Savara: The Binary Bet On Phase 3 Data (NASDAQ:SVRA) (2024)

Introduction

Savara Inc. (NASDAQ:SVRA) is a small clinical-stage biotech company focusing on the treatment of a single rare respiratory disease - autoimmune pulmonary alveolar proteinosis (aPAP). In April 2017, privately held Savara merged with the publicly traded Mast Therapeutics, with Savara holding 77% ownership in the new entity and Mast owning 23%. Post-merger, the company was renamed Savara Inc. and managed three drugs in its pipeline: Molgramostim, AeroVanc, and Aironite. In March 2018, Savara announced that Aironite failed to meet the primary endpoint in the Phase 2 INDIE study, and in December 2020, AeroVanc did not meet its primary endpoint in the Phase 3 AVAIL trial, leading to the discontinuation of both developments. In June 2019, the company released topline data from the IMPALA trial, which investigated Molgramostim in aPAP, causing the stock price to plummet from $10.57 to $2.62 per share. Although the trial missed its primary endpoint, it demonstrated significant potential through secondary endpoints, prompting the announcement of a follow-up trial, IMPALA-2, with a revised study design. The topline data from IMPALA-2 trial are expected by the end of Q2 2024, which will either de-risk the development and potentially boost the stock price or lead to market disappointment and a further decline in valuation.

Investment thesis

Savara, with a single drug in its current pipeline, presents a high-stakes situation due to its binary catalyst nature. Nebulized GM-CSFs, similar to Molgramostim, are already used in clinical practice for patients with aPAP, though their use is constrained by the drug’s availability and cost. The company’s valuation is heavily influenced by the results from both primary and secondary endpoints, which combined describe a comprehensive efficacy profile of the drug. However, as is common with clinical-stage biotech companies, the initial price action is strongly influenced by the statistical significance of the primary endpoint.

We believe that Molgramostim’s efficacy will be demonstrated in the upcoming trial readout, based on previous research and the scientific rationale behind the drug’s mechanism of action (MOA) in aPAP. Alveolar impairment in aPAP patients is believed to be mainly driven by anti-GM-CSF antibodies, thus administration of the cytokine should alleviate the symptoms. Furthermore, our in-house systematic review of clinical literature involving use of GM-CSF in aPAP cohorts suggested high heterogeneity and consistent level of efficacy.

Under our assumptions, the stock is undervalued, and following the topline data readout, we believe the market will recognize the discrepancy between the valuation and the probability of success (POS) of Molgramostim, due to the de-risked profile. However, the previous trial, IMPALA, has several shortcomings that might impact IMPALA-2 reported outcomes, which will be discussed further below. Moreover, the IMPALA-2 study design will be analyzed to assess the choice of inclusion/exclusion criteria and primary/secondary endpoints, and to explore the likely scenarios. The market potential of Molgramostim will also be discussed, as there is considerable uncertainty regarding the prevalence estimates and which patients would benefit from the drug in clinical practice.

Pulmonary Alveolar Proteinosis

First, it is important to understand clinical and physiologic features of the disease, as well as the epidemiological dynamics of the patient population.

Pulmonary alveolar proteinosis (PAP) is a rare pulmonary disorder characterized by compromised alveolar macrophage functionality, leading to an excessive accumulation of surfactant proteins within the alveoli. Disease progression may lead to secondary infections, fibrosis, and respiratory failure; however, literature indicates that spontaneous recovery occurs in 8% to 25% of cases.

The etiology of PAP may be linked to mutations in the granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor genes, secondary to inhalation of toxic substances (e.g., silicone), or stem from an autoimmune process where anti-GM-CSF antibodies disrupt alveolar macrophage differentiation and phagocytosis. Autoimmune PAP represents the most common form, accounting for approximately 90% of cases. The incidence of aPAP is estimated at 6 per million, with suggestions that this figure is an underestimate due to historical deficiencies in anti-GM-CSF testing, and independent research of Savara suggest true prevalence 4 times greater. Nonetheless, current consensus estimates indicate there are about 2000, 2300, and 750 cases of aPAP in the USA, EU, and Japan, respectively.

