Safety and preliminary efficacy of adding tocilizumab to cisplatin/docetaxel for the treatment of locally advanced triple-negative breast cancer patients: prospective phase 1/2 clinical trial.

Background
Triple-negative breast cancer is an aggressive breast cancer (BC) subtype, which accounts for approximately 15%–20% of newly diagnosed BC but with worse outcomes and higher recurrence rates than the other subtypes1,2. Conventional chemotherapy remains the primary treatment option for these patients, mostly as anthracycline-taxane-based regimens3. Neo-adjuvant chemotherapy (NACT) with platinum-based agents has produced therapeutic benefits, especially among chemo-sensitive TNBC patients4,5. Recently, the addition of immunotherapy to NACT has significantly increased pathological complete response and survival among patients with early TNBC6,7. However, improvements in clinical outcomes and toxicity profile are still needed.
TNBC cells secrete high levels of IL-6, which is a pleiotropic pro-inflammatory/-carcinogenic cytokine8. IL-6 acts through a specific receptor (IL-6R) and plays a major role in the pro-carcinogenic cross-talk between cancer cells and their microenvironment9,10. IL-6 is also one of the main determinants for TNBC’s cell proliferation and chemoresistance mechanisms11. Tocilizumab (Actemra) is a monoclonal anti-IL-6R antibody that binds to and inhibits both soluble and membrane-bound IL-6R (sIL-6R and mIL-6R) with equal affinity12. Tocilizumab is used to treat several IL-6-dependent chronic inflammatory diseases, including rheumatoid arthritis and cytokine release syndrome. Additionally, tocilizumab can reduce tumor growth and chemosensitize cancer cells to cytotoxic agents13–17. We have recently shown that tocilizumab can inhibit the IL-6/STAT3 signaling pathway in both breast cancer cells and their active stromal fibroblasts18,19. Furthermore, when administered first, tocilizumab potentiated the cytotoxic effects of cisplatin both in vitro and in humanized orthotopic breast tumor xenografts19. These preclinical findings prompted us to test the efficacy and tolerability of tocilizumab followed by chemotherapy in locally advanced TNBC patients as part of the first-in-human clinical trial.

Methods

Study design and oversight
This was an investigator-initiated, prospective, dose-escalation, open-label, non-randomized single-arm phase I/II clinical trial that took place at the King Faisal Specialist Hospital & Research Centre in Riyadh. Women were invited to participate in this study if they had histologically proven unilateral triple-negative locally advanced non-inflammatory breast cancer (clinical tumor size ≥ 4 cm, T2-T4, N0–N3, M0). Patients were eligible if they were at least 18 years of age, had an ECOG performance status ≤ 1, neither pregnant nor nursing. All patients are required to have adequate hematologic, cardiac, hepatic, and renal function. Patients were also excluded from participation if they had received surgery, radiotherapy, or chemotherapy for any malignancy except adequately treated basal cell carcinoma of the skin or in situ cervical cancer. All the patients were evaluated with a bilateral mammogram, mammary ultrasound, bone scan, computed tomography of the chest, abdomen, and pelvis, or PETCT scan ± MRI if indicated to rule out distant metastasis. All patients had to undergo a true-cut biopsy; surgical specimens were evaluated by a board-certified pathologist; the diagnosis of invasive breast cancer was confirmed. Assessment for the grade, lympho-vascular invasion, Ki67 labeling index, and stromal tumor-infiltrating lymphocytes (sTILs) was performed by hematoxylin and eosin staining and multiplex immunohistochemistry. sTILs were scored on a binary scale of ≤ 30% or > 30%, estrogen and progesterone receptors, and HER2 were determined on pretreatment biopsy by immunohistochemistry (IHC). Hormone receptor status was considered negative if ≤ 9% of tumor cells stained for ER and/or PR. HER2 status was determined by immunohistochemistry (IHC) and confirmed with fluorescent in situ hybridization if IHC equivocal (2+), a FISH ratio of > 2 was considered positive, or an average of > 6 copies of the HER2 gene per nucleus for systems without an internal central probe.

