Intermittent high dose proton pump inhibitor enhances the antitumor effects of chemotherapy in metastatic breast cancer

Extracellular acidity is a hallmark of malignant tumors with a pivotal role in the invasion, metastasis, drug resistance and selection of more aggressive cell phenotypes armed to survive in very hostile microenvironmental condition [1‐5]. The prime cause of tumor acidity is the fermentation of the sugars that invariably induce production of lactate, also called Warburg Effect [6] (Additional file 1: Figure S1). In addition, some proton-coupled monocarboxylate transporter (MCT) subtypes of the SLC16A gene family which rely on CD147, a chaperone to some MCTs, for expression contribute to regulate tumor acidity via the monocarboxylate (such as lactate) symport [7]. One of the most important mechanism allowing tumor cells to survive in acidic condition are some proton exchangers, including the vacuolar ATPase (V-ATPase). V-ATPase on one hand pump H+ from the cytosol within the internal vesicles, on the other hand eliminate H+ outside the cells, thus contributing to extracellular acidification (pHe), and to the citosolic alkalinization (pHi) as well; a pathway of reverse pH gradient that again is a peculiar characteristic of malignant tumors [8‐11]. Tumor cells use proton pumps to avoid a cascade of lytic enzymes activation that inevitably may lead to self-digestion [9, 12‐14]. Pre-clinical evidence demonstrated that an antiacidic approach based either on buffering molecules or inhibitors of proton exchangers may increase the efficacy of chemotherapeutic agents [1‐4, 6, 9, 15]. However, the same approach may also induce a direct antitumor effect by inhibiting H⁺ clearance from tumor cells, leading to tumor cell death [4, 16‐18]. Some further evidence supports also a role of tumor acidity as a tumor escape mechanism. In fact, proton pump inhibition increases the effect of immunotherapy and the spontaneous anti-tumor immune response [19]. Although these are features described for a wide spectrum of cancers, extracellular acidity was proven to have a key role in resistance of breast cancer cells to chemotherapeutics, through an impairment of drug uptake [20, 21]. In fact, both bicarbonate pretreatment and overexpression of Homo sapiens longevity assurance homolog 2 of yeast LAG1 (LASS2) which interacts with VPL (proteolipid subunit of vacuolar H+ ATPase) increases antitumor activity of either doxorubicin or mitoxantrone, in a mouse breast tumor model [22, 23]. More recent evidence suggests that V-ATPase inhibition induces cell death and growth inhibition in trastuzumab-resistant breast cancer cells, through a HER-2-mediated pathway [24]. Knock-down of the V-ATPase subunit c can also increase sensitivity of breast cancer cells to various cytotoxic agents, including cisplatin [15, 25]. Inhibition of proton pumps may re-establish a pH-gradient more suitable for both the uptake and the retention of various antitumor drugs with cancer cells, as it has been shown for the V-ATPase inhibitor bafilomycin, in the case of cisplatin as well [26, 27]. However, a direct V-ATPase inhibition through bafilomycin showed high level of systemic toxicity, due to the ubiquitous distribution of these enzymes in normal tissues [28]. As a consequence, the use of V-ATPase inhibitors, such as bafilomycin, as either potential anti-tumor drugs or simply chemosensitizers was abandoned due to their high level of systemic toxicity and did never get to phase I clinical trials. However, a class of proton pump inhibitors (PPIs) including esomeprazole (ESOM), omeprazole, lansoprazole, pantoprazole and rabeprazole, currently used in the treatment of peptic diseases, have been shown to represent a model of drugs with a high potential in the future anti-cancer strategies [14]. While they have been originally described as specific blockers of gastric H⁺-K⁺ ATPases, they can also inhibit V-ATPase activity [4]. PPI have been used by billions of people worldwide in the last decades, without significant side effects [29], even at high dosages (as in patients with Zollinger-Ellison syndrome) [30, 31]. Interestingly, the absence of toxicity for this class of drug is largely due to their dependence on an acidic pH for activation, inasmuch as PPI are administered as prodrugs, either orally or systemically, needing low pH in order to be transformed into the active compounds tetracyclic sulfonamide [32]. Thus, for PPI, differently to the vast majority of the drugs including anticancer drugs, protonation in an acidic environment leads to activation instead of neutralization. As lipophilic and weakly basic prodrugs, they easily penetrate cell membranes and concentrate in acidic compartments, where they are unstable and are converted into sulfonamide forms, which are the active inhibitors [33]. Based on these properties, PPIs have been extensively investigated for their potential to reduce tumor acidity and overcome the acid related chemoresistance. Furthermore, PPIs could have direct tumor cell toxicity by depriving them of a key mechanism for surviving in their aberrant acidic condition. A number of studies have now shown that PPIs can be useful in modulating tumor acidification and restoring chemotherapeutic sensitivity in drug-resistant cancer cells in in vitro and in vivo preclinical studies [15, 34‐36]. These preclinical data have been supported by clinical studies in both patients with osteosarcoma [37] and in companion animals with spontaneous tumors [38, 39]. In addition, specific cytotoxic effects of PPIs on tumor cells have been reported, including B cell lymphoma [40], melanoma [18], pancreatic cancer [36], esophageal cancer [41], gastric carcinoma [42], Ewing sarcoma [43], osteosarcoma, rhabdomyosarcoma and chondrosarcoma [37, 44]. As expected, the PPI induced cytotoxicity is strongly enhance in low pH culture conditions [18]. The effects of PPI were shown in models of breast cancer as well at various levels, including growth, invasion and metastasis [45‐47]. Moreover, long lasting treatment with PPI in patients with Barrett’s oesophagus significantly reduces the risk of oesophageal adenocarcinoma and/or high grade dysplasia [48, 49].