Although its efficacy has not been evaluated in a controlled trial, whole lung lavage (WLL) remains the primary treatment for aPAP. This invasive procedure necessitates general anesthesia and entails the physical removal of excess surfactant via the administration and subsequent drainage of up to 50 liters of saline (usually less). WLL is invasive and can lead to significant complications, such as pneumonia, hypoxemia, fever, pleural effusion, pneumothorax and in severe cases death. While WLL is primarily symptomatic and does not modify the disease course, some reports indicate a 70% success rate in preventing recurrence over a seven-year follow-up period. However, other studies suggest higher relapse rates, with 56% of patients experiencing recurrence within an average of 16.9 months. Despite these limitations, WLL continues to be the standard treatment due to its effectiveness in improving lung function in the short term, particularly recommended for severe cases (DSS-3 or higher).

The logistical challenges and safety concerns associated with WLL, coupled with the occurrence of non-responsive cases, have prompted the exploration of alternative treatments. Current therapeutic strategies reported in the literature include the use of statins, plasmapheresis, secretolytic agents (e.g., Ambroxol), Rituximab, subcutaneous or inhaled GM-CSF (Sargramostim, Leukine), and inhaled GM-CSF (Molgramostim).

Granulocyte-Macrophage Colony Stimulating Factor

Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) is a monomeric glycoprotein known to function as a cytokine, promoting white blood cell growth. In addition to stimulating proliferation of hematopoietic stem cells, the cytokine induces differentiation, activation, and survival of affected cells. In lung GM-CSF is produced by alveolar epithelial cells type II (AEC-II) and binds to receptor on alveolar macrophages (AM). This interaction promotes downstream changes in the cell metabolism, notably in the lipid homeostasis, and induces efflux of cholesterol. In aPAP, elevated anti-GM-CSF antibodies interfere with this pathway, leading to cholesterol accumulation in the AM, inhibiting their functionality and resulting in accumulation of surfactant protein. Excessive accumulation impairs the ability of the lung to diffuse oxygen through the alveolar epithelium, and therefore results in the symptoms described above.

The recognition of elevated anti-GM-CSF antibodies as a hallmark of aPAP, along with previous animal models implicating the peptide in disease pathology, has led to the adoption of inhaled and subcutaneous GM-CSF as therapeutic strategies. Initially, GM-CSF was employed to enhance bone marrow recovery in acute myeloid leukemia patients’ post-chemotherapy, reducing infection rates and mortality. Subsequent case reports have underscored the potential of GM-CSF in treating aPAP, although many claimed concerns regarding the absence of large-scale trials, the high costs associated with the inhaled form, difficulties in accessing the drug for off-label use, and the non-standardization of dosing protocols.

Recombinant human GM-CSF historically used in clinical practice involve Sargramostim and Molgramostim. Sargramostim is manufactured in yeast and is glycosylated, while Molgramostim is grown in E.coli cultures and is not glycosylated. According to Dr. Trapnell (PI of IMPALA) himself there is little evidence that would suggest that either is superior (1:13:27), which is further supported by literature data. Sargramostim has been used intravenously, but inhaled therapies were deemed more promising, due to better accumulation in the desired site. The nebulizers used to administer Sargramostim and Molgramostim use three mechanisms: jet, ultrasonic and mesh. It has been determined that the mesh system, chosen by Savara, imposes the least strain on the peptide during administration, and therefore could result in the best bioavailability of the active compound in the lung.

IMPALA trial

When Savara announced the topline data of the IMPALA trial, the stock price plummeted. This drop was likely driven by the study miss on statistical significance on the primary endpoint, the alveolar-arterial difference in oxygen concentration (A-aDO2) and clinically relevant secondary endpoint, 6-minute walk test (6MWT). The sponsor noted that four subjects had invalid data due to the administration of supplementary oxygen during the A-aDO2 test. Analysis of the MITT population, which excluded these four subjects, showed that the primary endpoint data was statistically significant at p=0.03. The secondary endpoints yielded mixed results - some surrogate lung function endpoints such as the diffusing capacity of the lung for carbon monoxide (DLCO), and patient self-reported questionnaires such as St. George’s Respiratory Questionnaire (SGRQ) were statistically significant, but 6MWT was not. Further market reaction occurred when the company announced via press release that the FDA advised against submitting a BLA due to insufficient data, leading to another decline in stock value. Nonetheless, the market regained some optimism in December 2019 when Savara announced that the FDA had granted them a breakthrough therapy designation, which led to the stock price doubling quickly, though it has traded sideways since then. Below, we will consider the data from the IMPALA trial to assess the efficacy of Molgramostim and evaluate the likelihood of this data being replicated in IMPALA-2.

