5 · Immune-related adverse events and endocrinopathies

Immune-related toxicity does not resemble cytotoxic toxicity in timing, organ distribution, or reversibility, because it is the mechanism rather than a side effect of it. Management is codified in guideline form with organ-specific branches34, and the general structure is set out in Immune-related toxicity and its management.

Three properties matter more than any individual algorithm.

Onset is not dose-related and is frequently delayed. Events appear months into treatment and after it has stopped, so tolerance of six cycles does not establish low risk.

Reversibility differs by organ. Colitis, hepatitis, and pneumonitis usually resolve with immunosuppression. Endocrinopathy frequently does not. Hypophysitis, primary thyroid failure, and immune-mediated diabetes leave lifelong replacement whether or not the drug is stopped.

Grade does not mean the same thing in every organ. Grade 2 dermatitis and grade 2 pneumonitis are not comparable events, which is why the algorithms branch by organ rather than by number.

The consideration specific to breast cancer is the population. Checkpoint blockade is used in early triple-negative disease with curative intent, and in KEYNOTE-522 more than half of the patients in the control arm reached a pathological complete response without pembrolizumab35. Permanent adrenal insufficiency in a patient who would have been cured without the drug is a different calculation from the same event in metastatic disease. That argument is developed in Immune-related toxicity in a curative-intent population and belongs in the conversation held before the first dose.

Caution

Fatigue in a patient on checkpoint blockade is not a nonspecific complaint. Hypothyroidism, hypophysitis, and adrenal insufficiency all present this way, and adrenal insufficiency is the one that kills. Check thyroid function and a morning cortisol before attributing fatigue to treatment or to the cancer.

6 · Cardio-oncology, surveillance and prevention

Anthracycline injury and HER2-directed injury differ in reversibility, and that single difference drives everything about how they are watched.

Anthracycline cardiotoxicity is cumulative-dose related and is largely irreversible once established. Its timing is narrower than clinical folklore suggests. In a cohort of patients receiving anthracycline-containing therapy, cardiotoxicity occurred in 9%. The median interval from the end of chemotherapy to the event was 3.5 months. Of all events, 98% occurred within the first year. Among affected patients treated with heart failure therapy, 11% had full recovery and 71% had partial recovery, and recovery depended on how early treatment was started36.

Trastuzumab-associated dysfunction behaves differently. A referral series examined 38 patients with suspected trastuzumab-related cardiotoxicity, all previously anthracycline-exposed. Mean ejection fraction fell from 0.61 to 0.43 and recovered to 0.56 after withdrawal. Mean time to recovery was 1.5 months. Of 25 patients re-treated with trastuzumab, three had recurrent left ventricular dysfunction37.

The surveillance consequence follows directly. For anthracyclines the aim is early detection in the first year, because the window in which the injury is modifiable is short. For HER2-directed therapy the aim is serial monitoring across the treatment course, because a fall is expected to be recoverable and rechallenge is often possible. Treating both with one schedule wastes imaging in one setting and misses events in the other. The HER2-specific schedule is in Cardiac safety across the adjuvant trials.

Baseline risk stratification before potentially cardiotoxic treatment is now formalised, using pre-existing cardiovascular disease, prior cardiotoxic exposure, age, and conventional risk factors38.

Radiation is the third exposure and behaves as a dose-response rather than a threshold. In a case-control study of 2,168 women irradiated for breast cancer, comprising 963 with major coronary events and 1,205 controls, the overall average mean heart dose was 4.9 Gy. Rates of major coronary events rose by 7.4% per gray of mean heart dose, with no apparent threshold39. Cardiac sparing technique is in Cardiac and pulmonary sparing and late effects.

In practice

Measure the baseline before the first cardiotoxic dose, not after the first symptom.

For anthracyclines, concentrate surveillance in the first 12 months after the last cycle.

Start heart failure therapy at the point of detection rather than at the point of symptoms. Recovery depends on the interval.

Treat blood pressure, lipids, glucose, and smoking. Conventional risk factor management is the only part of this that is fully modifiable.

7 · Pulmonary, neurologic, and hematologic complications

The pulmonary problem in current practice is that four different causes present identically. Conjugate-associated interstitial lung disease, checkpoint pneumonitis, radiation pneumonitis, and mTOR inhibitor pneumonitis all appear as new dyspnoea with ground-glass change. Infection and lymphangitic progression complete the differential. The distinction matters because the treatments diverge. Conjugate events require permanent discontinuation at grade 2 and above29. Checkpoint events require immunosuppression with a specified taper34. Radiation pneumonitis is confined to the treated volume and follows a defined time course. Infection requires the opposite of steroids. A patient on several of these agents at once presents the hardest version of the problem, and no test resolves it quickly.