With this background we set up a pilot, prospective, randomized, phase II clinical study (NCT01069081) with the purpose to investigate whether the PPI ESOM might improve the efficacy of chemotherapy in patients with metastatic breast cancer (MBC). Cisplatin is moderately active as first-line treatments in MBC, with an objective response rate (ORR) 40 to 60 % [50]. Cisplatin-based regimens are also widely tested in the first-line metastatic setting with median time to progression (TTP) of 7.2 to 11.7 months and acceptable toxicities [51‐56]. Based on these data and sufficient experience of our center [56‐58], the doublet TP regimen (docetaxel and cisplatin combination) was an acceptable option for the control arm and suitable for this proof of concept setting. In the setup of the treatment combining the TP regimen with ESOM we use 3 key criteria: (i) two fixed doses of 160 and 200 mg/day were established based on the pre-clinical evidence of ESOM [18]; (ii) pretreatment with ESOM, based on the pre-clinical evidence that only pre-treatment showed to increase chemosensitization [15]; and (iii) the intermittent schedule, based both on the rationale that an acidic pH is needed for a full PPI activation, in order to be transformed into the active molecule (sulfonamide) and on the in vivo evidence showing that tumor pHe is PPI-dependent, showing displayed an initial shift towards neutrality after ESOM treatment, returning to acidic values within 48 h after the stop of the treatment [18]; we thus used a weekly treatment schedule of consecutive 3 days of a full ESOM dosage, followed by 4 days ESOM off immediately followed by TP regimen-based chemotherapy.

