Part XIV · Special populations and uncommon presentations · Chapter 82

Older adults and complex comorbidity

Assessment is not a formality before the decision. It is part of the decision.

1 · Geriatric assessment in routine practice

Geriatric assessment is a structured evaluation of domains that an oncology history does not cover. Those domains are physical function, comorbidity burden, cognition, nutrition, psychological state, social support, and medications. Each is scored with a validated instrument rather than estimated from the doorway.

The argument for doing it is not that it is thorough. It is that it changes outcomes. In GAP70+, 40 community oncology practices were cluster-randomised. Oncologists in the intervention arm received a geriatric assessment summary with management recommendations, and those in the control arm received nothing. Enrolment covered 718 patients aged 70 years and over who were starting a new regimen for incurable cancer. Grade 3 to 5 toxic effects over 3 months occurred in 51% of the intervention group and 71% of the usual care group. The relative risk was 0.74, with a 95% confidence interval of 0.64 to 0.86. Falls occurred in 12% against 21%, and more medications were discontinued in the intervention arm¹.

Two limits should be stated with that result. The population had advanced incurable disease of mixed primary sites rather than breast cancer specifically. And the intervention was assessment plus a management plan, not assessment alone. Handing a clinician a score changes nothing. The trial tested a score attached to recommended actions.

The guideline position follows from this and has been restated across two editions^2,3. Assessment is recommended for patients aged 65 years and over receiving systemic therapy, with a brief screening instrument used to triage who needs the full evaluation.

2 · Frailty, function, cognition, and competing mortality

Frailty is reduced physiological reserve, which means a diminished capacity to recover from a stressor. Chemotherapy and surgery are stressors. The concept earns its place because it predicts how a patient will handle an insult rather than how much disease they have.

The prevalence figures are wide and the reason is measurement. A systematic review of 20 studies and 2,916 older patients with cancer reported a median frailty prevalence of 42%, with a range across studies of 6% to 86%. Pre-frailty had a median of 43% and a range of 13% to 79%⁴. That spread is not biological variation between cohorts. It is what happens when different instruments define the same word differently.

Function is the domain that carries the most predictive weight and the least measurement difficulty. Instrumental activities of daily living, gait speed, and a history of falls are recorded in minutes and outperform performance status, which was designed for a younger population and compresses the top of the range.

Cognition is assessed for two separate reasons. One is capacity to consent. The other is capacity to execute a regimen, which matters most for oral agents taken at home on a schedule with dose interruptions.

Competing mortality reframes the whole discussion. The relevant question is not what the breast cancer will do over 10 years. It is which of the patient's conditions will reach her first. A woman of 82 with node-negative hormone receptor positive disease, heart failure, and chronic kidney disease has a cancer that may never become the thing that limits her. Systemic therapy aimed at a 10-year endpoint is then being given for a benefit she has limited opportunity to collect. The biology of ageing that sits underneath this is developed in Aging and the changing normal microenvironment and Aging, immunosenescence, and clonal hematopoiesis.

3 · Treatment adaptation and tolerability

There are two ways to give less treatment and they are not equivalent. One is to select a population in whom a treatment can be omitted without loss. The other is to substitute a gentler version of the same treatment for everyone. The first is de-escalation. The second is dilution, and it has a clean negative result in this population.

CALGB 49907 randomised 633 women aged 65 years and over with early breast cancer to standard adjuvant chemotherapy or to single-agent capecitabine. At a median follow-up of 11.4 years, 10-year relapse-free survival was 56% with standard chemotherapy and 50% with capecitabine. All three hazard ratios below compare standard chemotherapy with capecitabine, so a value under 1 favours standard chemotherapy. For relapse-free survival it was 0.80. For breast cancer-specific survival, 88% against 82%, it was 0.62. For overall survival, 62% against 56%, it was 0.84, and that last comparison did not reach significance⁵. Substituting a better-tolerated drug cost disease control.

Selection works where substitution does not. In PRIME II, 1,326 women aged 65 years and over with hormone receptor positive, node-negative tumours of 3 cm or less were randomised after breast-conserving surgery to whole-breast irradiation or none, all receiving endocrine therapy. At a median follow-up of 9.1 years, cumulative local recurrence at 10 years was 9.5% without radiotherapy and 0.9% with it. Overall survival at 10 years was 80.8% and 80.7% respectively, and breast cancer-specific survival did not differ substantially⁶. The local control difference is real and large in relative terms. It did not translate into a survival difference because other causes of death dominated. That is de-escalation working through competing risk rather than through equivalence, and the distinction between those two claims is the subject of Designing de-escalation trials and the non-inferiority margin problem. Radiotherapy omission is developed further in Omission in low-risk and older patients.