Neurologic complications are dominated by peripheral neuropathy, covered in Acute and chronic cytotoxic toxicity, and by cognitive change, covered in Fatigue, cognition, and return to function. Both are chronic rather than acute, and both are underdocumented because neither generates an abnormal laboratory value.

Haematologic complications extend beyond the nadir. Prolonged cytopenias after multiple lines of treatment reflect cumulative marrow damage rather than a single regimen. PARP inhibitors produce anaemia that is dose-limiting and often requires transfusion26. Thrombosis risk is elevated by the cancer itself, by tamoxifen, by central venous access, and by immobility, and these are additive in a way that no single risk score captures well.

8 · Bone, metabolic, and endocrine late effects

Aromatase inhibition and ovarian suppression both accelerate bone loss, and they do so on top of whatever age-related loss is already occurring. Baseline bone density measurement before starting, with a defined monitoring interval, is the standard of care rather than a refinement.

Two separate indications for antiresorptive therapy exist in this disease and conflating them is common. Every randomised trial of a bone-modifying agent in early disease is in the table below, and the endpoint column is the one that separates the two indications.

Bone-modifying agents given alongside adjuvant treatment.

One of these trials was designed around a fracture. The others were designed around a recurrence. Read the endpoint column before the result column.

TrialPopulationExperimental vs controlPrimary endpointResultOverall survival
D-CARE403 · 4509stage II or III breast cancer at high risk, receiving neoadjuvant or adjuvant chemotherapy, at 389 centres in 39 countriesdenosumab 120 mg subcutaneously every 3 to 4 weeks with chemotherapy then every 12 weeks to five years, vs placebobone metastasis-free survivalbone metastasis-free survival was no different, median not reached in either arm, hazard ratio 0.97 (0.82-1.14), p=0.70not reported in the primary publication
AZURE413 · 3360stage II or III breast cancer receiving standard adjuvant systemic therapyzoledronic acid 4 mg intravenously added to standard therapy, six doses then at reducing frequency to five years, vs standard therapy alonedisease-free survivalat a median 117 months, disease-free survival was no different overall, hazard ratio 0.94 (0.84-1.06), p=0.34, nor invasive disease-free survival, 0.91 (0.82-1.02), p=0.116. In postmenopausal women the hazard ratios were 0.82 (0.67-1.00) and 0.78 (0.64-0.94). Bone metastasis as the first recurrence fell, 0.76 (0.63-0.92), p=0.005not reported for the whole population in this report. In the 79% of tested tumours without MAF amplification the hazard ratio was 0.69 (0.50-0.94)
ABCSG-18423 · 3420HR+ postmenopausal on aromatase inhibitordenosumab 60 mg subcutaneously every 6 months vs placebo, added to adjuvant aromatase inhibitor therapytime from randomisation to first clinical fracturetime to first clinical fracture favoured denosumab, hazard ratio 0.50 (0.39-0.65), p<0.0001, with 92 fractures against 176. The effect held both in patients with a baseline bone mineral density T-score of -1 or higher, 0.44 (0.31-0.64), and in those already below it, 0.57 (0.40-0.82). No case of osteonecrosis of the jaw was adjudicatednot an endpoint of this fracture-prevention trial; 93 patients (3%) died during the study period
ABCSG-12433 · 1803premenopausal endocrine-responsive early breast cancer, all receiving goserelin, median follow-up 47.8 monthsgoserelin plus anastrozole or tamoxifen, each with or without zoledronic acid, for 3 yearsdisease-free survivalanastrozole did not improve disease-free survival over tamoxifen, HR 1.10 (0.78-1.53), p=0.59; adding zoledronic acid reduced disease progression, HR 0.64 (0.46-0.91), p=0.01zoledronic acid did not significantly reduce death, HR 0.60 (0.32-1.11), p=0.11

The first indication is skeletal protection, and ABCSG-18 is the trial that establishes it42. Two features of it are worth holding. Its endpoint was a fracture rather than a recurrence. And the benefit held in women whose baseline bone density was still in the normal range, which makes this a treatment decision rather than an osteoporosis decision.