This pilot prospective, randomized, phase II trial evaluated the efficacy of ESOM, in association with TP regimen, in the first-line treatment of MBC patients. To our knowledge, this is the first clinical study to test the hypothesis that manipulation of pH gradient in tumor microenvironment can enhance antitumor effects of chemotherapy in MBC patients. Our study preliminarily showed a significant difference in both ORR and TTP between patients co-treated with ESOM and those who received exclusively TP regimen. Actually, this result was achieved not only through a simple addition of ESOM to the treatment schedule during TP regimen, but through a continuation of the PPI treatment for up to 66 weeks, until disease progression, death, withdrawal of informed consent, or unacceptable toxicity. This, of course may suggest that the clinical might be due to either or both the PPI-induced chemosensitization [15, 37‐39] and/or an additional effect due to the direct anti-tumor activity of PPI [18, 40]. Independently from any interpretation of the data, we have shown for the first time a potential synergistic effect of repeating intermittent high dose of PPI with a standard chemotherapeutic regimen based on cisplatin and docetaxel in patients with MBC, which is worthy of being validated in future phase III trial. Although with a comparable median TTP, the dose level with 80 mg bid showed a relatively higher ORR (primary endpoint) and median OS than the dose level with 100 mg bid. The relatively longer OS in lower dose arm can be partly attributed to higher proportion of subsequent systemic treatments (Additional file 3: Table S2). Taking into consideration the cost, risk and benefit, the dose level with 80 mg bid is recommended for phase III testing. Recently, significantly stronger immunofluorescence expression of H+/K + -ATPase proton pump was observed with the triple negative MDA-MB-468 cells compared to control MCF-10A cells, which was confirmed with Western blotting. Differences in sensitivity to ESOM were also detected in these two cell lines, with MDA-MB-468 cells found to be significantly more sensitive [46]. In our study, the effects of TTP were even more significant when TNBC patients were analyzed separately, indicating that ESOM has potentially improved the prognosis of TNBC patients which was consistent with the preclinical research [46], although in a small sample size (15 patients) (Fig. 3c).

Development of resistance to cisplatin is a major obstacle in the clinical treatment of some solid tumors, including TNBC [63] and ovarian cancer [64]. Cisplatin resistance is thought to involve several mechanisms, such as increased drug efflux and cellular thiols [65, 66] and increased DNA-repair activity [67, 68]. We have recently shown that particularly cisplatin resistance of human malignant tumors may be the result of both tumor acidity [15, 27] and the release of nanovesicles called exosome [69]. In turn, also exosome release from cancer cells is highly increased by environmental acidity and proton pump inhibitors or buffering procedure dramatically inhibit exosomes production by cancer cells [70]. Notably, exosome plasmatic levels directly correlate with the tumor size or mass [71], and PPI treatment markedly reduces the plasmatic levels of exosome released by human tumors in xenograft models [69]. PPI, together with inducing chemosensitization in both pre-clinical settings [15, 69] and clinical investigations in both humans [37] and domestic animals with spontaneous tumors [38, 39] can induce apoptosis in various tumor cell types and tumors [18, 40, 42], including breast cancer cells in which PPI inhibit both growth [46] and invasion and metastasis [47].

Actually, our study is the first providing the preliminary clinical proof of concept that PPI may induce chemosensitization but also contributing to control tumor progression in breast cancer patients. Recently, the effectiveness of PPI in improving chemotherapy was shown in a clinical study on osteosarcoma patients [37], thus supporting the results of the present study. Notably, PPI treatment was continued after the stop of chemotherapy without any evidence of additional or specific toxicity, suggesting that PPI may well be used in a chronic treatment of breast cancer patients with the aim to either prevent disease relapses or controlling the disease progression.

In conclusion, our trial is the first study that prospectively evaluates the potential use of PPI in treatment of breast cancer patients. The preliminary results provide the evidence that intermittent high dose PPI enhances the antitumor effects of chemotherapy in MBC patients without evidence of additional toxicity, in term of both ORR and TTP. Taking into consideration the cost, risk and benefit, the dose level with 80 mg bid is recommended for phase III testing. Moreover, the PPI treatment was proven particularly efficient in prolonging the TTP in the subgroup of TNBC, which unfortunately have currently very poor treatment option. The results of this pilot clinical trial support the setup of a larger-sample, multicenter, randomized, phase III clinical trial, in order to provide a definitive validation for the use of PPI in future strategy against breast cancer, and hopefully other poorly treatable cancer.