|
Brent B. Siler |
VIA EDGAR AND HAND DELIVERY |
|
(703) 456-8058 |
|
|
bsiler@cooley.com |
|
November 4, 2013
United States Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549
Attention: Jeffrey P. Riedler, Assistant Director
RE: Trevena, Inc.
Registration Statement on Form S-1
Filed October 9, 2013
Response Letter Dated October 31, 2013
File No.: 333-191643
Gentlemen and Ladies:
On behalf of our client, Trevena, Inc. (“Trevena” or the “Company”), we are responding to the comments (the “Comments”) of the staff (the “Staff”) of the Securities and Exchange Commission (the “Commission”) contained in its letter dated October 31, 2013 (the “Comment Letter”), relating to the above referenced Registration Statement (the “Registration Statement”).
We also describe below the changes the Company is proposing to make in response to the Comments in an Amendment No. 3 to the Registration Statement on Form S-1 (the “Amendment”) and the prospectus included therein (the “Prospectus”). These proposed changes are attached to this letter as Exhibit A (the “Exhibit”).
FORM S-1
Critical Accounting Policies and Significant Judgments and Estimates
Fair Market Value Estimates, page 62
We have reviewed your response to comment one issued on October 29, 2013 and have the following comment:
· Regarding the first bullet point, we do not believe that the backsolve method is appropriate since the Series C preferred stock transaction does not appear to be considered an arms’ length transaction. Please revise your methodology or tell us why you believe the fair values used are appropriate.
1299 PENNSYLVANIA AVENUE, NW, SUITE 700, WASHINGTON, DC 20004-2400 T: (202) 842-7800 F: (202) 842-7899 WWW.COOLEY.COM
Response to first bullet:
In response to the Staff’s Comments, the Company advises the Staff as follows: The American Institute of Certified Public Accountants practice aid entitled Valuation of Privately-Held-Company Equity Securities Issued as Compensation (the “Practice Guide”) provides that valuations made pursuant to the guide should be based on the FASB ASC 718 and 505-50 definition of fair value as “[t]he amount at which an asset . . . could be bought . . . or sold . . . in a current transaction between willing parties, that is, other than in a forced or liquidation sale.” Although the definition refers to a hypothetical transaction, an actual transaction that otherwise meets the terms of the definition would evidence fair value. Because both Forest Labs and the Company were willing and informed parties in the Series C financing, and neither was forced to transact, the amount at which Forest Labs purchased an interest in the Company reflects the fair value of the Company on that date. Although the Company also transferred an option to Forest Labs concurrent with the stock purchase, the value of the option was negligible, so the inclusion of the option in the transaction does not alter the fair value assigned to the Company’s equity, nor does the option agreement render Forest Labs a related party with respect to the Series C investment. Finally, the participation of the Company’s existing investors in the Series C financing on the same terms as Forest Labs does not change that these terms were set by competitive negotiations between willing but uncompelled parties, i.e., Forest Labs and the Company. For all these reasons, described in greater detail below, the Series C preferred stock transaction revealed the fair value of the Company at the time of the transaction and the Company’s use of the backsolve method utilizing this fair value is both appropriate and consistent with the Practice Guide.
The May 3, 2013 transaction between the Company and Forest Labs was between willing parties neither of whom was forced to enter into the transaction
On May 3, 2013, in return for $30 million, the Company sold Forest Labs 18.4 million preferred shares and entered into an option agreement to license TRV027 upon the payment by Forest of a $65 million option fee as well as the license agreement that would become effective on the exercise of the option (all three together the “May Transaction”). At the same time, the Company’s existing investors also purchased $30 million of preferred stock pro rata based on their prior ownership in the Company and at the same price that Forest Labs paid per share. The May Transaction was the culmination of almost six months of discussion and negotiation between Forest Labs and the Company.
The Company first approached Forest Labs about a combined equity investment and option agreement in November 2012. In this same timeframe, the Company also received competing term sheets from two groups of existing investors in the Company, each proposing an equity investment in the Company at a valuation substantially lower than what was ultimately obtained in the Series C investment. The Company rejected these valuations as too low and, having received an initial term sheet from Forest Labs in December 2012, began negotiating the terms of the option and license agreements with Forest Labs on January 5, 2013. Also in January, the Company initiated separate
negotiations for a capital raise with two well-known venture capital funds who were not existing investors in the Company and began trading draft term sheets with these firms.