The final analysis of ABCSG-18, at a median follow-up of eight years, added a disease outcome44. Disease-free survival events numbered 309 with denosumab and 368 with placebo, a hazard ratio of 0.83 with a 95% confidence interval of 0.71 to 0.97. Bone metastasis-free survival and overall survival pointed the same way, at 0.81 and 0.80, with intervals reaching 1.00 and 1.01. These were secondary endpoints in a trial powered for fractures, which limits how far they carry.

The second indication is antitumour effect, and it belongs to bisphosphonates rather than to bone protection in general. A meta-analysis of individual patient data from randomised trials examined adjuvant bisphosphonates. In women who were postmenopausal when treatment began, recurrence fell with a rate ratio of 0.86 and a 95% confidence interval of 0.78 to 0.94. Distant recurrence fell with a ratio of 0.82 and bone recurrence with a ratio of 0.72. Breast cancer mortality fell with a ratio of 0.82 and an interval of 0.73 to 0.93. Fractures were reduced with a ratio of 0.85. Benefit was definite only in women postmenopausal at the start of treatment45.

The two individual trials in that table are where the distinction is easiest to see. AZURE gave adjuvant zoledronic acid and separated nothing across the whole population41. Its first report had crossed a boundary for lack of benefit, with disease-free survival of 77% in each group46. ABCSG-12 added zoledronic acid to ovarian suppression with anastrozole or tamoxifen in 1,803 premenopausal women, and its benefit persisted at 62 months47. A trial-level meta-analysis found fewer fractures and a survival benefit over placebo or no treatment, with osteonecrosis of the jaw in 0.52%48. What benefit it found sat in the postmenopausal stratum, which is the meta-analysis result appearing inside a single randomisation. D-CARE gave denosumab at the dose and interval used for skeletal protection in metastatic disease, and it did not improve bone metastasis-free survival40. The drug that prevents fractures did not prevent metastases, at a far higher exposure than the fracture indication uses. The ABCSG-18 secondary endpoints, at the lower dose, pointed the other way44.

Caution

These are different questions with different answers. A drug that prevents fractures has not thereby been shown to prevent recurrence, and a drug that reduces recurrence in postmenopausal women has not been shown to do so in premenopausal women. The menopausal status restriction in the bisphosphonate data is the sharpest illustration, because the same drug in the same disease produced benefit in one group and not the other.

Premature menopause from chemotherapy or ovarian suppression compresses years of endocrine change into months, and its bone and cardiovascular consequences arrive early in a young patient Parenting, relationships, and long-horizon survivorship. Metabolic late effects include weight gain during adjuvant treatment, corticosteroid exposure, and the hyperglycaemia of PI3K and AKT inhibition described in Targeted agent toxicity by class. Thyroid dysfunction follows supraclavicular irradiation and checkpoint blockade through entirely different mechanisms, and both need a check rather than an assumption.

9 · Treatment-related second malignancies and clonal hematopoiesis

The best current quantification comes from a population cohort of 476,373 women with breast cancer as their first invasive cancer, followed for up to 28 years. By 20 years, 13.6% had developed a non-breast primary cancer, which was 2.1 percentage points more than expected in the general population. A contralateral breast cancer had developed in 5.6%, which was 3.1 percentage points more than expected. Radiotherapy was associated with increased contralateral breast and lung cancer, endocrine therapy with increased uterine cancer and reduced contralateral breast cancer, and chemotherapy with increased acute leukaemia. Absolute excess risk at 20 years was under 1% for every individual non-breast cancer type. Roughly 7% of the excess second cancers in the cohort were estimated to be attributable to adjuvant therapies49.

Caution

Standardised incidence ratios above 1.5 and absolute excesses below 1% are both correct descriptions of the same data. The ratio is the right number for understanding causation. The absolute excess is the only number that belongs in a treatment conversation. Quoting the ratio to a patient deciding about adjuvant chemotherapy misrepresents the magnitude by an order of magnitude or more.

Therapy-related myeloid neoplasms arise by two recognised routes. Alkylating agents produce a latency of several years with characteristic chromosomal losses, and topoisomerase II inhibitors produce a shorter latency with balanced translocations. Both are rare after standard breast cancer regimens.

Clonal haematopoiesis is the mechanistic bridge and it is now measurable. Somatic mutations in blood cells without cytopenia increase steeply with age and are associated with adverse outcomes including subsequent haematologic cancer50. In patients with non-haematologic cancers, therapy-related clonal haematopoiesis is common and associated with adverse clinical outcomes51. Cancer therapy is not a neutral observer of this process. It shapes which clones expand, with specific therapy classes selecting for mutations in defined DNA damage response genes52. The ageing marrow context sits in Aging, immunosenescence, and clonal hematopoiesis.