Surrogate endpoints of respiratory function, such as A-aDO2 and DLCO, are commonly used in clinical contexts for diagnostic purposes and provide useful insights into lung functionality. However, by the FDA these are considered inferior to clinically relevant endpoints like the 6MWT. This is largely because, while surrogate measures can indicate improvement, the meaningfulness of these effects for patients is often questionable.

The primary endpoint, A-aDO2, exhibited improvements in both the placebo and treatment arms, suggesting that even patients receiving sham inhalations can experience enhancements in lung function. This could be attributed to several factors: patients could regress to the mean, WLL was permitted throughout the treatment, and some patients might spontaneously recover. DLCO, an exploratory endpoint somewhat analogous to A-aDO2, showed significant improvements over the placebo. SGRQ, another exploratory measure, also heavily favored the treatment group. The clinical measure of interest, the 6MWT, while displaying numerically superior levels, failed to reach statistical significance due primarily to a high degree of variance (some patients showed tremendous improvements, while others showed no improvement or worsening). These outcomes are depicted in figure 2 of Trapnell et al.; however, we believe that figure S11 from the supplementary material provides more insight as it includes data on both continuous and intermittent administration.

Interestingly, both the intermittent and continuous treatments exhibit similar trends in A-aDO2 and SGRQ, yet the intermittent arm shows little to no improvement in the 6MWT and slight inferiority compared to continuous regimen in DLCO. This inconsistency is concerning, as it makes the reproducibility of any relevant clinical improvements appear less convincing. It raises the question: why would patients in the intermittent group experience similar improvements in surrogate measures but almost no improvement in the most clinically relevant test, both to patients and the FDA? These discrepancies might be partially due to differences in baseline statistics of the study groups.

From the supplementary material Table S2, it is evident that both the placebo and intermittent arms had a higher number of current smokers and a greater baseline distance walked on the 6MWT. DLCO, a measure known to be confounded by smoking status, should have been adjusted for smoking baseline status in the statistical analysis, yet it was not. The adjustments were made for trial arm, baseline WLL status and geographic region. The observed DLCO score in the continuous group may appear superior due to these unadjusted confounders, and, in reality, it might be closer to the score observed in the intermittent group. Furthermore, the baseline difference in 6MWT distance between the continuous group and the placebo is similar to the observed estimated treatment difference (~35m baseline difference, ~25m estimated treatment difference). If one were to plot the improvements in distance walked from baseline and not as a change, the figure would appear much less convincing (example below, numbers are not exact, the figure serves to approximate a trend rather than directly describe the data from the study). When plotted this way, the values instead of diverging and showcasing a drug's efficacy, converge, possibly due to the effect of the intervention, or other factors such as regression to mean.

It is also important to recognize that the continuous treatment group had the largest fraction of the most severely symptomatic patients, which could be interpreted either as a benefit (superior treatment effect over placebo in more severely ill patients) or as a flaw (higher likelihood of regression to the mean). It could be argued that regression to the mean and spontaneous recoveries are unlikely in respiratory surrogate measures, given that the screening period was based on the stability of symptoms.

The significance of improvements in surrogate measures can also be evaluated from the perspective of the minimally clinically important difference (MCID). While aPAP is a rare disease and lacks official guidelines or defined improvements (to our knowledge), other respiratory conditions like chronic obstructive pulmonary disease (COPD) have well-defined MCIDs, and these can be partially analogous to aPAP. In COPD, it is suggested that an MCID of 11% in DLCO differentiates levels of pathology severity. Trapnell et al. reported an estimated treatment difference of 7.8%, which falls below the MCID threshold for COPD, but is arguably close to this value (with a crude gain in the treatment arm of 11.5% over baseline) and if a treatment period longer than 24 weeks would result in continued improvements, they would likely exceed 11% (based on follow-up trends).

Notably, DLCO can be confounded by hemoglobin levels but can be appropriately adjusted. The authors adjusted the data, and, in fact, the supplementary material provides data on changes in hemoglobin levels in treated subjects.