As the discussions with Forest Labs progressed, the Company continued negotiating with the two venture capital funds due to the uncertainty of concluding a transaction with Forest Labs. On March 7, 2013, as the terms of the option and license agreements became more settled, the Company and Forest Labs expanded their negotiations to address the terms and pricing of the proposed equity investment that was intended to occur concurrently with the option and license agreements. Shortly before the beginning of valuation discussions between the Company and Forest Labs, the board of directors of the Company authorized management to enter into the funding transaction with Forest Labs at an implied enterprise value of the Company that met or exceeded a set level. This threshold was set at a value higher than the valuations previously proposed by the two inside investor groups. Also in March, one of the two venture capital firms with whom the Company had been negotiating submitted a final term sheet valuing the Company at an amount substantially lower than what was ultimately obtained in the Series C transaction.
Towards the end of April, the terms of the option and license agreements and the stock purchase agreement between Forest Labs and the Company were largely settled, with a $125 million enterprise valuation of the Company implied by the purchase price materially above the threshold authorized by the Company’s board of directors. With the transaction near completion, the Company contacted the two venture capital firms to gauge their interest in participating in the Series C round at the same price established through negotiations with Forest Labs. Both firms declined, stating that the price was too high in their estimation. The firm that had not already submitted a term sheet also indicated that they had been considering a value closer to what the other firm had proposed.
On May 3, 2013, the Company and Forest Labs signed the option and license agreements and the stock purchase agreement for the Series C investment. The Company also entered into stock purchase agreements with its existing investors for a purchase of the same size — $30 million — and on the same terms as the one made by Forest Labs.
The Company’s rejection of a series of investment proposals with valuations the Company deemed too low demonstrates that the Company was not forced to enter into the May Transaction. That the Company sought and received competing valuations shows that the Company actively tested the market for its preferred shares to obtain the best terms and was well informed when making a final decision on the valuation it was willing to accept. The Company’s negotiation of a final Series C valuation that was above the threshold established by its board of directors to authorize a transaction confirms that the Company fully exercised its fiduciary duty to its shareholders to obtain the greatest value possible in its sale of an equity interest in the Company. Empirical data support the competitiveness of the $125 million enterprise value that the Company obtained in the Series C. First, the median enterprise valuation of the seven
biotechnology companies that completed IPOs from the beginning of 2013 through the date of the Series C was $109M. Additionally, as of the date on which the backsolve to the Series C valuation was completed (three days before the signing of the Series C stock purchase agreements), the median enterprise value of the Company’s comparable companies was $104 million.
The extensive process the Company followed to obtain multiple offers for its equity and then to negotiate the best and highest value possible with the party with whom it eventually concluded the May Transaction places this value squarely within the Practice Guide’s definition of fair value. Both the Company and Forest Labs were willing parties to the transaction and neither was forced to enter into it. Additionally, the May Transaction was a current transaction in that it was unrelated to any prior transaction between the parties. The Company believes that it would be inconsistent with the definition of fair value and the principles of the Practice Guide to ignore persuasive and reliable evidence of enterprise fair value from an arm’s length contemporaneous transaction in the Company’s own securities and place greater weight on after-the-fact assessments based on changed circumstances and events. Because the Series C investment reflected the fair value of the Company at that time, using the backsolve method to establish a fair value for April 30, 2013, a date within three days of the Series C financing, was both appropriate and consistent with the approaches for determining fair value prescribed by the Practice Guide.
The entry of Forest Labs into option and license agreements with the Company as part of the May Transaction does not render Forest Labs a related party for purposes of the Series C transaction
We understand that the Staff may be concerned that the entry of the Company into option and license agreements with Forest Labs at the same time as the Series C investment may render Forest Labs a related party such that the entry into the Series C financing was not an arm’s length transaction. In this regard it is essential to recall that the Company and Forest Labs entered into all three agreements — the option agreement, the license agreement, and the Series C stock purchase agreement — simultaneously as part of the May Transaction. Immediately before the May Transaction, there was no relationship between the two parties other than as negotiators of a potential agreement between them. As a result they were not related parties as that term is defined in the Practice Guide and FASB ASC 850. (Practice Guide, Glossary.) Even if after the May Transaction was complete Forest Labs and the Company became related parties because of the May Transaction, they would not have been related parties at any time they negotiated and concluded the May Transaction. Accordingly, the May Transaction was an arm’s length transaction. (The Company notes that, as a general matter, the entry into the option and license agreements concurrently with the Series C financing could raise the question of the allocation of value to each of the transactions. As described more fully below, all of the value paid by Forest Labs as part of the May Transaction is attributable to the Series C investment.)