Treatment schedule
The dose escalation scheme used a traditional 3 + 3 design, with an initial dose of tocilizumab of 8 mg/Kg IV (intravenous) on day 1, and with the dose of cisplatin and docetaxel escalated by 10 mg/m2 IV in each cohort given on day 2 of a 4-week schedule for a total of 6 cycles. The study had two phases: a dose-escalation phase and an expansion phase. The objective of phase 1 was to evaluate the maximum tolerated dose (MTD) and the dose-limiting toxicity (DLT) to determine the recommended dose (RD) to be utilized in the sequential combination of tocilizumab with cisplatin/docetaxel. The phase 1 part of the trial was conducted following the classical 3 + 3 study design, 3 patients were assigned to each level, and if no evidence of DLT was observed, the next dose level was initiated. In case one of the 3 patients showed DLT, an additional 3 patients were entered at that dose level before dose escalation. If 2 or more of the 6 patients demonstrated DLT or 2 of the first 3 patients have shown DLT to a dose, this particular dose was defined as MTD.
DLT was defined as follows: (1) Grade 4 neutropenia (2) grade 4 thrombocytopenia (3) Grade 3 neutropenia with fever > 38c (4) Grade 3 nonhematological toxicities except for nausea, vomiting, appetite loss, and general fatigue. The objective of phase 1 was to assess the safety and efficacy of the tocilizumab followed by cisplatin/docetaxel. The primary endpoint was pCR in all eligible patients, including patients who received treatment at the RD level in the phase 1 study. The secondary end point was progression-free survival and overall survival (Fig. 1). Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria version 5.0. Dose discontinuations, dose modifications, and delays of the allocated interventions followed their approved labelling.

Clinical response was assessed by physical examination using a caliper at each follow-up before each treatment cycle and defined as a complete response, partial response, or progressive disease according to Response Evaluation Criteria in Solid Tumors. No radiological evaluations were performed during the study. Once the patients underwent surgery, pCR was defined as the complete absence of viable invasive tumor cells in the breast and axillary nodes, including surgical margins without ductal carcinoma in situ (ypT0/ypN0) according to NSABP B-1820. RCB was scored for all patients using Symman’s criteria21. Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria version 5.0.
A sample of 34 subjects was anticipated, with up to 12 subjects accrued during a dose-ranging phase, and 16 additional subjects were accrued during a dose-expansion phase. In the dose-ranging phase, the planned cohort size was not based on a specific statistical hypothesis but on experience in conducting similar trials in cancer patients. In the dose expansion phase, the enrollment of 28 subjects (12 from the dose-ranging portion and 16 from the dose-expansion portion) was estimated to allow a preliminary evaluation of the overall response rate (ORR) associated with the study treatment.
The study followed good clinical practice (GCP) and was conducted according to the Declaration of Helsinki. All patients provided their written informed consent to participate in the study. This study was approved by the institutional review board (IRB) of King Faisal Specialist Hospital and Research Centre (KFSHRC) under IRB RAC# 2181156 on 04/17/2019. Further approval was obtained from the Saudi Food and Drug Authority (SFDA) under (SCTR NO: 1911002). The study was conducted from July 2019 to September 2022.

Statistical analysis
Due to the exploratory nature of the study, no confirmatory inferential analyses were planned. The statistical analyses included descriptive statistics, presented as percentage or mean values with standard deviations, medians, and ranges to summarize the patients’ characteristics, treatment administration/compliance, safety parameters, and antitumor activity endpoints. A total of 30 patients were planned for enrollment at the MTD and/or the recommended phase 2 dose (RP2D). pCR and safety outcomes were estimated using proportions with confidence intervals calculated using the Wilson binomial method, which provides more reliable estimation for binomial proportions, especially in small samples. Survival analysis was conducted using the Kaplan–Meier method to calculate OS and DFS. Overall survival (OS) was defined from the date of diagnosis to death or last follow-up, and disease-free survival (DFS) from the date of diagnosis to relapse or death; patients without an event were censored at the last follow-up. The log-rank test was used to determine whether survival rates corresponded to pCR in the breast and axilla. A two-sided P value of < 0.05 was considered statistically significant. All statistical analyses were performed using SAS (version 9.1).