The authors suggest that the observed changes might be linked to the drug’s efficacy, noting that in a hypoxic state, the body produces an excess of red blood cells (RBCs).

The safety analysis reported no major differences in adverse effects; however, it is important to mention that patients in the continuous administration group experienced chest pain at significantly higher rates (~20%). The relative tolerability of the treatment can be inferred from the high compliance observed during the study period, indicating that patients generally tolerate the inhaled drug well.

Overall, IMPALA trial revealed that the promising preclinical research and case study reports of efficacy of inhaled GM-CSF were not easily shown in a properly conducted RCT. There are many possible mechanisms that could limit the drug’s functionality; while it has been determined that the mesh nebulizer variant induces the least stress on GM-CSF during administration, it might still limit the bioactivity. Another issue might be the alveolar surfactant microenvironment, which is naturally rich in proteases that can expose the active compound to chemical stress and catalyzed cleavage. The concentration of the drug that reaches the alveoli will also likely be much smaller than the concentration in the proximal parts of the respiratory tract, due to the primary innate defense of the lung - mucociliary clearance - constantly moving the mucus away from the distal portion of the lung. Lastly, in more severe disease stages associated with higher surfactant burden, the macrophages might be damaged beyond the point where the administration of the factor will rescue their functionality.

IMPALA-2

IMPALA 2 represents the sponsor’s effort to refine the design from the IMPALA trial, aiming to achieve meaningful data on key measures of interest. The topline readout is critical to the company’s future, as a failure to deliver positive results will likely lead to a rapid decrease in market capitalization, particularly due to the absence of other projects in the pipeline. We will analyze Savara’s study design, critically assess it, and speculate on whether the data can meet the expectations of investors and, more importantly, satisfy FDA requirements for approval.

The figure provided by Savara in their current corporate presentation contrasts the IMPALA and IMPALA-2 designs side by side. The primary change is the selection of DLCO as the primary endpoint, replacing A-aDO2, based on its robust improvement in the IMPALA trial and its more standardized and repeatable profile. The screening criteria remain similar, though the sponsor has adjusted the cut-off for change in the DLCO score to 15% from the previous 10% and chose to exclude subjects based on WLL received within 3 months prior to enrollment instead of just 1 month. SGRQ continues as a secondary endpoint but is now subdivided into total score and activity score. The time to/requirement for WLL has been completely removed, indicating a possible lack of confidence in this measure by the sponsor, and the 6MWT has been replaced with an Exercise treadmill test. Additionally, with the trial featuring only continuous molgramostim dosing, there will be just two arms, each with approximately 80 patients.

While the rationale to change the primary endpoint from A-aDO2 to DLCO may be supported by IMPALA data, it might not be as clinically relevant as portrayed. It is concerning that the sponsor did not designate a clinical measure such as the 6MWT or the Exercise treadmill test as the primary endpoint, as these are outcomes of significant interest to the FDA. This decision might be explained by the sponsor’s lack of confidence in achieving this endpoint, or possibly because, with the limited sample size associated with the rarity of the condition, achieving a statistically significant result on such an outcome would be challenging. This argument is further supported by the decision to report the SGRQ total score and activity score separately, likely intending to demonstrate significant patient-reported improvements in activity, regardless of the treadmill’s clinical functionality measure. However, these are at best educated guesses, and a deeper understanding of whether the primary endpoint is likely to be met - which is usually the headline of topline data press release - could be more pertinent.

Concerns regarding the reproducibility of DLCO were mentioned previously. The lack of consistency between intermittent and continuous responses on different respiratory surrogate measures, combined with differences in baseline statistics, suggests that the true performance of the treatment arm in DLCO might have been overestimated in the trial. Given the current lack of evidence that other rhGM-CSF treatments would be superior or inferior to molgramostim, we conducted an in-house systematic review of such treatments in aPAP patients.

It is crucial to note that while the analysis incorporated data from various rhGM-CSFs and dosing regimens, the low heterogeneity (I²=0%, p=0.58) provides moderate evidence suggesting strong similarities between the studies. The random effects model predicted a treatment difference of 6.33 (p<0.01), which slightly exceeds the 5.7% change threshold that has a 90% power to be detected in IMPALA-2, according to Savara’s corporate presentation. Given that most studies utilize intermittent dosing, this effect size might underestimate the data observed in IMPALA-2, assuming that continuous dosing correlates with a greater treatment effect.