The Series C transaction between Forest Labs and the Company was an arm’s length transaction notwithstanding that inside investors also participated in the funding round
We understand that the Staff may be concerned that the participation of the Company’s inside investors alongside Forest Labs in the Series C financing could indicate that the Series C investment by Forest Labs was not an arm’s length transaction and that consequently the Series C financing did not establish the fair value of the Company. The Staff may be informed in this view by a statement in the Practice Guide that “some might consider the sale of preferred stock in a second round of financing to an existing investor a related-party transaction, even if other preferred shares in the same round are sold to new shareholders.” Practice Guide, 1.12, fn 6. We note that the Practice Guide is suggesting that “the sale of preferred stock in a second round of financing to an existing investor” could be considered a related-party transaction, but it does not suggest that the new shareholders would or should also be considered related parties. Where a transaction occurs between a company and both a related party and an unrelated party, and the terms of the transaction are set by negotiation between the company and the unrelated party, these terms indicate fair value, even if related parties subsequently participate in the transaction.
In the text to which the above quoted language is footnoted, the Practice Guide states that,
privately held enterprises may sometimes engage in arm’s-length cash transactions with unrelated parties for issuances of their equity securities, and under certain conditions, the cash exchanged in such a transaction is an observable input. Those conditions are . . .
. . .
b. the transaction is a current transaction between willing parties (that is, other than in a forced or liquidation sale and other than under terms or conditions arising from a previous transaction).
Practice Guide, 1.12 (emphasis in original). The Practice Guide here returns to the central question of fair value in observable transactions, which is whether the parties are willing and unforced and the transaction is current. Transactions with preexisting shareholders may fail this test because the transaction is not viewed as current, but as an extension of a previous funding transaction. At the same time, the transaction with the new shareholders may pass the test, and if so, fair value will have been established by the transaction with the new shareholders.
The Company need not draw a conclusion as to whether the participation of existing shareholders in the Series C financing was an arm’s length transaction. That transaction adopted the terms of the transaction between Forest Labs and the Company. The Practice Guide states that,
[w]hen [a preferred stock financing] involves new investors, and the company is not in distress, the company and existing investors, if any, have an incentive to
negotiate the best possible price for the shares; however, the new investors have an incentive not to overpay. These transactions generally are relevant in estimating the fair value of the securities within the enterprise.
Practice Guide, 8.03.a.1. As described extensively above, the Series C transaction between Forest Labs and the Company was a current transaction (i.e., one unrelated to any prior transaction between the parties) between parties who were willing and unforced to enter into the transaction, concluded after extensive negotiations in which each party sought to obtain the best price. The subsequent participation of existing investors does not preclude this conclusion since the existing investors did not force the Company to enter into the transaction and, other than setting a minimum acceptable value, did not interfere with the Company’s negotiation of the transaction. Accordingly, the Series C transaction between Forest Labs and the Company indicates the fair value of the Company at the time of the transaction.
The option and license agreements were of negligible value in the May Transaction and so do not reduce the allocation of the full value to the Series C funding
As described above, the entry of Forest Labs and the Company into option and license agreements concurrent with the Series C stock purchase agreement as part of the May Transaction does not render Forest Labs a related party for purposes of the Series C transaction. According to the Practice Guide, however, “[w]hen an investor invests in [a] company and also negotiates a formal license to use the company’s products in the investor’s business, it may be necessary to allocate value between the equity investment and license.” Practice Guide, 8.03.b.ii. However both the option agreement and the license agreement which was the subject of the option had negligible value as of May 3, 2013. Because of the historical challenges in developing acute heart failure drugs, there is a significant probability that Forest Labs will never receive any value under the optioned license agreement, particularly after paying the $65 million option exercise fee and subsequent milestones. This assessment is consistent with Forest Labs’ treatment of the option and license agreements in their own second quarter 2013 financial statements filed with the Commission on Form 10-Q. There, Forest Labs allocated the full amount of the $30 million investment to “Marketable securities and investments” and none to Research & Development. If the option had value in Forest Labs’ view, some allocation to Research & Development would have been necessary.