There is a measurement consequence that reaches beyond haematology. Clonal haematopoiesis is a leading source of false positive results in plasma sequencing, because the variant is genuinely in the blood and is not from the tumour. That problem is developed in Clonal hematopoiesis and other false positives.

10 · Toxicity monitoring and coordinated care

Most monitoring failures are not knowledge failures. They are ownership failures. Everyone assumed someone else was checking the echocardiogram, the thyroid function, the bone density, or the glucose.

The remedy is structural and unglamorous. Every regimen should have a written monitoring plan that names the test, the interval, and the person responsible. That is three fields, and their absence accounts for most of the missed surveillance in Cardio-oncology, surveillance and prevention and Bone, metabolic, and endocrine late effects.

Three transitions are where plans break. The first is from intravenous to oral therapy, because the patient stops attending an infusion unit where monitoring happened by default. The second is from the cancer centre to community or primary care, which is the subject of Survivorship care models and transition to primary care. The third is between lines of therapy, because the monitoring requirements change completely and the previous plan is often carried forward.

There is a measurement asymmetry worth naming. Clinician-graded adverse events record the clinician's impression of the patient's experience. When the same events are graded by patients directly, the reported severity and frequency are systematically higher. Both are measurements, and they are measurements of different things. Routine collection of patient-reported outcomes is developed in Patient-reported outcomes as routine clinical data.

In practice

Write the monitoring plan when you write the prescription, not at the first review.

Name one person per recurring test. Shared responsibility for a six-monthly echocardiogram means no responsibility.

Reconcile the plan at every line change and at every handover to another service.

Ask the patient directly about the symptoms that do not generate laboratory abnormalities. Neuropathy, arthralgia, fatigue, sexual function, and cognitive change will not otherwise be recorded.

Toxicity in older adults carries additional predictors that this chapter does not repeat, and they are in Treatment adaptation and tolerability.