While the SGRQ data was not assessed quantitatively, based on the IMPALA results and data from Tian et al., there appears to be a consistent effect on the patient subjective questionnaire in favor of the drug, suggesting that this measure is also probable to be met in IMPALA-2.

The functional activity measures appear to be the most challenging to achieve statistical significance on and, in our view, represent the greatest uncertainty regarding Savara’s success and short-term price action. It seems probable that the surrogate measures and patient questionnaire outcomes will achieve statistical significance; however, the clinically relevant measures might fail to demonstrate significant improvements. This scenario would likely lead to increased uncertainty for investors, as Savara has attempted to advocate for their drug based on IMPALA data, yet the FDA recommended another trial (FDA normally requires two phase 3 trials or one trial with univocal results; p<0.025 on relevant measures). This recommendation was likely due to concerns over data validity stemming from the exclusion of patients with incorrect A-aDO2 measurements and failures on endpoints such as the 6MWT. Nonetheless, if the sponsor can meet the primary endpoint and some secondary endpoints, they will be able to present a compelling case for approval, which could stand firm given the orphan disease status and the current off-label use of GM-CSF in clinical practice. In other words, if patient reported subjective outcomes and surrogate endpoints were to showcase efficacy, considering the favorable safety data, the treatment would seem to present a positive benefit-risk profile. FDA approvals for respiratory diseases based on primary surrogate endpoints happened in the past, for example, Trikafta was approved in 2019 for cystic fibrosis based on improvements in (forced vital capacity) FVC measure. Notably, Trikafta trial data was highly convincing, with clear improvements over placebo on major secondary and primary endpoints (p<0.001).

Savara - Valuation

Savara’s valuation can be modeled based on industry standards and a combination of literature and data provided by the company, particularly concerning pricing and prevalence. We modeled future Molgramostim sales in the US, EU, and Japan to determine the drug’s value, which will likely closely match the company’s valuation under best-case-scenario, given that it is the sole program currently in development.

We modeled the entry of Molgramostim into the US, EU, and Japan markets simultaneously in 2026, with a BLA submission anticipated in the first half of 2025, per company guidance. The sales projection includes a typical five-year ramp-up period and a three-in-four penetration rate. The adherence rate was modeled at 85%, reflecting the industry average, while the Average Selling Price (ASP) was modeled at 75% of the Average Wholesale Price (AWP). The initial price in the US was set at $300,000, which is the lower bound of the range expected by the company, with a 2% annual price increase until the price was reduced to 50% of its value upon loss of exclusivity in a given region. Prices in the EU and Japan were set at half that of the US and were modeled to decrease by 2% annually.

The percent values were primarily adapted from research published in David et al. 2017. Cost of Goods Sold (COGS) was modeled at 10% of revenue, while R&D expenses were kept constant at $10 million per year to cover minor developmental costs associated with ongoing drug research post-market entry. Selling, General, and Administrative (SG&A) expenses were set at 34% of revenue. Assuming a discount rate of 10%, the Net Present Value (NPV) of Molgramostim was estimated to be approximately $1.2 billion, corresponding to about $10 per share, based on a share count of 138 million, a figure which includes cash.

This analysis suggests that the market’s perceived probability of success (POS) for Molgramostim is around 40%, significantly below the average for drugs in phase 3 clinical development, which typically stands at about 60%. This gap is likely attributable to initial feedback from the FDA on the IMPALA trial, as well as several issues related to the methodology and the questionability of the replicability of findings highlighted in this article. Additionally, the rarity of disease and the addressable patient population is also another risk that the market is likely pricing in. Despite that, based on the available literature and adjustments made in the IMPALA-2 trial design, it is reasonable to anticipate that the upcoming trial is more likely to achieve success on several outcome measures, which supports the view that Savara is currently undervalued.

If the stock were to trade at POS levels consistent with the industry average, the per-share price would be approximately $6, representing about a 50% upside from the current level of $4.09. Promising topline data could significantly de-risk the approval chances for Molgramostim, potentially driving the stock price towards the modeled value of $10 per share. Conversely, a mixed readout or trial failure could cause the stock price to plummet to or below the cash basis, which would be approximately $1-2 per share, given the $143 million in cash on hand and relatively small debt of ~$25 million. This scenario represents a substantial downside risk, and has already happened when IMPALA trial results were published, when the company lost ~75% of market capitalization (dropping from ~$400 million market capitalization to ~$100 million, cash and equivalents were ~$110 million in Q2 2019).