Hypothetically, even if the option had greater than a de minimis current value, the allocation of a portion of the $30 million spent by Forest Labs in the May Transaction to the option and license agreements would reduce the value imputed to the Company’s equity. That is, Forest Labs would have purchased the same proportion of the Company’s equity for less than the $30 million now attributed to this purchase, suggesting that the total value of the Company was proportionally lower. Our understanding is that the Staff’s concern with the valuation of the Company indicated by the Series C financing is that it may be understated, and therefore that stock options granted with exercise prices derived from that valuation would have been priced too low.
Attributing hypothetical value to Forest Labs’ purchase of the option and license agreements would only exacerbate this concern.
Conclusion
Forest Labs and the Company were willing and informed parties in the Series C financing, and neither was forced to transact. As a result, the amount at which Forest Labs purchased an interest in the Company reflects the fair value of the Company on that date. Although the Company also transferred an option to Forest Labs concurrent with the Series C investment, the value of the option was negligible, so the inclusion of the option in the May Transaction does not alter the fair value assigned to the Company’s equity, nor does the option agreement render Forest Labs a related party with respect to the Series C investment. Finally, the participation of the Company’s existing investors in the Series C raise on the same terms as Forest Labs does not change the fact that these terms were set by competitive negotiations between willing but unforced parties, i.e., Forest Labs and the Company. For all these reasons, the Series C preferred stock transaction revealed the fair value of the Company at the time of the transaction and the Company’s use of the backsolve method utilizing this fair value is both appropriate and consistent with the Practice Guide.
· Regarding the second bullet, it is unclear why use of the indices to adjust the enterprise value is appropriate. The use of comparable companies, adjusted for your specific facts and circumstances appears more appropriate than the use of indices to adjust the enterprise value. Please revise your disclosure accordingly.
Response to second bullet:
In response to the Staff’s Comments, the Company advises the Staff as follows: The purpose of the first adjustment the Company made to its enterprise value calculated in the April 30, 2013 valuation as part of its August 15, 2013 valuation was to apply the effect of changes in market conditions affecting the biotechnology industry since April 30, 2013. To make this adjustment, the Company began by examining the changes in enterprise value of its comparable companies. Because the range and variance of the changes in value of these companies over the 3.5 month period of interest was so large, as described more fully in the immediately following response, the Company concluded that these changes in value more likely reflected company-specific factors and events rather than broader market conditions. For this reason, the Company did not use the change in values of the comparable companies in determining the market adjustment factor. Instead, the Company concluded it would be more appropriate to use the 9.7% mean change in value of three indices measuring the relevant market — the NASDAQ Biotechnology Index, iShares NASDAQ Biotechnology Index, and the S&P Biotech ETF. By coincidence, the mean of the change in enterprise value of the comparable companies (10.9%) was close to the mean change in the indices, but for the reasons just explained, was not appropriate for use as a market adjustment factor.
The Company proposes to modify the disclosure in the Amendment as shown in the Exhibit to more clearly refer to the use of this adjustment to capture the change in market conditions between the two valuation dates.
· Regarding the second bullet, your discussion regarding the comparable companies appears inconsistent. We do not understand the relationship between the wide range of changes in the enterprise values for comparable companies that your analysis refers to and the average change in the enterprise values of the comparable companies from April 30, 2013 to August 15, 2013 which was 10.9%.
Response to third bullet:
In response to the Staff’s Comments, the Company advises the Staff as follows: In the set of comparable companies the Company reviewed, the two highest increases in enterprise value over the period from April 30, 2013 and August 15, 2013 were +76.0% and +35.9% while the two largest decreases in enterprise value were -61.3% and -27.6%. The arithmetic mean of all the changes across the full set of comparable companies, including the four just cited, was 10.9%. The wide range referred to in the prior response was the difference between the increase in one company’s enterprise value of 76.0% and another company’s simultaneous decrease in enterprise value of -61.3% over the period of interest — a difference of 137.3 percentage points. This wide range is consistent with a mean value that falls near the center of the range. (The mean need not fall near the center of the range, of course, depending on the number and distribution of values included in the computation of the mean.)