References

All 52 references
  1. Hesketh PJ, Kris MG, Basch E, et al. Antiemetics: ASCO Guideline Update. J Clin Oncol 2020 38:2782-2797. PMID 32658626
  2. Nangia J, Wang T, Osborne C, et al. Effect of a scalp cooling device on alopecia in women undergoing chemotherapy for breast cancer: the SCALP randomized clinical trial. JAMA 2017 317:596-605. PMID 28196254
  3. Rugo HS, Klein P, Melin SA, et al. Association between use of a scalp cooling device and alopecia after chemotherapy for breast cancer. JAMA 2017 317:606-614. PMID 28196257
  4. Vasconcelos I, Wiesske A, Schoenegg W. Scalp cooling successfully prevents alopecia in breast cancer patients undergoing anthracycline/taxane-based chemotherapy. Breast 2018 40:1-3. PMID 29660543
  5. Kinoshita T, Nakayama T, Fukuma E, et al. Efficacy of Scalp Cooling in Preventing and Recovering From Chemotherapy-Induced Alopecia in Breast Cancer Patients: The HOPE Study. Front Oncol 2019 9:733. PMID 31448235
  6. Rugo HS, Melin SA, Voigt J. Scalp cooling with adjuvant/neoadjuvant chemotherapy for breast cancer and the risk of scalp metastases: systematic review and meta-analysis. Breast Cancer Res Treat 2017 163:199-205. PMID 28275922
  7. Kruse M, Abraham J. Management of Chemotherapy-Induced Alopecia With Scalp Cooling. J Oncol Pract 2018 14:149-154. PMID 29529389
  8. Seretny M, Currie GL, Sena ES, et al. Incidence, prevalence, and predictors of chemotherapy-induced peripheral neuropathy: A systematic review and meta-analysis. Pain 2014 155:2461-2470. PMID 25261162
  9. Loprinzi CL, Lacchetti C, Bleeker J, et al. Prevention and management of chemotherapy-induced peripheral neuropathy in survivors of adult cancers: ASCO guideline update. J Clin Oncol 2020 38:3325-3348. PMID 32663120
  10. Hanai A, Ishiguro H, Sozu T, et al. Effects of Cryotherapy on Objective and Subjective Symptoms of Paclitaxel-Induced Neuropathy: Prospective Self-Controlled Trial. J Natl Cancer Inst 2018 110:141-148. PMID 29924336
  11. Sphar BG, Bowe C, Dains JE. The Impact of Peripheral Cooling on Chemotherapy-Induced Peripheral Neuropathy: An Integrative Review. J Adv Pract Oncol 2020 11:845-857. PMID 33489425
  12. Michel LL, Schwarz D, Romar P, et al. Efficacy of Hand Cooling and Compression in Preventing Taxane-Induced Neuropathy: The POLAR Randomized Clinical Trial. JAMA Oncol 2025 11:408-415. PMID 40048176
  13. Henricks LM, Lunenburg CATC, de Man FM, et al. DPYD genotype-guided dose individualisation of fluoropyrimidine therapy in patients with cancer: a prospective safety analysis. Lancet Oncol 2018 19:1459-1467. PMID 30348537
  14. Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 early-stage breast cancer patients. J Clin Oncol 2010 28:4120-4128. PMID 20585090
  15. Loprinzi CL, Kugler JW, Sloan JA, et al. Venlafaxine in management of hot flashes in survivors of breast cancer: a randomised controlled trial. Lancet 2000 356:2059-63. PMID 11145492
  16. Boekhout AH, Vincent AD, Dalesio OB, et al. Management of hot flashes in patients who have breast cancer with venlafaxine and clonidine: a randomized, double-blind, placebo-controlled trial. J Clin Oncol 2011 29:3862-8. PMID 21911720
  17. Kaplan M, Mahon S, Cope D, Keating E, Hill S, Jacobson M. Putting evidence into practice: evidence-based interventions for hot flashes resulting from cancer therapies. Clin J Oncol Nurs 2011 15:149-57. PMID 21444282
  18. Bordeleau L, Pritchard KI, Loprinzi CL, et al. Multicenter, randomized, cross-over clinical trial of venlafaxine versus gabapentin for the management of hot flashes in breast cancer survivors. J Clin Oncol 2010 28:5147-52. PMID 21060031
  19. Rae JM, Drury S, Hayes DF, et al. CYP2D6 and UGT2B7 genotype and risk of recurrence in tamoxifen-treated breast cancer patients. J Natl Cancer Inst 2012 104:452-460. PMID 22395643
  20. Jin Y, Desta Z, Stearns V, et al. CYP2D6 genotype, antidepressant use, and tamoxifen metabolism during adjuvant breast cancer treatment. J Natl Cancer Inst 2005 97:30-9. PMID 15632378
  21. Henry NL, Stearns V, Flockhart DA, Hayes DF, Riba M. Drug interactions and pharmacogenomics in the treatment of breast cancer and depression. Am J Psychiatry 2008 165:1251-5. PMID 18829880
  22. Ahern TP, Pedersen L, Cronin-Fenton DP, Sørensen HT, Lash TL. No increase in breast cancer recurrence with concurrent use of tamoxifen and some CYP2D6-inhibiting medications. Cancer Epidemiol Biomarkers Prev 2009 18:2562-4. PMID 19690182
  23. Andre F, Ciruelos E, Rubovszky G, et al. Alpelisib for PIK3CA-mutated, hormone receptor-positive advanced breast cancer. N Engl J Med 2019 380:1929-1940. PMID 31091374
  24. Turner NC, Oliveira M, Howell SJ, et al. Capivasertib in hormone receptor-positive advanced breast cancer. N Engl J Med 2023 388:2058-2070. PMID 37256976
  25. Baselga J, Campone M, Piccart M, et al. Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer. N Engl J Med 2012 366:520-529. PMID 22149876
