The safety and efficacy of low-dose interleukin-2 for the treatment of SR-cGvHD

Steroid-refractory chronic graft-versus-host-disease (SR-cGvHD) is characterized by clinical manifestations in multiple organ systems that greatly vary among patients following allogeneic stem cell transplantation.¹ It represents a difficult to treat condition, leading to morbidity and poor quality of life. Therefore, new and emerging treatment options present the opportunity to improve response and survival rates in SR-cGvHD.¹

Here, we discuss the combined results of five phase I or II clinical trials (NCT00529035, NCT01366092, NCT01937468, NCT02318082, and NCT02340676) conducted at the Dana-Farber Cancer Institute between 2008 and 2017, which have been published in Blood Advances by Kim, et al.¹ These trials investigated the efficacy and safety of daily low-dose interleukin-2 (LD IL-2) therapy to manage SR-cGvHD in patients with organ involvement. IL-2 is a regulator of CD4⁺ regulatory T cell (Treg) differentiation and proliferation; therefore, IL-2 can be utilized to enhance immune responses in patients with GvHD.¹

Study design

A total of 123 patients were enrolled across the five trials. The initial treatment period was 8 weeks for all trials, with the exception of NCT01366092, which had an initial treatment of 12 weeks. In NCT01937468, LD IL-2 was administered alongside donor Tregs, and in NCT02340676, LD IL-2 was administered with extracorporeal photopheresis. For all trials, treatment with LD IL-2 after the initial 8 weeks was allowed to be continued for patients with complete response, partial response, or stable disease.

Results

Baseline characteristics

Of the 123 patients enrolled, 14 were not evaluated for response. Four patients were involved in multiple clinical trials; in these cases, only data from the first trial are included. Patient characteristics of the 105 patients evaluated are shown below in Table 1.

Table 1. Baseline patient characteristics*

Safety and efficacy

The median follow-up across all five trials was 69 months (range, 2–145 months). Across the cohorts, there were a total of 41 patient deaths; 19 from GvHD, five from disease recurrence, and five from new malignancies.

The 5-year overall survival (OS) and progression-free survival was 63%, with a 5-year OS of patients with lung, liver, or gastrointestinal involvement of 62%, compared with 64% in patients without these involvements (p = 0.95). Involvement of different organs in cGvHD and response rates in those organs are shown in Table 2. Eyes, mouth, and genital tract were involved in 64%, 50%, and 5% of patients, respectively; these organs however were not assessed for response. While these variables did not seem to influence the survival outcomes, age was found to be significantly associated with OS in univariable and multivariable analyses (age, ≥50 years vs <50 years; hazard ratio, 2.4; p = 0.028).

Table 2. Organ involvement and organ-specific response rates*

Overall response rate was evaluated after either 8 or 12 weeks of LD IL-2 therapy and was found to be 48.6% and 53.3%, respectively. Fewer prior therapies (<4 vs ≥4) and less time to therapy initiation (<1.6 years vs ≥1.6 years) was associated with overall response and skin response in further analyses.

All trials apart from NCT00529035 collected data on the cGvHD symptom score, as reported by the patients. Across all patients, both skin and muscle symptoms significantly improved from pre-treatment to post-treatment analysis (median score for skin, 5 and 3, respectively [p = 0.001]; median score for muscle 6 and 4.5, respectively [p = 0.015]).

Immunologic response

Treatment with LD IL-2 led to elevated levels of CD4⁺ Tregs above baseline levels for the whole duration of treatment; with these levels peaking at Weeks 2–4 of treatment. There was also an increase in natural killer cells both during and after treatment. There was found to be no significant differences in the CD3⁺ and CD8⁺ T cell counts in responding versus non-responding patients; however, the ratio of CD4⁺ Tregs to CD4⁺ conventional T cells increased from baseline levels in responding and non-responding patients, with a higher increase in responders. This suggests IL-2 was able to effectively enhance immune responses in these patients.

In patients with skin response, B cell count was higher at 1 year of follow-up compared with non-responders (median, 12% vs 4%, respectively; p = 0.001), with a significantly higher increase in the CD4⁺ Treg to CD4⁺ conventional T cell ratio in responders. In patients with joint/muscle/fascia involvement, B cell counts and the CD4⁺ Treg to CD4⁺ conventional T cell ratio was similar to patients with skin involvement.

For patients with lung involvement, results did not overlap with skin response. There was no significant difference in the CD4⁺ Treg to CD4⁺ conventional T cell ratio between lung responders and non-responders, with lower T cell counts in responders after 4 weeks of treatment.

Conclusion 

As shown in the analysis, skin and joint, muscle, and fascia were the most common organs involved in cGvHD for this cohort of patients. However, response rates in these organs were only around 30%, compared with a 66% response in the liver after treatment with LD IL-2. The increase in the CD4⁺ Treg to CD4⁺ conventional T cell ratio after treatment with LD IL-2 demonstrates that this therapy is effective in increasing immune tolerance in these patients, which in turn can improve their GvHD symptoms. Despite this, response rates in patients treated with prior therapies is poor, and potentially combining multiple drugs to target different signaling pathways could result in higher organ-specific and overall response rates in SR-cGvHD.