Time to next treatment and safety assessment in cutaneous-T-cell lymphomas: a retrospective analysis on patients treated with bexarotene and acitretin

Abstract

Dear Editor, Oral retinoids are among the most common systemic therapies prescribed to patients with cutaneous T-cell lymphomas (CTCLs),1 with response rates ranging from 45% to 77·3%.2, 3 As patients with CTCL commonly endure high symptom burden, with multiple lines of treatment frequently resulting in incomplete responses and toxicities, effectively assessing the durability of clinical benefit has been one of the greatest challenges in past years.4, 5 Recently, the surrogate endpoint ‘time to next treatment’ (TTNT), defined as the interval from the date of initiation of a treatment to the date of commencement of the next line of therapy (excluding skin-directed therapies and short-term treatment gaps), has been regarded as a promising tool in assessing clinical benefit in CTCL. Previously, TTNT values between 5·4 and 22·1 months were reported for patients treated with retinoids.6, 7 To shed light on the efficacy and safety of bexarotene and acitretin, the most prescribed retinoids for CTCL in our centre, we performed a retrospective cohort study, screening all patients who visited between January 2019 and January 2022. Inclusion criteria were a biopsy-confirmed diagnosis of CTCL, the presence of complete medical charts, and recorded acitretin and/or bexarotene therapy duration time. Best overall response (BOR) was assessed in conformity with standardized skin response definitions.8 Patients who received both retinoids were included in the analysis group of the earlier administered therapy. Therapy duration was recorded and median TTNT compared through a log-rank test (Stata Statistical Software Release 17; StataCorp LLC, College Station, TX, USA). Unacceptable toxicities leading to treatment switch were also evaluated. Out of 304 patients with CTCL, 87 (28·6% of our cohort) received retinoids and were analysed. Among them, five patients lacked a complete medical record and were excluded. Characteristics of the 82 analysed patients are summarized in Table 1. (The data that support the findings of this study are available from the corresponding author upon reasonable request.) In the bexarotene group, median TTNT was 299 days (range 68–2499), i.e. 9·8 months, while in the acitretin group, median TTNT was 441·5 days (range 74–1826), i.e. 14·5 months. For patients undergoing the second line of retinoids, recorded median TTNT in the bexarotene followed by acitretin subgroup was 293 days (range 49–1503), i.e. 9·6 months, while in the acitretin followed by bexarotene subgroup, median TTNT was 345 days (range 28–951), i.e. 11·3 months. No statistically significant difference (P = 0·625) regarding median TTNT was found through log-rank test (P > 0·05). TTNT triggered by unacceptable toxicity was twice as likely in the bexarotene compared with the acitretin group (38·1% vs·15%), with a statistically significant difference according to a χ2-test (P = 0·018, significant at P < 0·05). To date, no conclusion as to superiority in clinical efficacy of one retinoid over the other has been made, due to heterogeneity of the published evidence and the absence of direct comparisons. Therefore, molecule choice is usually based on clinician’s preference.1 According to our data, some interesting observations can be made. Firstly, bexarotene and acitretin display similarities in terms of BOR (ranging from 66·6% to 75%), with comparable therapy duration times. Secondly, TTNT triggered by toxicity was twice as likely in the bexarotene compared with the acitretin group (38·1% vs.15%) (P = 0·018, significant at P < 0·05), with blood lipid anomalies as the most common cause (13 patients), followed by skin toxicity (four patients). Interestingly, in our study, patients with more advanced disease (i.e. IIB) were more likely to receive bexarotene as the first therapeutic option, while acitretin was more commonly prescribed to older patients (median age 73·5 years vs. 71 years) and as the first option in patients with milder disease (i.e. 42·9% of patients in the bexarotene group had previously received treatment with acitretin). Though the small sample size of this study, along with its retrospective nature, prevents us from drawing definitive conclusions regarding which retinoid should be prescribed as first choice, our study confirms that both agents, bexarotene and acitretin, play an important role in tackling CTCL. Overall response rates are in line with those previously reported in the literature,2, 3 confirming a higher activity of retinoids in early stages (Table 1). This evidence consolidates the emerging paradigm of CTCL treatment as a circular model, where therapy options can be upgraded and downgraded based on clinical improvement or worsening. Finally, it confirms the promising role of TTNT as a powerful tool in encompassing all those aspects – such as treatment efficacy, toxicities, duration of clinical benefit and patient compliance – that need to be considered when treating patients with accumulated exposures to multiple lines of therapy and low likelihoods of cure.5 Gabriele Roccuzzo: Conceptualization (lead); data curation (equal); supervision (equal); writing – original draft (equal); writing – review and editing (equal). Paolo Fava: Conceptualization (lead); data curation (equal); supervision (lead); writing – original draft (equal); writing – review and editing (equal). Gianluca Avallone: Data curation (supporting). Carola Aquino: Data curation (equal). Sara Boskovic: Data curation (equal). Nicole macagno: Data curation (equal). Simone Ribero: Supervision (equal). Pietro QUAGLINO: Supervision (lead); validation (lead); writing – review and editing (lead). Open Access Funding provided by Universita degli Studi di Torino within the CRUI-CARE Agreement.

Topics & Concepts