Cabozantinib for different endocrine tumours: killing two birds with one stone. A systematic review of the literature

Systematic research of articles published between 2009 and 2023 on MEDLINE (PubMed), Scopus and EMBASE was performed. For search strategy and search terms refer to review protocol (Supplementary File 1). Reference lists were manually screened for further relevant articles. The search strategy was based on the PICO approach (a standardised way of defining research questions, focusing on Patients, Intervention, Comparison, and Outcome) [4]. The review was registered with PROSPERO registration number CRD42023403886 (https://​www.​crd.​york.​ac.​uk/​prospero/​).

Titles and abstracts were first reviewed. Eligible studies were assessed on the basis of their full text and referenced using Zotero Software v6.0.26 (Zotero.org). Preferred Reported Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed (see Supplementary File 2) [5].

MTC is a rare neuroendocrine disease arising from parafollicular C-cells of the thyroid, thus producing CT. It encompasses about 2–5% of all thyroid cancers, occurring as part of MEN2 syndrome in about 25% of cases and sporadically in 75% [13]. RET mutations can be detected at a germline level in familial forms, as well as at a somatic level in up to 50–90% of histological samples, according to literature results [14]. Irrespective of RET mutation status, MTC discloses poor prognosis in the event of distant metastases at diagnosis or during the follow-up [13]. At early stages, complete surgical resection represents a curative treatment; while the management of advanced and progressive MTC remains challenging [15]. Cytotoxic chemotherapy and radiotherapy showed limited efficacy, due to different mechanisms, including higher expression of Multi Drug Resistant 1 (MDR-1) proteins and enhancement of Heat Shock Protein pathways [16]. By reason of the low PDL1 expression on C-cells, disappointing results have been observed with immune-checkpoint inhibitors [17]. Besides, experience with peptide receptor radionuclide therapy (PRRT) is still limited, and an objective response is observed in only 10–15% of patients [18]. The best results in the treatment of advanced and progressive MTC came from TKI drugs. In particular, vandetanib and CBZ (Cometriq®), had been approved by the United States Food and Drug Administration (FDA) in 2011 and 2021, and by the European Medicine Agency (EMA) in 2012 and 2022, respectively. From a molecular perspective, in vitro and in vivo studies on mouse models demonstrated that, besides the robust MET and VEGFR2 antagonism, CBZ inhibits both the wild type and the mutated forms of RET, preventing MTC cells and xenograft from growing and metastasising [19]. As regards clinical trials, in phase I (Study of XL184 (Cabozantinib) in Adults With Advanced Malignancies) and III trials (EXAM trial), CBZ showed an acceptable safety profile (Table 2), a longer progression-free survival (PFS) (11.2 months versus 4.0 months in the placebo arm, p < 0.001) and overall survival (OS) (26.6 versus 21.1 months, although the difference did not reach statistical significance, p = 0.24) [20, 21]. These promising results were obtained mainly in patients with RETM918T mutation (which was present in almost 65–70% of the tumours) [22]. In fact, according to the exploratory analysis of this study, RETM918T subgroup achieved the greatest observed PFS benefit for CBZ versus placebo (HR, 0.15; 95% CI, 0.08–0.28; P < 0.0001) [23]. Furthermore, an indirect treatment comparison between CBZ and vandetanib showed a positive trend in favour of CBZ in terms of PFS [24]. As regarding adverse effects, CBZ showed a dose-dependent relation but the EXAMINER trial failed to prove the non-inferiority of the 60 mg/day tablets versus the 140 mg/day capsules [25]. Moreover, Koehler et al. study found that both PFS and OS were significantly longer in younger patients experiencing more than five AEs, thus suggesting a dose-related correlation with the tumour response [26]. In conclusion, in MTC, CBZ appeared to be effective in prolonging survival, although many patients often do not receive full dosage or need a discontinuation of the treatment due to poor tolerance or toxicities (Table 2).

DTCs, which account for about 90% of all thyroid cancers, usually have favourable long-term prognosis following surgery±radioiodine treatment, with mortality rates ranging from 0.3–0.5/100,000 population [27, 28].

However, two-thirds of patients with recurrent or metastatic DTC are RAIR with a 10-year survival rate of only 10% after diagnosis, resulting in a challenge for physicians [29].