The Company referred to the wide range in changes in enterprise value across the comparable companies in its earlier response to explain why it instead chose to use the change in value of biotechnology market indexes to adjust the original enterprise value from its April 30, 2013 valuation for use in its subsequent August 15, 2013 valuation. As described in the response above, this adjustment was intended to apply the effect of changes in market conditions affecting the biotechnology industry since the last valuation date to the Company’s enterprise value. To make this adjustment, the Company first examined the changes in enterprise value of its comparable companies. Because the range and variance of the changes in value of these companies over the 3.5 month period of interest was so large, the Company concluded that these changes in value more likely reflected company-specific factors and events (for example the results of clinical trials or development decisions made by collaboration partners) rather than broader market conditions. For this reason, the Company did not use the change in values of the comparable companies in determining the market adjustment factor. Instead, the Company used the 9.7% mean change in value of the three biotech indices described above. By coincidence, the mean of the change in enterprise value of the comparable companies (10.9%) was close to the mean change in the indices, but for the reasons just explained, was not appropriate for use as a market adjustment factor.
· Regarding the second bullet, please tell us how you determined the 10.9% increase for the comparable companies and why this number was not used.
Response to fourth bullet:
In response to the Staff’s Comments, the Company advises the Staff as follows: To determine the 10.9% increase for the comparable companies, the Company collected enterprise value data for each comparable company as of April 30, 2013 and August 15, 2013. The Company then calculated the percentage change in the enterprise value of each company over this period. Finally, the Company calculated the arithmetic mean of all of these percentage changes to yield an average increase of 10.9% for the set of comparable companies. If the Company had value-weighted the individual company changes in enterprise value, for example by summing the enterprise values for the companies at each date and calculating the percentage change in these sums, the weighted-average growth would have been 18.2%.
The reasons why the Company did not view the mean of the change in enterprise value of the comparable companies as appropriate for use as a market adjustment factor are described in the immediately preceding response.
· Regarding the third bullet, you state that the incremental value attributed to the assumed annual 20% return, after considering the offsetting effect of the $12 million reduction in the Company’s cash balance, did not materially impact the overall enterprise value of the Company or the estimated fair value of the Company’s common stock. However, your disclosure states that you adjusted the enterprise value by 13%, which appears material. Please revise to clarify.
Response to fifth bullet:
In response to the Staff’s Comments, the Company advises the Staff as follows: In calculating the Company’s enterprise value in its August 15, 2013 valuation, the Company began with the enterprise value calculated in the April 30, 2013 valuation and first adjusted it by 9.7% to account for the change in value in the market for biotechnology company stock, as described above. The Company’s next adjustment reflected the use of its cash between the two valuation dates and the effect of this investment on the enterprise value of the Company.
Between April 30, 2013 and August 15, 2013, the Company spent approximately $11.7 million of cash. The purpose of this investment was to grow the value of the Company. Enterprise value as used here explicitly excludes net cash, and equity value — enterprise value, plus cash, less debt — does not include the value of cash that has already been spent except to the extent reflected in enterprise value. To estimate the effect of this investment on enterprise value, the Company applied the $11.7 million cash expense to enterprise value which resulted in a 12.9% increase in value. As described in our previous letter, the Company also assumed a 20% return on this investment, which for the 3.5 month period over which the cash was used, resulted in a further 0.4% increase in
value yielding a total increase in enterprise value of 13.3%. It is the 0.4% to which the Company was referring when it stated that “the incremental value attributed to this assumed annual 20% return, after considering the offsetting effect of the $12 million reduction in the Company’s cash balance, did not materially impact the overall enterprise value of the Company or the estimated fair value of the Company’s common stock”.
The Company proposes to modify the disclosure in the Amendment as shown in the Exhibit to more clearly distinguish between these two contributors to enterprise value growth.
· Regarding the fourth bullet, the 50% probability of an IPO occurring appears low considering the timing of your filing the initial registration statement. Please tell us why you believe that percentage is appropriate.
Response to sixth bullet:
In response to the Staff’s comment, the Company advises the Staff as follows: The first conversation between management and its board of directors (the “Board”) regarding a possible IPO occurred on June 17, 2013 as part of a broad discussion of various future financing options for the Company — including further private offerings — and business development opportunities. At the time of that discussion, February 2014 was viewed as the earliest realistic date for an IPO.
The minutes of the Company’s July 10, 2013 Board meeting reflect that movement towards a decision to pursue an IPO had not progressed materially since the last Board meeting. There was significant disagreement among board members concerning the desirability of an IPO, including very strong opposition to an IPO by some of the board members. In the meeting, the Board reviewed the significant risks associated with pursuing an IPO. These included high development success risk since management and the Board agreed that TRV130 would need to be very successful in its ongoing Phase 1b clinical trial in order for an IPO to be possible. This trial was designed to directly compare TRV130 to morphine with respect to efficacy, safety and tolerability providing proof of concept for the program (more usually obtained in Phase 2). At the time of the Board’s discussion, the data were not expected to be available until December 2013.