  26. Robson M, Im SA, Senkus E, et al. Olaparib for metastatic breast cancer in patients with a germline BRCA mutation. N Engl J Med 2017 377:523-533. PMID 28578601
  27. Litton JK, Rugo HS, Ettl J, et al. Talazoparib in patients with advanced breast cancer and a germline BRCA mutation. N Engl J Med 2018 379:753-763. PMID 30110579
  28. Murthy RK, Loi S, Okines A, et al. Tucatinib, trastuzumab, and capecitabine for HER2-positive metastatic breast cancer. N Engl J Med 2020 382:597-609. PMID 31825569
  29. Powell CA, Modi S, Iwata H, et al. Pooled analysis of drug-related interstitial lung disease and/or pneumonitis in nine trastuzumab deruxtecan monotherapy studies. ESMO Open 2022 7:100554. PMID 35963179
  30. Cortes J, Kim SB, Chung WP, et al. Trastuzumab deruxtecan versus trastuzumab emtansine for breast cancer. N Engl J Med 2022 386:1143-1154. PMID 35320644
  31. Heist RS, Sands J, Bardia A, et al. Clinical management, monitoring, and prophylaxis of adverse events of special interest associated with datopotamab deruxtecan. Cancer Treat Rev 2024 125:102720. PMID 38502995
  32. Bardia A, Hurvitz SA, Tolaney SM, et al. Sacituzumab govitecan in metastatic triple-negative breast cancer. N Engl J Med 2021 384:1529-1541. PMID 33882206
  33. Verma S, Miles D, Gianni L, et al. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med 2012 367:1783-1791. PMID 23020162
  34. Schneider BJ, Naidoo J, Santomasso BD, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO guideline update. J Clin Oncol 2021 39:4073-4126. PMID 34724392
  35. Schmid P, Cortes J, Pusztai L, et al. Pembrolizumab for early triple-negative breast cancer. N Engl J Med 2020 382:810-821. PMID 32101663
  36. Cardinale D, Colombo A, Bacchiani G, et al. Early detection of anthracycline cardiotoxicity and improvement with heart failure therapy. Circulation 2015 131:1981-1988. PMID 25948538
  37. Ewer MS, Vooletich MT, Durand JB, et al. Reversibility of trastuzumab-related cardiotoxicity: new insights based on clinical course and response to medical treatment. J Clin Oncol 2005 23:7820-6. PMID 16258084
  38. Lyon AR, López-Fernández T, Couch LS, et al. 2022 ESC guidelines on cardio-oncology developed in collaboration with the European Hematology Association, the European Society for Therapeutic Radiology and Oncology and the International Cardio-Oncology Society. Eur Heart J 2022 23:e333-e465. PMID 36017568
  39. Darby SC, Ewertz M, McGale P, et al. Risk of ischemic heart disease in women after radiotherapy for breast cancer. N Engl J Med 2013 368:987-998. PMID 23484825
  40. Coleman R, Finkelstein DM, Barrios C, et al. Adjuvant denosumab in early breast cancer (D-CARE): an international, multicentre, randomised, controlled, phase 3 trial. Lancet Oncol 2019 21:60-72. PMID 31806543
  41. Coleman RE, Collinson M, Gregory W, et al. Benefits and risks of adjuvant treatment with zoledronic acid in stage II/III breast cancer: 10 years follow-up of the AZURE randomized clinical trial (BIG 01/04). J Bone Oncol 2018 13:123-135. PMID 30591866
  42. Gnant M, Pfeiler G, Dubsky PC, et al. Adjuvant denosumab in breast cancer (ABCSG-18): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet 2015 386:433-43. PMID 26040499
  43. Gnant M, Mlineritsch B, Schippinger W, et al. Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N Engl J Med 2009 360:679-691. PMID 19213681
  44. Gnant M, Frantal S, Pfeiler G, et al. Long-Term Outcomes of Adjuvant Denosumab in Breast Cancer. NEJM Evid 2022 1:EVIDoa2200162. PMID 38319865
  45. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials. Lancet 2015 386:1353-1361. PMID 26211824
  46. Coleman RE, Marshall H, Cameron D, et al. Breast-cancer adjuvant therapy with zoledronic acid. N Engl J Med 2011 365:1396-405. PMID 21995387
  47. Gnant M, Mlineritsch B, Stoeger H, et al. Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62-month follow-up from the ABCSG-12 randomised trial. Lancet Oncol 2011 12:631-41. PMID 21641868
  48. Valachis A, Polyzos NP, Coleman RE, et al. Adjuvant therapy with zoledronic acid in patients with breast cancer: a systematic review and meta-analysis. Oncologist 2013 18:353-61. PMID 23404816
  49. McGale P, Dodwell D, Challenger A, et al. Second cancers in 475 000 women with early invasive breast cancer diagnosed in England during 1993-2016: population based observational cohort study. BMJ 2025 390:e083975. PMID 40865997
  50. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med 2014 371:2488-2498. PMID 25426837
  51. Coombs CC, Zehir A, Devlin SM, et al. Therapy-related clonal hematopoiesis in patients with non-hematologic cancers is common and associated with adverse clinical outcomes. Cell Stem Cell 2017 21:374-382. PMID 28803919
  52. Bolton KL, Ptashkin RN, Gao T, et al. Cancer therapy shapes the fitness landscape of clonal hematopoiesis. Nat Genet 2020 52:1219-1226. PMID 33106634