While indolent and oligometastatic RAIR-DTC can be followed up with suppressive levothyroxine dose alone; symptomatic, progressive low-burden disease may benefit from locoregional therapies i.e., surgery, external radiotherapy, or thermal ablation. The major challenge lies in rapidly progressive and inoperable cases or high-burden metastatic RAIR-DTC, which requires systemic therapies. In those patients, the treatment with cytotoxic chemotherapy showed disappointing results, with selective benefits only in patients who fail to respond to TKIs (data are mainly anecdotal) [30]. Hence, TKIs are considered in the event of disease progression, mainly after surgical and radiation therapy approaches [31]. Until fairly recently, lenvatinib, a wide spectrum anti-angiogenetic TKI, was the only available option for advanced RAIR-DTC and, at the time of progression to this drug, the therapeutic chances were poor [32]. In 2021 and 2022, CBZ (Cabometyx®) was finally approved by FDA and EMA for RAIR-DTC patients, who disclosed progressive disease after prior treatments with VEGFR inhibitors. This endorsement was based on the promising data from the COSMIC-311 study: a multicentral, randomised, double-blind trial in which patients, with locally advanced or metastatic RAIR-DTC, progressing during or after treatment with at least one VEGFR-targeting TKI, were treated with either CBZ 60 mg orally once daily or placebo [3]. The median PFS was 11.0 months [95% confidence interval (CI), 7.4–13.8] in the CBZ arm compared with 1.9 months (95% CI, 1.9–3.7) in the control arm [3]. The final analysis of COSMIC-311 with longer follow-up confirmed the superiority of CBZ versus placebo with a manageable safety profile [3]. The PFS benefit was consistent with the interim analysis and irrespective of prior VEGFR-targeted therapy [33]. Phase I and II trials that preceded the COSMIC trial showed that CBZ presented a safety profile similar to other multitargeted VEGFR and clinical and statistically significant activity in DTC patients who have progressed on prior VEGFR targeted therapies [34‐36]. Furthermore, Konda et al. performed a prospective analysis as part of a multicenter International Thyroid Oncology Group phase II study of CBZ in patients with RAIR-DTC with bone metastases who progressed on prior VEGFR‐targeted therapy: CBZ was associated with a highly significant reduction in all bone turnover markers (even when they were within normal limits at baseline) and a minor decreases in bone lesion uptake was noted on Sodium Fluoride PET in four out of six patients [37]. Finally, CBZ provided the same efficacy advantage when in monotherapy and when associated with immune-checkpoint inhibitors such as nivolumab and ipilimumab, while it showed a durable response and high rate of disease control when in combination with atezolizumab [38, 39]. In 2018, Brose et al. presented the first study documenting CBZ anti-tumoral activity in patients with RAIR-DTC as a first-line therapy [40]. A phase I trial on the simultaneous use of CBZ and PD-1 inhibitors in HIV-positive patients is now recruiting (Table 3 - https://​clinicaltrials.​gov).

MCC is a rare, highly aggressive neuroendocrine skin tumour that usually develops in sun-exposed areas [41]. Despite major advantages in understanding its carcinogenesis process, the MCC origin has not been completely understood: on one hand, Merkel cell precursor has been speculated as a source of MCC (normal Merkel Cells are terminally differentiated, thus allegedly not able to devolve into cancer); on the other hand, pro-B/pre-B lymphocytes, fibroblasts, dermal mesenchymal stem cells or aberrant keratinocytes have also been suspected as oncological precursors [42].

