Radiotherapy for nonagenarians: the value of biological versus chronological age

All cancer patients aged ≥90 years who received radiotherapy between 2009 and 2019 at the University of Freiburg Medical Center were included in this analysis. Demographic characteristics, radiotherapy specifications and clinical data including toxicity reports were retrospectively surveyed using the electronic patient records. Comorbidities were quantified using the validated Adult Comorbidity Evaluation 27 (ACE-27) index, as it has been shown that comorbidity is an independent prognostic factor for cancer patients [21]. Tumor classification of most solid tumors was encoded according to the 8th edition of the UICC TNM classification, while brain tumors were encoded according to the WHO Classification of Tumors of the Central Nervous System. The Ann-Arbor classification was used for non-Hodgkin lymphoma, and myeloma was classified in accordance with the Salmon-Durie classification. The Independent Ethics Committee of the Medical Faculty, University of Freiburg approved this analysis (record no. 376/19).

Depending on the time period of treatment and tumor localization, patients were treated with intensity-modulated radiotherapy (IMRT), three-dimensional conformal radiotherapy (3D radiotherapy) or electron beam radiotherapy for superficial skin cancer lesions (Fig. 1). Oncentra MasterPlan® (Nucletron BV, Veenendaal, the Netherlands) and Eclipse™ planning systems (Varian Medical Systems) were used for radiotherapy planning. Depending on the target volume and primary malignancy, individual concepts regarding the gross tumor volume (GTV), clinical target volume (CTV) and planning target volume (PTV) were applied according to the institutional guidelines. Treatments using moderate hypofractionation were usually applied for adjuvant breast cancer radiotherapy (e.g. 40.05 Gy applied in 15 fractions), painful bone metastases (30 Gy or 36 Gy in 10 fractions or 8 Gy in 1 fraction) or highly palliative situations such as bleeding gynaecological tumors (30 Gy in 10 fractions or 20 Gy in 5 fractions) in our cohort. Normofractionated treatments were typically used for definitive treatments with curative intention, e.g. for non-melanoma skin cancers (60 Gy in 30 fractions). Few patients received stereotactic body radiotherapy (SBRT) for primary or secondary lung tumors; in these cases, 35 Gy in 5 fractions or 37.5 Gy in 3 fractions were used. Radiotherapy characteristics are summarized in Table 1.

For curative radiotherapy, follow-up consultations were performed regularly, whereby the intervals varied depending on the primary malignancy and patient wishes. Patients who were treated in a palliative intention commonly had at least one follow-up consultation at 6 weeks after completion of radiotherapy in order to evaluate the treatment response and acute toxicities. OS was defined as the time span from the start of radiotherapy to the last follow-up or death, whereby censoring was applied for patients who were lost to follow-up. Missing survival data were obtained by contacting the state authorities of Baden-Württemberg, Germany. Acute radiotherapy-related toxicities were retrospectively classified based on the Common Terminology Criteria for Adverse Events (CTCAE) v.5.0.

One hundred nineteen nonagenarian patients underwent radiotherapy between 2009 and 2019 and were included in our analysis. In line with an increased female-to-male-ratio in this elderly age group, 81 female patients (68.1%) and 39 male patients (31.9%) received radiotherapy (Table 2). The age in the study cohort ranged between 90 and 99 years with a median age of 91 years. The majority of patients demonstrated a reasonably good performance status with ECOG 1 (n = 58, 48.7%) being most prevalent. However, about one third of the study cohort was found to have a performance status ranging between ECOG 2 and ECOG 3. Existing comorbidities prior to treatment initiation were quantified according to the ACE-27 index and revealed that the large majority of the study population exhibited moderate (ACE-27-score of 2; n = 28, 23.5%) or severe (ACE-27-score of 3; n = 53, 44.5%) comorbidities. Skin cancers including basal cell carcinoma, squamous cell carcinoma and melanoma were the most common primary tumors in our study with 44 patients (37.0%) suffering from these malignancies, followed by breast cancer (n = 15, 12.6%) and other gynecological malignancies (n = 11, 9.2%) (detailed diagnoses are listed in Table 2). The majority of patients (n = 61, 51.3%) exhibited UICC stage IV tumors, followed by stage III (n = 20, 16.8%), stage II (n = 13, 10.9%) and stage I tumors (n = 10, 8.4%). Whereas 38 (31.9%) patients received curative radiotherapy, 81 (68.1%) were treated in a palliative setting. Fifty-four patients (45.4%) were treated on an outpatient basis, and 65 patients (54.6%) received their treatment as inpatients. On average, inpatients were hospitalized for a median of 16 days (range 2–43 days), and the majority of inpatients were discharged home (n = 49, 75.4%) or to nursing homes (n = 11, 16.9%) after radiotherapy.