Results
The full analysis set consisted of 30 patients enrolled between July 2019 to September 2022 with a median follow-up of 21.3 (95%CI: 19.2–23.4) months. Two of these patients were ineligible (one had metaplastic carcinoma and one with secretary low-grade carcinoma), 28 patients completed the study, (12 patients were enrolled in the phase 1 portion and 16 in the phase 2 portion), 26 patients underwent definitive surgery (one declined surgery and is still on follow up, while the other one was deceased, but the cause of death was not related to the study). As shown in Table 1, the median age was 42 years (range 25–65), 23% clinical stage IIB, 77% stage III, 80% grade III, 67% had nodal involvement at baseline, 27% had deleterious mutation in BRCA 1/2 (other genes such as PALP2, CHEK2 and RAD51c were not sequenced), 17% had lumpectomy and 77% mastectomy, all patients who had risk of relapse had adjuvant radiotherapy as per institutional guidelines. Indeed, 21 out of 28 patients (75%) received adjuvant radiation therapy, despite a pathologic complete response in both breast and axilla. The adjuvant radiation therapy to breast/chest wall with regional nodal irradiation was administered based on the initial clinical stage at presentation (T3-T4/N2-N3), in accordance with contemporary standards of care. Adjuvant PARP inhibitors or capecitabine were not allowed for patients who had residual disease. Furthermore, as mandated by the SFDA research committee, tocilizumab was not given postoperatively.

In the secondary endpoint analysis, the overall pCR (RCB class 0) and (RCB class 0 + 1) were 62% and 69% respectively; pCR in patients with BRCA-associated mutations and wild type were 62% and 100%, respectively; while pCR in the axilla was achieved in 81% of patients (Table 2). Recently, the concept of near-pCR was proposed, and according to the most common consensus, it was the residual tumor of ≤ 10 mm with negative axilla. In our study, 16 patients (62%) achieved pCR, and 5 patients (19%) achieved near-pCR. Therefore, the overall pCR and near-pCR were 21 patients (81%), and none of these patients developed recurrent disease (Table 3).

The Kaplan-Meier curves for 3-year disease-free survival (DFS) and 3-year overall survival (OS) showed 80% and 90% survival rates, respectively (Fig. 2). The median follow-up for both OS and DFS had not been reached at the time of analysis.

The safety analysis set included 28 patients. All patients were assessable for toxicity; according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (NCI CTCAE) version 5.0. There was no treatment-related death in the study protocol. All cases of serious adverse events (AEs) were improved or resolved with appropriate treatment. The most common hematologic and non-hematologic toxicities are summarized in Table 4. The most frequently reported AEs were grade 1–2: nausea (53%), joint and muscle pain (29%), vomiting (25%), abdominal pain and diarrhea (18%), peripheral neuropathy (14%), febrile neutropenia (11%), and skin rash (11%) (Table 4). Potential tocilizumab-induced effects on total cholesterol and triglyceride were reported in 50% of patients (Table 4). Grade 3–4 AEs were neutropenia (39%), febrile neutropenia (21%), increased AST and ALT (11%), nausea (3.5%), vomiting (7%), and mucositis (3.5%) (Table 4). Dose de-escalation of chemotherapy based on toxicity occurred after a median of 3 cycles in only 3 patients.

