PRO-CTCAE reveals under-recognition of dermatologic symptom burden in hospitalized cancer patients

Abstract

Purpose

Dermatologic adverse events (dAEs) occur frequently in hospitalized patients and can significantly reduce quality of life. Physicians grade dAEs using the Common Terminology Criteria of Adverse Events (CTCAE). However, they often underestimate symptom frequency and severity. The patient-reported outcomes (PRO) version of the CTCAE (PRO-CTCAE) was developed to assess symptoms from the patient’s perspective. In this study, we assessed the patient-reported burden of dAEs via the PRO-CTCAE questionnaire and compared results with dAE assessment by treating oncologists and dermatologists.

Methods

Patients admitted to Memorial Sloan Kettering Cancer Center from 6/1/2018 to 4/30/2019 and received a dermatology consultation were eligible. Once enrolled, participants completed a PRO-CTCAE questionnaire on 14 dermatologic symptoms. CTCAE grades assigned by oncology and dermatology were obtained from clinical notes, and kappa statistics were calculated to evaluate the level of agreement between physician and patient evaluations.

Results

A total of 100 patients (mean age 59.4, 55% male) were prospectively enrolled. The most common patient-reported dAEs were rash (72%), swelling (67%), pruritus (64%), bruising (53%), and hives (37%). Oncologists and dermatologists underreported dAEs except for rash (median kappa values 0.3 [0.02-0.84] and 0.32 [0.02-0.87], respectively). Oncologists and dermatologists were concordant with each other’s documented assessment of dAEs (median kappa value 0.985 [0.55-1]).

Conclusion

Oncology patient-reported dAEs in a tertiary academic oncologic referral center were under-recognized by providers. PRO-CTCAE may be a useful tool to optimize inpatient dermatologic care for cancer patients by detecting and allowing management of patient-reported dAEs.

Introduction

Dermatologic adverse events (dAE) are common complications of anticancer therapies and can significantly reduce patient quality of life (QoL) [1–3]. Most dAEs can be managed with proper interventions, but if severe, may lead to interruption of a patient’s oncologic regimen. The severity of dAEs is graded by providers using the Common Terminology Criteria of Adverse Events (CTCAE v5.0) [4]. However, physicians tend to underestimate dAE symptom frequency/severity, reporting these later than patients [5–7]. The patient-reported outcomes (PRO) version of the CTCAE (PRO-CTCAE) was developed by the National Cancer Institute (NCI) as a standardized tool to assess toxicities from the patient perspective [8, 9]. Integration of PRO-CTCAE into the evaluation of toxicities allows for earlier detection of symptoms and has been associated with improved survival in patients with cancer [10].

In this single-center study, we prospectively assessed the burden of patient-reported dAEs via the PRO-CTCAE questionnaire in hospitalized cancer patients who received a dermatology consultation between 6/2018 and 4/2019 and compared results with dAE assessment by treating oncologists (ONC) and dermatologists (DERM).

Methods

Patients admitted to Memorial Sloan Kettering Cancer Center who received a dermatology consultation during their hospitalization 6/1/2018—4/30/2019 were eligible for this study. Patients who agreed to enroll completed a dermatology-specific PRO-CTCAE questionnaire.

This questionnaire evaluated 14 symptoms with 17 items derived from the CTCAE criteria (swelling, rash, dry skin, acne, hair loss, itch, hives, hand-foot syndrome (HFS), changes on fingernails or toenails, bed sores, skin burns from radiation, skin darkening, bruising, and body odor) and was created via the web-based form builder [11]. CTCAE grades by ONC and DERM were extracted from the patient electronic medical record; dAEs reported by patients, but not documented by providers were marked as ‘not evaluated’ and excluded from analysis.

Patient characteristics were summarized using descriptive statistics and relative frequencies. Physician and patient assessments of the patient’s symptoms were cross classified to evaluate the level of agreement between PRO-CTCAE and physician assessments. Kappa and weight kappa statistics were calculated to evaluate level of agreement between physician and patient evaluations (RStudio v.4.2.2 and Stata/MP v16.1).