Ninety-nine patients (83.2%) completed their prescribed course of radiotherapy, while 20 patients (16.8%) discontinued their treatment, most commonly due to a deterioration of their general condition. A split-course radiotherapy concept was applied in 14 patients (10.2%), and 3 patients (21.4%) of these did not receive the intended radiotherapy dose. Most patients were treated in the head-and-neck region (n = 49, 35.8%), followed by the pelvic (n = 28, 20.4%) and thoracic regions (n = 21, 15.3%). About one in four patients (n = 27, 22.7%) was treated with hormonal therapy for either breast or prostate cancer, while only 9 patients (7.6%) received concomitant chemotherapy. Treatment was administered to a median dose of 39.0 Gy (range 1.8 Gy – 72.0 Gy), and patients received a median of 14 fractions (range 1–33).

Median OS ranged at 10 months with a 1-year OS, 2-year OS and 3-year OS of 42.6, 21.6 and 11.1%, respectively (Fig. 2a). At 3 months after the start of radiotherapy, 76.8% of patients were still alive. Patients who were treated in a curative setting were shown to have superior OS rates compared to patients receiving palliative radiotherapy treatment (HR = 2.84, 95% CI 1.68–4.81, p < 0.001) (Fig. 2b, Table 3). The median OS for patients treated with curative intention was shown to be 24 months which was three times higher than the median OS of 8 months for patients receiving radiotherapy with palliative intention. Whereas survival was comparable between patients aged 90–94 years and patients ≥95 years (HR = 1.44, 95% CI 0.84–2.47, p = 0.19), higher values in the ACE-27 instrument as an indicator for a higher burden of comorbidities as well as a reduced performance status were found to be significantly associated with an impaired OS (ACE-score: HR = 2.00, 95% CI 1.00–4.10, p < 0.05; ECOG-status: HR = 1.56, 95% CI 1.00–2.45, p < 0.05) (Fig. 3a-b). Additionally, patients who were hospitalized during radiotherapy exhibited a lower OS than patients treated on an outpatient basis (HR = 1.64, 95% CI 1.05–2.66, p < 0.05) (Fig. 3c). Neither concomitant chemotherapy (HR = 1.02, 95% CI 0.32–3.26, p = 0.98) nor endocrine treatment (HR = 1.26, 95% CI 0.75–2.47, p = 0.19) influenced the survival rates after radiotherapy. Interestingly, skin cancer patients did not exhibit superior survival rates compared to non-skin cancer patients (HR = 0.87 in favor of non-skin cancer patients, 95% CI 0.56–1.36, p = 0.54) (Supplementary Figure 1A). Survival rates did significantly differ between the different UICC stages with the best OS rates for patients with UICC stage I (median OS of 43 months) and the worst OS rates for patients with UICC stage IV (median OS of 8 months) (UICC stage III-IV versus UICC stage I-II: HR = 2.21, 95% CI 1.14–4.26, p < 0.05) (Supplementary Figure 1B). Unplanned discontinuation (HR = 7.17, 95% CI 3.88–13.26, p < 0.001) due to worsening of a patient’s general condition as well as split-course concepts (HR = 2.05, 95% CI 1.07–3.94, p < 0.05) were observed to result in reduced survival rates (Fig. 4a-b).

In the multivariate analysis of significant pre-therapeutic parameters, split-course treatments (HR = 2.21, 95% CI 1.10–4.37, p < 0.05) and palliative treatments (HR = 3.19, 95% CI 1.77–5.75, p < 0.001) remained significant risk factors for reduced OS. In contrast, the ECOG performance status (HR = 1.49, 95% CI 0.91–2.43, p = 0.11), the comorbidity burden (HR = 2.37, 95% CI 0.91–6.18, p = 0.08) and the hospitalization status (HR = 1.18, 95% CI 0.71–1.94, p = 0.53) were found to be not significant regarding OS upon multivariate analysis.

In order to analyze whether the comorbidity burden or ECOG performance status differed between the split-course- and continuous course-group, Chi-square tests were conducted and showed no imbalances regarding these parameters between both groups (ACE-score 0 versus 1–3: p = 0.43, χ2-test; ECOG-score 0–1 versus 2–3: p = 0.78, χ2-test).

As the majority of patients received palliative treatments with reduced radiation doses, toxicity rates in our cohort were found to be relatively low. About one third of the study population (n = 40, 33.6%) did not report any radiotherapy-related acute toxicities, and 66 patients (55.5%) only exhibited CTCAE grade 1–2 adverse reactions (Table 4). Thirteen patients (10.9%) suffered from at least one grade 3 toxicity, most commonly dermatitis (n = 6), mucositis (n = 4) or pain (n = 3) (Supplementary Table 1). However, no grade 4–5 acute toxicities were observed in our study population. The median follow-up time and number of follow-up visits was insufficient to robustly assess chronic toxicities.