Discussion
To our knowledge, the present clinical trial is the first study addressing the possible use of the anti-IL-6R (tocilizumab) in the treatment of TNBC patients. In this trial, 28 patients completed the treatment protocol and 26 underwent definitive surgery, 62% achieved RCB-0 and 69% RCB 0 + 1, pCR in breast and axilla was reached by 65% and 81% of patients, respectively (Table 2). Only, two patients experienced disease relapse. The Kaplan- Meier curve for DFS and OS showed 80% and 90% survival rates, respectively. Six patients had residual disease RCB2 and 2 patients had RCB3. In our previous study, treatment of locally advanced TNBC patients with frontline doxorubicin followed by cisplatin/docetaxel, pCR in breast and axilla was achieved in only 24%22. In the neoSTOP trial, which included a low-risk population, a pCR rate of only 54% was reached in breast and axilla with 6 cycles of carboplatin/docetaxel23. However, in the present study, 76% of the patients were stage III, and 50% of them were stage IIIC, unfortunately in our study the relationship between TNBC molecular profiling and different treatment response was not evaluated. Importantly, 8 of 26 patients (27%) with deleterious BRCA mutations achieved 100% pCR rate compared with 18 of 26 patients with BRCA wild type that achieved a 66% pCR rate, with an absolute difference of 34%. In fact, patients with tumors bearing BRCA mutations show an average pCR rate higher than in BRCA non-carrier patients (59% vs. 30%). In one study, neoadjuvant treatment of TNBC patients with carboplatin/docetaxel achieved a 59% pCR rate, similar to what Byrski et al. have reported (pCR rate of 61% from BRCA1 positive breast cancer patients)24,25. This suggests that addition of tocilizumab to platinum/docetaxel regimen improves the outcome of both BRCA-negative and -positive breast cancer patients.
In the present study, sTILs were categorized into high and low groups with a cut-off value of 30%. While patients with high sTILs (10/26, 34%) showed a pCR rate of 70%, patients with low sTILs (16/26, 66%) exhibited a pCR rate of 59%. Therefore, in the presence of tocilizumab, the efficacy of the treatment was similar for both groups, suggesting tocilizumab-dependent improvement of the treatment of patients with low sTILs. In fact, high sTILs count in TNBC has been associated with higher pCR to neoadjuvant chemotherapy and favorable outcomes in terms of DFS and OS25–27. IL-6 signaling is a major determinant of TNBC cell proliferation and viability, and this inflammatory cytokine is responsible for resistance mechanisms in TNBC11. Chung et al. have previously shown that adding tocilizumab to docetaxel delays tumor progression in TNBC patients compared to docetaxel alone by negating autocrine IL-1α-dependent induction of IL-6 and neutralizing IL-6 signaling by antagonizing the IL-6 receptor. In a phase I recurrent ovarian cancer trial, it has been shown that tocilizumab, combined with carboplatin, doxorubicin, and interferon-α2b, had no dose-limiting toxicities, and IL-6 receptor inhibition was determined as a safe and feasible method to sensitize tumor cells to chemotherapy28. In addition, a phase I/II open-label multicenter, randomized umbrella trial is underway to evaluate the efficacy and safety of multiple immunotherapies, including tocilizumab, in patients with metastatic TNBC (Morpheus-PanBC, NCT 034244005).
Tocilizumab has been shown to worsen dyslipidemia, a significant adverse metabolic effect that is attributed to the blockade of IL-6 and its ability to induce the expression of apolipoproteins in the liver29. In the present study, total cholesterol and triglycerides levels were significantly increased in 14 patients (50%), these changes were observed after the second cycle, and all patients were well-controlled with statins.
The most frequently observed dose-dependent and cumulative cisplatin-caused adverse effects, seen in 30%−40% of patients, is nephrotoxicity, which is manifested as non-oliguric acute kidney injury30. The formation of reactive oxygen species and the high expression proinflammatory cytokines such as TNF-α, IL-6, or IL-1β contribute to kidney injury. It has been previously reported that using anti-IL6 and anti-TNFα provided the greatest resistance to cisplatin nephrotoxicity31. Importantly, in the current study only 2 patients (7%) presented with mild reversible non-oliguric acute kidney injury, and no fatigue or weight loss were reported despite the use of platinum-based chemotherapy. It has been recently reported that tocilizumab is an efficient treatment for severe paraneoplastic inflammatory syndrome (PIS) offering significant improvement in clinical symptoms and biological abnormalities, including fever, fatigue, weight loss, and cachexia in patients suffering solid tumors32.

Conclusion
Despite its limitations (early phase, enrolling a small cohort of patients, adjuvant PARP inhibitors or capecitabine were not allowed for patients who had residual disease, and tocilizumab was not given post-operatively), this clinical trial shows that the sequential administration of tocilizumab followed by cisplatin/docetaxel is well tolerated and safe as a neoadjuvant therapeutic regimen for locally advanced TNBC patients.