Toxicity can be predicted rather than feared. The Cancer and Aging Research Group breast cancer score was developed and validated in 473 patients aged 65 and over receiving neoadjuvant or adjuvant chemotherapy, of whom 46% had grade 3 to 5 toxicity. Eight weighted predictors were retained: anthracycline use, stage II or III disease, planned treatment duration over 3 months, abnormal liver function, low haemoglobin, falls, limited walking, and lack of social support. In the development cohort, grade 3 to 5 toxicity occurred in 19%, 54%, and 87% of the low, intermediate, and high risk groups⁷.

Three of those eight predictors are falls, walking limitation, and social support. None appears on a standard oncology intake. The tool works partly because it asks questions oncologists do not otherwise ask.

Dosing is a further adaptation with a measurement problem underneath it. Body surface area assumes that a patient's drug-metabolising mass scales with height and weight. In sarcopenia it does not. Among patients with metastatic breast cancer receiving capecitabine, about 25% were sarcopenic on computed tomography, and sarcopenia occurred at normal weight, overweight, and obese body mass index. Toxicity after the first cycle occurred in 50% of sarcopenic patients against 20% of non-sarcopenic patients⁸. The same milligram dose is a different exposure.

4 · Cardiovascular, renal, and hepatic comorbidity

Comorbidity changes treatment through three separate routes. It alters drug handling, it alters the tolerance of an organ for injury, and it competes as a cause of death. Those should be assessed separately because they have different remedies.

Cardiac reserve constrains anthracyclines and HER2-directed antibodies. Baseline risk stratification before potentially cardiotoxic therapy is now formalised, with pre-existing cardiovascular disease, prior anthracycline exposure, age, hypertension, and diabetes as the contributing variables⁹. A reduced baseline ejection fraction is not an absolute contraindication to every agent. It is a reason to choose a non-anthracycline regimen, to co-manage with cardiology, and to shorten the surveillance interval. The full surveillance argument is in Cardio-oncology, surveillance and prevention.

Renal function governs capecitabine dosing, bisphosphonate and denosumab selection, and contrast imaging. The measurement point is specific and often missed. Creatinine-based estimating equations assume a normal relationship between muscle mass and creatinine production. In a sarcopenic patient that assumption fails, and the estimated glomerular filtration rate overstates true renal function. The patient looks safer on paper than she is.

Hepatic impairment affects taxanes, several tyrosine kinase inhibitors, and the conjugates. Baseline abnormal liver function is one of the eight predictors in the toxicity score above⁷.

5 · Polypharmacy and drug interactions

Polypharmacy is usually defined as five or more regular medications, and that threshold is a convention rather than a discovered boundary. In a systematic review of 47 studies, 57% used the five-medication definition. Polypharmacy was associated with chemotherapy toxicity in four of nine studies that examined it and with falls in three of three¹⁰. In a prospective cohort of 500 adults aged 65 and over starting chemotherapy, the mean number of daily medications was five, with a range from none to 23¹¹.

Three interaction classes recur in breast oncology and each has a specific remedy.

The first is cytochrome P450 3A4. CDK4/6 inhibitors, everolimus, and several tyrosine kinase inhibitors are substrates, and strong inducers or inhibitors shift exposure substantially. The common offenders in an older patient's list are azole antifungals, macrolides, and anticonvulsants.

The second is additive QT prolongation. Ribociclib prolongs the QT interval, and so do several antiemetics, antidepressants, and antiarrhythmics that the patient is already taking. The interaction is pharmacodynamic, so a drug-level check does not detect it.

The third is absorption. Acid-suppressing drugs reduce the absorption of several oral oncology agents, and long-term proton pump inhibitor use is common and frequently unexamined.

Two pharmacogenetic situations are worth separating because the evidence differs sharply. Genotype-guided dose reduction for DPYD variant carriers before fluoropyrimidine therapy reduced severe toxicity in a prospective safety analysis, and the variant carrier is identified before the first dose rather than after the event¹². CYP2D6 genotyping before tamoxifen has a coherent mechanism and no established clinical utility. In an analysis of tamoxifen-treated patients, CYP2D6 genotype was not associated with recurrence¹³. The pharmacological argument for avoiding strong CYP2D6 inhibitors alongside tamoxifen remains reasonable. The case for genotyping does not follow from it.

6 · Treating outside trial eligibility

Older adults are underrepresented in the trials that define their treatment, and the mechanism has shifted rather than disappeared. Explicit upper age limits are now rare. Eligibility criteria based on organ function, performance status, prior therapy, and life expectancy exclude the same people without naming them. The general form of this problem and its remedies sit in Representation, generalizability, and equity by design and Clinical trial representation and the mechanisms of exclusion.