The Board also believed there was high risk to an IPO by the Company both for reasons of market conditions and for reasons specific to the Company’s profile. For example, at the time of the Board’s discussion, most companies achieving a successful IPO in 2013 had very different profiles from Trevena, and one company with a similar profile — another pain therapeutics company, Iroko Pharmaceuticals Inc. — had recently withdrawn its IPO due to lack of interest. The Company had not yet obtained investor feedback that might have increased its assessment of the possibility of a successful IPO, and the feedback provided by a number of potential underwriters was that the Company would be a good candidate for an IPO only if it received good data in its ongoing Phase 1b trial for TRV130, which at the time was not expected until December 2013. Most underwriters
declined to discuss supporting the Company’s IPO until the availability of such data. Finally, since there had been more IPOs in the first half of 2013 than in the whole of 2012, both management and the Board were concerned that the market for biotechnology IPOs would become saturated and demand would decline by the time the Company was ready to begin or was already in the process of going public. As a result, in July there was significant uncertainty as to whether the Company would decide to pursue an IPO, and the probability of successfully consummating an IPO was also highly uncertain.
Discussions on a potential IPO did not culminate in a decision to proceed until August, when interim data from 10 subjects in TRV130’s key trial became available suggesting a full complement of positive data could be available as early as October. The Board authorized the Company to move forward in preparing a draft registration statement on Form S-1 for confidential filing with the Commission to allow the Company to complete the IPO process should the final TRV130 1b data be positive and should the window for biotech IPOs remain open until the time of the data’s availability. The target pricing date under this plan had been advanced to mid-November 2013.
As the Staff is aware, the decision to confidentially submit an initial registration statement does not assure the success of the IPO process. Even though the Company had decided to proceed with the confidential submission, a public filing was explicitly dependent on obtaining positive final data from the TRV 130 Phase 1b trial, which the Company ultimately received in early October. In addition, the Company had no way of knowing whether the window for biotech IPOs would remain open through the target pricing date. Moreover, the Company was still uncertain about market receptivity to the Company’s profile in the absence of any investor feedback. With this much uncertainty, and with a final decision resting on as-yet unknown clinical trial data, the Company believes that a 50% probability of an IPO occurring successfully as of August 15, 2013 was reasonable and appropriate.
· Regarding the fifth bullet, you state that the increase in value per share of common stock between April 30, 2013 and August 15, 2013 was primarily driven by the increased likelihood of an IPO in the near term. We note that you have assumed a 50% probability for the non-IPO scenario, but your proposed disclosure does not support the value you determined for the non-IPO scenario, which was weighted 50%. You state in your proposed disclosure that the average implied value per share of common stock in the IPO scenarios was $12.44 per share. If that amount was essentially weighted 50% and presumably the non- IPO value is lower, please tell us how you derived a $7.44 fair value of your common stock. Please revise to clarify.
Response to seventh bullet:
In response to the Staff’s comment, the Company advises the Staff as follows: The August 15, 2013 valuation was based on a full PWERM analysis combining a set of three IPO scenarios, bearing a weighted average valuation of $12.40 per share, with a non-IPO scenario valuation of $2.48 per share. (The value of $12.44 per share for the IPO scenarios was included in the prior disclosure in error. The Company proposes to
modify the disclosure to correct this.) The Company believes that the weighted average IPO scenario valuation was well-founded in light of the currently proposed price range of $12.00 to $14.00 per share. The non-IPO value was based on a market-adjusted OPM that began with the April 30, 2013 valuation. This was in turn determined by backsolving to the Series C valuation, as discussed above. For the reasons described elsewhere in this response, the use of a backsolve to the Series C valuation is appropriate and, accordingly, this non-IPO scenario value is also well-founded. The increase in valuation was therefore driven primarily by the increase in the probability of an IPO between April 30, 2013 to 50% on August 15, 2013.
In response to the Staff’s Comments, the Company proposes to modify the disclosure in the Amendment as shown in the Exhibit to provide more information about the various IPO and non-IPO scenarios that were used to determine the $7.44 valuation of the Company’s common stock.