Despite the stock price being down over 50% compared to pre-IMPALA topline readout, due to a series of dilutions, on a market capitalization basis the company is currently worth almost 50% more (~$600 million in Q2 2024 compared to ~$400 million in Q2 2019). The total share count approximately tripled since 2019, going from an initial ~40 million to ~138 million, with ~45 million pre-founded outstanding warrants. It is also crucial to consider the valuation assuming pre-funded warrants were exercised and converted to common stock, representing a more conservative approach. At a total share count of ~182 million (figure currently reported in EPS calculations by Savara), our DCF model values the company at $7.50 per share.

Cash Burn

The company asserts that they are well capitalized into 2026, a claim that appears reasonable given the nearing completion of the IMPALA-2 clinical trials and their recent cash burn rates. In 2023, the cash burn amounted to $55 million, with a quarterly breakdown of $11 million in Q1 and Q2, $17 million in Q3, and $16 million in Q4. In 2019, when IMPALA data was first reported, the operating expenses equaled to $51 million, of which $39 million was related to R&D and $13 million to G&A. Between 2020 and 2022, the total expenses stood chronologically at $49 million, $41 million, and $38 million. The net burn for Q1 of 2024 escalated to $20 million, marking a 25% growth quarter-over-quarter. Despite this increase, it appears that the company maintains sufficient capital to complete the ongoing study and file the Biologics License Application (BLA). Extrapolating current cash burn thorough the next 5 quarters (end of H1 2025), the total would add up to $100 million, still about $40 million shy from current cash and cash equivalents.

However, should the company present strong data, we find it likely that some form of capital raising, likely through dilution, could occur to address the increased costs associated with commercialization. It is important to note that despite the upcoming topline data readout, trial expenses will continue to accrue, particularly due to the open-label phase of the trial, which is scheduled to last up to 144 weeks. This possibility is further supported by Savara's recent issuance of a mixed shelf offering for up to $400 million. Although we believe that the management did not mention anticipating such activities in the upcoming future, investors should be cautious.

Risks

To summarize, we identify three major risks associated with the stock's future performance. The primary concern is the upcoming IMPALA-2 trial data, which, to the contrary of our belief, could miss on the primary endpoint and key secondary endpoints. As outlined above, this would likely result in a rapid stock sell-off. Another reason for investors to stay wary is the level of dilution the stock has experienced and could possibly experience. Between 2019 and 2024 the number of shares outstanding grew from ~40 million to ~138 million. The pre-funded outstanding warrants combined with recent S3 filling, show that further dilution is not out of question, especially considering that if the trial were to succeed, the company would have to engage in a legislative process, and invest in a sales force. Even assuming study success, there still remains uncertainty regarding terminal market penetration of the drug and risk associated with use of other emerging treatments.

Conclusion

In conclusion, Savara’s Molgramostim presents a possible opportunity in the biotech sector. The science underpinning Molgramostim is sufficient, and while the data from trials like IMPALA and the anticipated IMPALA-2 show promising results, the clinical relevance continues to pose questions. Nevertheless, Molgramostim is already being utilized in clinical practice, underscoring its potential and acceptance within the medical community despite the uncertainties (based on the IMPALA cohort, incidence of previous use of GM-CSF therapies was ~15%).

Looking ahead, the primary catalyst for Savara’s stock will be the topline data readout from the IMPALA-2 trial. Based on the adjustments made to the study design and the initial efficacy observed, we expect statistically significant results on the primary endpoint. We believe this outcome is likely to significantly influence the stock price, reflecting the market’s response to positive trial results.

Considering the current valuation and the potential upside from successful trial outcomes, we view the Savara stock as a Buy, albeit not a particularly exciting one. While we anticipate that the surrogate endpoints are likely to be met, there remains significant uncertainty regarding the functional measures and how the FDA might interpret mixed results. Additionally, even with approval, introducing the drug to the orphan disease market presents notable challenges. Given the absence of other ongoing developments within the company, a buy-out might align best with management’s interests, suggesting a cautious investment approach with moderate expectations for a significant upside.

AR Capital Menagement