The consequence is not that the evidence is unusable. It is that two different quantities travel differently out of the trial population.

Relative efficacy usually transfers. A drug that reduces the hazard of recurrence by a given proportion in a fit 60-year-old is not expected to lose that mechanism at 80. There is no strong biological reason for the ratio to change, and subgroup analyses by age rarely show one.

Absolute benefit does not transfer, and it shrinks for a reason that has nothing to do with the drug. Absolute benefit is the relative reduction applied to a baseline risk over a horizon the patient must survive to reach. Competing mortality truncates that horizon. The same hazard ratio buys fewer events prevented.

Toxicity transfers worst of all. Trial toxicity rates are generated in a population selected for organ function and independence. An 82-year-old with three comorbidities was not in the denominator. This is why the prediction tools in Treatment adaptation and tolerability exist, and why they were built in cohorts rather than derived from trial data.

7 · Caregiver support and treatment burden

Treatment burden is the work a regimen transfers to the patient and household. It includes travel to appointments, blood tests between cycles, managing an oral agent with dose interruptions, recognising toxicity, and absorbing the cost. That work is invisible in a protocol and it is the thing that most often determines whether treatment is completed.

The burden is heavier here for structural reasons. The caregiver of an older patient is frequently an older spouse with their own conditions, or an adult child with employment and dependent children. Transport, cognition, and the ability to attend at short notice are all constrained at once.

The oncology-facing consequence is concrete. Lack of social support is one of the eight predictors in a validated chemotherapy toxicity score in this population⁷. That is not a psychosocial observation appended to the medical one. It is a medical variable with a coefficient.

Two adjustments follow. The first is that regimen choice should account for visit frequency and monitoring intensity, not only for efficacy and toxicity. The second is that the caregiver needs a named point of contact, because in practice they are the person who calls.

The honest position on interventions is that the evidence base is thinner than the problem. The caregiver experience is developed in Family, partner, and caregiver experience, and the structural barriers of transport, cost, and geography in Transportation, employment, childcare, and digital barriers and Geographic access, rural care, and treatment deserts.

References

All 13 references

1. Mohile SG, Mohamed MR, Xu H, et al. Evaluation of geriatric assessment and management on the toxic effects of cancer treatment (GAP70+): a cluster-randomised study. Lancet 2021 398:1894-1904. PMID 34741815
2. Mohile SG, Dale W, Somerfield MR, et al. Practical Assessment and Management of Vulnerabilities in Older Patients Receiving Chemotherapy: ASCO Guideline for Geriatric Oncology. J Clin Oncol 2018 36:2326-2347. PMID 29782209
3. Dale W, Klepin HD, Williams GR, et al. Practical Assessment and Management of Vulnerabilities in Older Patients Receiving Systemic Cancer Therapy: ASCO Guideline Update. J Clin Oncol 2023 41:4293-4312. PMID 37459573
4. Handforth C, Clegg A, Young C, et al. The prevalence and outcomes of frailty in older cancer patients: a systematic review. Ann Oncol 2015 26:1091-1101. PMID 25403592
5. Muss HB, Polley MC, Berry DA, et al. Randomized Trial of Standard Adjuvant Chemotherapy Regimens Versus Capecitabine in Older Women With Early Breast Cancer: 10-Year Update of the CALGB 49907 Trial. J Clin Oncol 2019 37:2338-2348. PMID 31339827
6. Kunkler IH, Williams LJ, Jack WJL, Cameron DA, Dixon JM. Breast-conserving surgery with or without irradiation in early breast cancer. N Engl J Med 2023 388:585-594. PMID 36791159
7. Magnuson A, Sedrak MS, Gross CP, et al. Development and Validation of a Risk Tool for Predicting Severe Toxicity in Older Adults Receiving Chemotherapy for Early-Stage Breast Cancer. J Clin Oncol 2021 39:608-618. PMID 33444080
8. Prado CMM, Baracos VE, McCargar LJ, et al. Sarcopenia as a determinant of chemotherapy toxicity and time to tumor progression in metastatic breast cancer patients receiving capecitabine treatment. Clin Cancer Res 2009 15:2920-2926. PMID 19351764
9. 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
10. Mohamed MR, Ramsdale E, Loh KP, et al. Associations of Polypharmacy and Inappropriate Medications with Adverse Outcomes in Older Adults with Cancer: A Systematic Review and Meta-Analysis. Oncologist 2020 25:e94-e108. PMID 31570516
11. Maggiore RJ, Dale W, Gross CP, et al. Polypharmacy and potentially inappropriate medication use in older adults with cancer undergoing chemotherapy: effect on chemotherapy-related toxicity and hospitalization during treatment. J Am Geriatr Soc 2014 62:1505-12. PMID 25041361
12. 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
13. 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