ECP in the spotLIGHT

In this issue of Blood, Gao et al demonstrate that extracorporeal photopheresis (ECP), when used as monotherapy in the first-, second-, or third-line therapy of patients with Sézary syndrome (SS) or erythrodermic mycosis fungoides (eMF), leads to prolonged disease control, as demonstrated by a median time to next treatment (TTNT) of 42 months and is superior to alternative non-ECP therapies used in the first-line setting (median TTNT, 3.5 months).¹ 

In contrast to early-stage mycosis fungoides (MF), which is largely managed with skin-directed therapies (“lotions and light”), patients with late-stage disease, including SS and eMF, benefit from a smorgasbord of systemic therapeutic options. Although largely incurable, immunomodulatory therapies, including interferon-α and ECP, may be associated with prolonged and durable remissions (reviewed in Wilcox² ). In stark contrast, and as previously demonstrated by this same group of investigators,³  responses achieved with conventional chemotherapeutic agents are transient and rarely durable.

ECP, as pioneered by Edelson et al,⁴  passes leukopheresis-enriched peripheral blood mononuclear leukocytes through a thin plate in which cells are exposed to UV-A–activated 8-methoxypsoralen, culminating in DNA crosslinking and cell death. Although disease classification and response criteria have certainly evolved over the 3 decades since ECP’s approval by the US Food and Drug Administration, the response rate initially reported by Edelson among erythrodermic patients (24 out of 29, 83%) is not dissimilar from the 69% response rate (in skin) reported here by Gao et al. Given the evolution of response criteria and the variable inclusion of additional agents with ECP in retrospective studies, perhaps most notably interferon-α, it is difficult to precisely estimate the proportion of exceptional responders for whom a complete response (in skin, blood, lymph node, and viscera) may be anticipated with ECP alone, but ≈10% may be a reasonable estimate.⁵  Although ECP is safe, efficacious, and cost-effective,⁶  its availability is largely confined to academic centers, and although different treatment schedules have been used, treatment is frequently given on 2 consecutive days every 4 weeks for at least 6 months, followed by a prolonged maintenance phase. Therefore, ECP is not universally available and poses significant logistical challenges for many patients. Although a head-to-head comparison is impossible, it is reassuring that the outcomes reported by Gao et al are comparable with many previous studies,^5,7  despite the use of an attenuated schedule (ECP once weekly for 6 weeks, then every 2 weeks for 12 weeks, followed by monthly treatment thereafter).

Despite its effectiveness and tolerability, widespread use,⁸  and endorsement by National Comprehensive Cancer Network (among other) guidelines, improved mechanistic understanding of its immunologic effects has proven challenging. This challenge, at least until recently, was largely explained by the absence of a murine model system. Nonetheless, a significant body of evidence, largely from Edelson’s group, demonstrates that ECP-dependent platelet immobilization (and activation) leads to significant transcriptional and functional changes in adherent peripheral blood monocytes, converting them into proficient antigen-presenting cells. More recently, studies performed in syngeneic tumor models using a miniaturized ECP (or “transimmunization”) chamber have convincingly demonstrated that ECP induces a tumor-specific immune response that is monocyte, platelet, and T cell dependent.⁹  Furthermore, the adoptive transfer of T cells obtained from ECP-treated mice to untreated recipients conferred protection to subsequent tumor challenge.

Increasing appreciation of both the genomic landscape in MF/SS, including its complexity and relatively high mutational burden, and the exceptional responses achieved with immunomodulatory agents, including abscopal effects (eg, with toll-like receptor agonists), suggest that malignant lymphocytes in MF/SS are sufficiently immunogenic to instigate an effective antitumor immune response. Therefore, in the absence of host- or tumor-related barriers to effective immunity, ECP alone may be highly effective, thus explaining the minority (≈10%) of exceptional responders (see figure). Unfortunately, the immunologic incompetence of the host, and the tumor microenvironment, which is profoundly immune suppressive, pose significant challenges for ECP and other immunotherapeutic strategies. Both disease progression and contraction of the normal T-cell repertoire, and prior therapies, particularly those that are cytotoxic and lymphodepleting, likely render the host resistant to immunotherapeutic strategies. This contention may be supported by the observation that an impressive median TTNT of 42 months was observed for patients who received early (treatment lines 1 to 3) ECP as monotherapy, but TTNT progressively shortened as ECP was initiated later (7 months) or combined with alternative agents (median TTNT 12 months when ECP was used early, either alone or in combination). These findings have significant clinical implications, because ECP may be underutilized in the first-line setting and overutilized in those who have received multiple lines of therapy.⁸ 

Of course, immunologic checkpoints (eg, programmed death-ligand 1¹⁰ ) and constituents of the tumor microenvironment (eg, macrophages, regulatory T cells) create a microenvironment that is hostile to effector T cells and poses a significant, but not insurmountable, barrier to immunotherapeutic strategies. Therefore, the identification of surrogate biomarkers for tumor immunogenicity (eg, genomic complexity and mutational burden), host immunologic competence (eg, T-cell repertoire), and the tumor microenvironment (eg, expression of immunologic “checkpoints”) may improve our ability to distinguish patients for whom ECP monotherapy is sufficient from those who may benefit from ECP in combination with other agents (see figure). However, at least for now, the findings reported by Gao et al are clear: in the course of SS/eMF treatment, ECP should take center stage early, enjoy the spotlight in the first act, be ushered off stage in the second, and rarely (if ever) be invited for an encore.