Elderly subjects with a story of sun exposition and/or immunosuppressant therapy are at higher risk for this disease [43, 44]. Local and regional lymph node metastasis can occur at the early stages, and the long-term prognosis is poor, dropping to 14% at five years from the diagnosis in the event of metastatic disease [11, 45]. Whether possible, first-line therapy encompasses a wide tumour excision, along with sentinel lymph node dissection and/or lymphadenectomy. High-dose adjuvant radiotherapy (50–60 Gray) to the primary site is usually delivered during the post-operative takeover, but this treatment also seems beneficial for locoregional control in patients with unresectable tumours [45, 46]. The historically favoured systemic therapy consisted of carboplatin/cisplatin-etoposide chemotherapy, while any second lines were based on anthracyclines, cyclophosphamide, vincristine, bleomycin, and 5-fluorouracil regimens [47]. Although cytotoxic chemotherapy has an objective response rate of >50%, responses are rarely durable with a median PFS of only three months [44]. Moreover, several safety concerns are well-known, making the traditional regimens mostly reserved for palliative care, i.e. patients with advanced or locally recurrent MCC [48]. During the last few years, thanks to the advancement in immunotherapy, different molecules have been proposed: immune-checkpoint inhibitors (PD-1 and PDL-1 ligand inhibitors) are the favoured agents. In particular, first-line therapy with avelumab showed a PFS in up to 72% of the cases in an international multicentre phase II trial (JAVELIN Merkel 200) [49]. Despite the success of immune therapy, a considerable proportion of patients with metastatic MCC do not respond to PD-1/PD-L1 blockade [49]. In this setting, TKIs have been proposed due to their capability of targeting angiogenesis and tumour-spreading processes as MCC expresses VEGF-A, VEGF-C, VEGF-R2, and platelet-derived growth factor (PDGF)-b in 91, 75, 88 and 72%, respectively [50, 51]. Despite these promising data, very little literature regarding anti-VEGF therapy on MCC is available: Tarabadkar described five cases of successful use of anti-VEGF TKIs such as pazopanib and CBZ in patients with advanced MCC [52]. In particular, a single case report of a 3.5-year PFS in a patient receiving CBZ after platinum-based chemotherapy has been described. Nonetheless, Rabinowits et al. failed to demonstrate the efficacy of single-agent CBZ and the 8 patients-trial prematurely failed, because of the lack of efficacy and toxicity [53]. However, all the included patients have been already pre-treated with heavy chemotherapy regimens - such as platin-etoposide - and received CBZ as a second or further line, which could have affected the safety and the tolerance to this drug. After all, no trial on CBZ in MCC is currently ongoing, since the phase II study on CBZ in recurrent MCC (NCT02036476), failed for lack of recruital.

Pheochromocytomas and paragangliomas (PPGLs) represent a rare group of neoplasia issued from chromaffin cells. In particular, pheochromocytomas arise from the adrenal medulla, while extra-adrenal lesions develop from the extra-adrenal paraganglia.

Surgical removal is the therapy of choice in PPGLs with either a curative intent in non-metastatic disease or, in metastatic disease, to minimise symptoms derived from catecholamine excess and from compression of the surrounding structures [54]. Surgical-related morbidity, however, needs to be carefully evaluated in particular in head and neck non-functioning PPGL. In fact, the closeness to cranial nerves and vases often induces to favour a non-surgical approach, such as watchful waiting or radiosurgery (gamma-knife/cyberknife), in the event of compression of the surrounding structures [55]. Aside from radiotherapeutic options, the only officially approved treatment for advanced tumours is the high-specific iodine-131-meta-iodobenzylguanidine [(131 I)MIBG] therapy (Ultratrace), while chemotherapy and different targeted therapies are widely used outside controlled clinical trials [56‐58]. Recently, TKIs have been in the spotlight, due to their ability to inhibit tumour neoangiogenesis. Furthermore, they also seemed to play a role in the bone microenvironment, by interfering with bone turnover, thus reducing eventual pain [59].

So far, the best studied TKI is sunitinib, which seems to have better efficacy in RET and SDHx mutated patients: in a large prospective phase II multicentre study, the total disease control rate (including stable disease or partial response) in patients with germline mutations (SDHA, SDHB, RET) was 83% (95% CI: 61–95%) [60]. At the moment, sunitinib is under investigation in the first randomised placebo-controlled phase II clinical trial in advanced PPGLs (FIRSTMAPPP, NCT01371202). Despite the promising data on sunitinib, other TKI such as CBZ have not yet been completely studied. In fact, in the literature, there is a case report of a 32-year-old male patient suffering from an advanced metastatic paraganglioma, showing a progressive disease when enrolled in a CBZ-based trial as a monotherapy treatment. Nonetheless, he showed prolonged disease control with a significant tumour reduction and decrease in metastatic burden when enrolled in the study by Apolo (started on CBZ plus nivolumab) [61, 62]. A phase II clinical trial on CBZ in monotherapy is now recruiting patients with unresectable metastatic paraganglioma or pheochromocytoma (Table 3 - https://​clinicaltrials.​gov). Preliminary results on patients treated with 60 mg daily (titrated down to 40 or 20 mg depending on the tolerability) showed tumour mass shrinkage and stability of the disease [63]. Moreover, a phase II trial on the combination therapy (CBZ plus atezolizumab) in patients suffering from advanced and progressive neoplasms of the endocrine system is ongoing, although not actively recruiting (Table 3 - https://​clinicaltrials.​gov).