Results

One hundred patients (15.2% out of 658 eligible) with a mean age of 59.4 years (SD 14.8) were enrolled and completed the questionnaire. Most were male (55%) and White (76%), Black (10%), Asian (9%), or not specified (5%). Most patients had hematologic cancers, including acute leukemia (33%), lymphoma (17%), chronic leukemia (9%), myelodysplasia/myeloproliferative disorders (2%), and multiple myeloma (1%). The remaining had solid tumors: GI (7%), urogenital (6%), ENT (5%), sarcoma (4%), renal (3%), breast (3%), renal (3%), skin (2%), pancreas (2%), lung (2%), and brain (1%). Therapies administered to these patients included chemotherapy (45%), targeted therapy (22%), immunotherapy (12%), allogeneic transplant (15%), surgery (6%), radiation (4%), autologous transplant (3%), and endocrine therapy (1%). In addition, 8 patients received no therapy and 19 received multiple.

The following dAEs were reported: swelling (67, 31 severe), rash (72), xerosis (76, 25 severe), acne (13, 8 severe), hair loss (13, 7 severe), pruritus (64, 30 severe), hives (37), HFS (14), nail changes (11), bed sores (4), radiation burns (3), skin darkening (11), bruising (53), and body odor (8).

Kappa values between patient-oncologist, patient-dermatologist, and oncologist-dermatologist, respectively, are as follows: swelling (0.46, 0.34, 0.72), rash (0.84, 0.87, 0.97), xerosis (0.09, 0.1, 0.55), acne (0.39, 0.39, 1.0), pruritus (0.47, 0.52, 0.82), hives (0.09, 0.17, 0.79), HFS (0.21, 0.21, 1.0), bed sores (0.32, 0.32, 1.0), radiation burns (0.49, 0.49, 1.0), skin darkening (0.28, 0.4, 0.8), bruising (0.02, 0.02, 1.0), and body odor (0.1, 0.1, 1.0). The kappa between patient-dermatologist for nail changes was 0.15 and was not calculable between patient-oncologist or oncologist-dermatologist as no ONC reported this symptom. Patient and provider reporting of the 3 most common symptoms—swelling, xerosis, and pruritus—are depicted in Fig. 1.

Fig. 1.

A Flow diagram of reported dry skin by patient, oncologist, and dermatologist. B Flow diagram of reported pruritus by: patient, oncologist, and dermatologist. C Flow diagram of reported swelling by: patient, oncologist, and dermatologist

Discussion

In this study, we characterized the burden of dAEs using PRO and compared this with assessments from providers to evaluate patient and clinician concordance. Xerosis, rash, swelling, pruritus, and easy bruising were the most frequently reported all grade dAEs by patients, and swelling, pruritus, and xerosis were the most frequently reported high grade dAEs (grade ≥ 3).

All symptoms in the questionnaire except rash were underreported by both ONC and DERM, suggesting a discrepancy between the physician’s and the patient’s perception and/or documentation of the symptom burden. For most symptoms, DERM and ONC gradings were concordant (kappa > 0.8), which may be due to the standardization and wide interdisciplinary use of CTCAE grading in oncologic patients. Notably, xerosis was reported more among DERM and swelling more among ONC, possibly reflecting the concern for a systemic etiology of edema in these patients.

Rash was overreported by providers relative to PRO. In addition, there were cases where providers documented the presence of a dAE while patients did not (Fig. 1). We hypothesize this reduced patient-reporting occurred because the dAEs were present at baseline, overshadowed by other toxicities, or not considered significant by the patient.

DAEs can significantly reduce QoL, patient emotion, and patient function [12]. As such, prompt recognition and management are integral to patient care. Unfortunately, our findings reveal that providers tend to underestimate dermatologic toxicity severity among cancer patients [5, 7, 13]. The integration of PRO into the care of these patients may help to reduce this gap and improve patient satisfaction [14].

This study is limited by its single-center analysis and lack of patient/provider randomization. Only a small percentage of patients who were eligible enrolled, which may have led to selection bias favoring those with ability/interest to participate. CTCAE grading was based on chart documentation, which may under-report physician recognition of symptoms; however, without documentation of management, it is likely that even if these symptoms were recognized, they were under-addressed.

Conclusion

DAEs occur frequently among hospitalized oncology patients but may be underreported or under-addressed by ONC and DERM providers. PRO-CTCAE is a useful tool to optimize inpatient dermatologic care for cancer patients by detecting and allowing management of patient-reported dAEs.

Data Availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable de-indentified request. Data are located in controlled access data storage at Memorial Sloan Kettering Cancer Center.