Thalidomide 100 mg/day edges out 50 mg/day for camrelizumab-related skin toxicity prevention
Camrelizumab’s most distinctive skin toxicity has become a familiar problem in oncology clinics: reactive cutaneous capillary endothelial proliferation, or RCCEP. In this multicenter randomized phase II trial, prophylactic thalidomide at 100 mg/day satisfied the prespecified incidence-based selection rule more convincingly than 50 mg/day, but the higher dose also brought more grade 3 or higher thalidomide-related toxicity. The signal is straightforward, even if the clinical tradeoff is not. A stronger antiangiogenic push appears to reduce the chance of RCCEP, yet the same dose increases the burden of off-target adverse effects.
Why RCCEP remains a practical problem during PD-1 therapy
Camrelizumab is a PD-1 inhibitor used in first-line treatment for advanced esophageal squamous cell carcinoma and non-small cell lung cancer. Like other immune checkpoint inhibitors, it restores antitumor T-cell activity by interrupting PD-1 signaling. RCCEP is unusual among cutaneous adverse events because it reflects a vascular proliferative process rather than a classic inflammatory rash. Lesions often appear as red papules or nodules, sometimes solitary and sometimes multifocal, and may ulcerate or bleed. Histology is characterized by capillary-like proliferation and endothelial hyperplasia in the dermis.
The biology makes prophylaxis appealing. If angiogenic drive helps produce the lesion, then an agent with antiangiogenic activity should, in theory, blunt its emergence. Thalidomide fits that logic. Its vascular effects have been explored for years, and earlier studies suggested that thalidomide prophylaxis could reduce camrelizumab-associated RCCEP. This trial asked a narrower question: which dose is more promising for future evaluation, 50 mg/day or 100 mg/day?
Randomized dosing in 72 patients across 11 centers
The prevention cohort enrolled 72 patients and randomized them equally to two groups. Group A received thalidomide 50 mg/day, and Group B received 100 mg/day. All patients also received camrelizumab 200 mg every 3 weeks plus platinum-based doublet chemotherapy as first-line treatment for advanced NSCLC or esophageal squamous cell carcinoma. Thalidomide began on the same day as the first camrelizumab dose and continued until 4 weeks after the last camrelizumab dose. The study was open-label, with skin and mucosal examinations every 3 weeks, and investigators used prespecified grading criteria based on the 2023 CSCO guideline for immune checkpoint inhibitor-related toxicities.
The primary endpoint was overall RCCEP incidence. Secondary endpoints included grade 3 or higher RCCEP, time to onset, landmark RCCEP incidence at week 6 and week 9, and thalidomide-related adverse events. No formal superiority test was planned. Instead, the protocol used a selection framework: an absolute difference of more than 5 percentage points in overall RCCEP incidence would identify the more promising regimen.
The lower dose did not win on incidence
RCCEP developed in 12 of 36 patients in the 50 mg/day group, or 33.3%, and in 9 of 36 patients in the 100 mg/day group, or 25.0%. The absolute difference favored the higher dose, although the gap stayed modest. Week 6 RCCEP incidence was 13.9% with 50 mg/day and 19.4% with 100 mg/day; by week 9, the values were 25.0% and 19.4%, respectively. The overall pattern suggests that RCCEP can emerge early during treatment and that weekly or cycle-based surveillance matters because the lesion may appear before maintenance therapy begins.
No grade 3 or higher RCCEP occurred in either arm. All first RCCEP events were grade 1 except for one grade 2 event in the 50 mg/day group. Among patients who did develop RCCEP, median onset time was 1.49 months in the lower-dose group and 1.31 months in the higher-dose group. The event timing matters at the bedside because prophylaxis must be in place before the vascular lesion appears; once lesions are established, the opportunity for prevention has already passed.
Toxicity rose with the 100 mg/day dose
Safety tilted in the opposite direction. Thalidomide-related adverse events occurred in 22.2% of patients receiving 50 mg/day and 41.7% of those receiving 100 mg/day. Grade 3 or higher thalidomide-related adverse events occurred in 8.3% and 22.2% of patients, respectively. Serious adverse events occurred in 11.1% of the lower-dose group and 13.9% of the higher-dose group.
This is the central tension in the trial. A higher dose improved the incidence-based selection result, but the cost was a higher burden of clinically important toxicity. For a supportive intervention meant to preserve tolerability during cancer therapy, dose escalation cannot be judged only by lesion prevention. A prophylactic drug that prevents a dermatologic complication but creates a substantial new adverse-event load may still fail the bedside test, especially in patients already receiving platinum chemotherapy and immunotherapy together.
What this selection trial can and cannot answer
The study did what a phase II selection trial is supposed to do: it identified the dose that looked more promising for a future controlled evaluation. By the prespecified rule, thalidomide 100 mg/day met the incidence-based threshold compared with 50 mg/day. Yet the trial does not prove that 100 mg/day is the best long-term choice for routine practice, and it does not establish a definitive risk-benefit balance.
Several design features keep the interpretation appropriately restrained. The study was open-label. RCCEP assessors were not masked. No central adjudication was performed, even though investigators used standardized definitions, photographic documentation, and dermatology consultation for atypical cases. The trial also did not prespecify a safety-based dose-selection rule, so the higher toxicity burden does not automatically disqualify the higher dose, but it clearly complicates the choice.
The treatment cohort from the parent study was stopped early because thalidomide showed insufficient therapeutic activity for established RCCEP, leaving this paper focused only on prevention. That distinction is clinically important: prevention and treatment are not interchangeable questions, and success in one setting does not guarantee success in the other. The preventive signal here is real, but it sits inside a narrow frame defined by modest sample size, open-label assessment, and a tradeoff between vascular prophylaxis and systemic toxicity.
Where the field now stands
For clinicians managing camrelizumab-based regimens in advanced esophageal squamous cell carcinoma or non-small cell lung cancer, the message is not that thalidomide should be added reflexively. The message is that prophylaxis against RCCEP is biologically plausible and clinically relevant, and that the 100 mg/day dose deserves further controlled study because it better met the prespecified selection criterion. Any follow-up trial will need to answer a harder question than incidence alone: can a higher prophylactic dose prevent enough RCCEP to justify the added toxicity burden across a population already exposed to combined systemic therapy?