Adrenocortical carcinoma (ACC) is a rare tumour arising from the adrenocortical gland, usually producing steroid hormones [64]. The prognosis is often poor since patients at low risk of recurrence are the minority and they can also frequently further develop recurrent disease [65]. In localised stages, complete surgical resection is the treatment of choice, while mitotane is the only approved drug that showed efficacy in prolonging recurrence-free survival when administrated as adjuvant treatment in patients at higher risk of recurrence [66]. Other adjuvant therapies, including chemotherapy, had been under investigation: although the European Society of Endocrinology did not reach a definitive consensus on adjuvant use of cytotoxic drugs for ACC, based on phase III trial, the combination of etoposide, doxorubicin and cisplatin (EDP) with mitotane (EDP-M) is recommended in patients with very high risk of recurrence [67]. Although the rates of response and PFS were significantly better with EDP-M treated patients, the study failed to show a significant difference in terms of overall survival [67]. In this setting, TKIs have been considered as a second-line therapy for patients who showed disease progression under EDP-M [68]. The rationale for the use of TKIs is based on the study of Pnan et al. who demonstrated an increased phosphorylation of c-MET in the immunohistochemistry of ACC tissue samples compared to adrenocortical adenomas and normal cortex, thus speculating that targeting c-MET could provide a breakthrough in the management of these aggressive diseases [69]. Moreover, an in vivo experiment in a mouse model confirmed that the knockout of c-MET inhibited adrenocortical carcinoma growth, tumour-related angiogenesis, chemotherapy resistance and cell survival, further suggesting this molecular pathway as a valuable therapeutic target for adrenocortical carcinoma [69]. Based on those promising preclinical data, a retrospective cohort study on CBZ use in patients with progressive disease under mitotane or cytotoxic chemotherapy regimens has been performed: best response was partial response in three out of 16 subjects, stable disease in five and progressive disease in eight patients. Although eight subjects (50% of the patients) showed tumour progression, this result was considered favourable (PFS was 16 weeks and OS was 58 weeks), compared to other treatments that failed to demonstrate a prolonged overall survival [68, 70]. A single institution experience showed a stable disease lasting six months in one patient treated with CBZ after previously progressing through chemotherapy and pembrolizumab [71].

To date, there is a phase II active trial actively recruiting patients suffering from unresectable or metastatic ACC in order to assess the efficacy and safety of CBZ (Table 3- www.​ClinicalTrials.​gov).

Neuroendocrine neoplasms (NENs) are a heterogeneous group of neoplasms originating from neuroendocrine cells localised in different organs, mostly within the digestive tract and in the lungs [72]. Therapeutic options rely on the tumour’s primary site, morphology, grade and stage, encompassing the “watchful and wait” attitude, surgical resection, locoregional treatment, or systemic therapies, including somatostatin receptor analogues, targeted therapies or traditional chemotherapy, in the event of advanced and progressive disease not responding to other treatments [73‐75]. As regards multi-TKIs, sunitinib is the only one that obtained EMA and FDA approval for pancreatic NENs, while other TKIs are currently under evaluation [76]. In particular, Reuther and colleagues found out that, while c-MET inhibition alone is not sufficient to exert direct antitumor or antimigratory effects in NEN cells, the multi-TKIs CBZ, as well as tivantinib, disclose therapeutic efficacy, probably due to the suppression of other molecular pathways [77]. Preclinical studies showed a reduced invasion and metastatic activity in mice with pancreatic NEN treated with CBZ: in particular, the simultaneous inhibition of VEGFR and MET pathways resulted in a smaller tumour mass and no metastases for CBZ-treated mice, compared to only anti-VEGF-treated mice [78]. The multi-TKI CBZ has also shown promising antitumor activity in a preliminary report of a phase II trial for pancreatic and extra-pancreatic NENs, showing a PFS of 21.8 months and 31.4 months in patients with pancreatic and non-pancreatic NENs, respectively [79]. Based on those results, five clinical trials are actively recruiting while three more trials are still active but not recruiting (Table 3 - www.​ClinicalTrials.​gov) [80].