· Your proposed disclosure states two primary drivers for the increased value per share of common stock between April 30, 2013 and August 15, 2013. In that disclosure you state that the increased valuations associated with early-stage biotech IPOs was one factor. Please clarify how this affected your value at each grant date as it does not appear that the changes in value reflect this factor.
Response to eighth bullet:
In response to the Staff’s comment, the Company advises the Staff as follows: The Staff is correct that the valuations associated with early-stage biotech IPOs were not included in the valuation conducted on April 30, 2013. These valuations were, however, included in the IPO scenarios included in the valuation conducted on August 15, 2013. Had such early-stage biotech IPO valuations been lower, the August 15, 2013 valuation would have been lower and there would have been a smaller change in value between April 30, 2013 and August 15, 2013. We therefore believe that the overall change in sentiment to biotech valuations and biotech IPOs between April 30, 2013 and August 15, 2013 does help to explain why the value of the common stock changed significantly between these dates.
The Company proposes to modify the disclosure in the Amendment as shown in the Exhibit to further clarify this point.
· Regarding the sixth bullet, there is a significant gap in the fair values used and the IPO price that your disclosures do not clarify. Please revise.
Response to ninth bullet:
In response to the Staff’s comment, the Company proposes to modify the disclosure in the Amendment to clarify the reasons for the difference between the fair value used in
the August and September option grants and the mid-point of the proposed offering range.
* * * *
As requested by the Staff, the Company acknowledges that:
· should the Commission or the staff, acting pursuant to delegated authority, declare the filing effective, it does not foreclose the Commission from taking any action with respect to the filing;
· the action of the Commission or the staff, acting pursuant to delegated authority, in declaring the filing effective, does not relieve the Company from its full responsibility for the adequacy and accuracy of the disclosure in the filing; and
· the Company may not assert staff comments and the declaration of effectiveness as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States.
* * * *
Please send any additional comment letters concerning the Amendment to Maxine Gowen, Chief Executive Officer of the Company, and Jim Fulton of Cooley LLP at the email addresses previously provided to the Staff. Please direct any questions or comments concerning the Amendment or this response letter to either the undersigned at (703) 456-8058 or to Mr. Fulton at (212) 479-6103.
|
|
Very truly yours, |
|
|
|
|
|
|
|
|
Brent B. Siler |
|
cc: |
Maxine Gowen, Ph.D., Trevena, Inc. |
|
|
Peter N. Handrinos, Latham & Watkins LLP |
Exhibit A
Fair Market Value Estimates
We are required to estimate the fair market value of the common stock underlying our share-based awards when performing the fair value calculations with the Black-Scholes option pricing model. The fair market value of the common stock underlying our share-based awards was determined on each grant date by our board of directors, with input from management. All options to purchase shares of our common stock are intended to be granted with an exercise price per share no less than the fair market value per share of our common stock underlying those options on the date of grant, based on the information known to us on the date of grant. In the absence of a public trading market for our common stock, on each grant date, we develop an estimate of the fair market value of our common stock in order to determine an exercise price for the option grants. We determined the fair market value of our common stock using methodologies, approaches and assumptions consistent with the American Institute of Certified Public Accountants, or AICPA, Audit and Accounting Practice Aid Series: Valuation of Privately Held Company Equity Securities Issued as Compensation, or the AICPA Practice Guide. In addition, we considered various objective and subjective factors, along with input from management and contemporaneous valuations, to determine the fair market value of our common stock, including:
· external market conditions affecting the biotechnology industry;
· trends within the biotechnology industry;
· the prices at which we sold shares of preferred stock;
· the superior rights and preferences of the preferred stock relative to our common stock at the time of each grant;
· our results of operations and financial position;
· the status of our research and development efforts;
· our stage of development and business strategy;
· the lack of an active public market for our capital stock; and
· the likelihood of achieving a liquidity event, such as an IPO or sale of our company in light of prevailing market conditions.
The per share estimated fair market value of our common stock in the table below represents the determination by our board of directors of the fair market value of our common stock as of the date of grant, taking into consideration the various objective and subjective factors described above, including the conclusions, if applicable, of contemporaneous independent third-party valuations of our common stock as discussed below. We computed the per share weighted average estimated fair value for stock option grants based on the Black-Scholes option pricing model. The following table sets forth information about our stock option grants since January 1, 2012:
|
Date of Issuance |
|
Number of Shares |
|
Exercise |
|
Estimated |
|
Estimated Fair |
| |||
|
February 2, 2012 |
|
63,707 |
|
$ |
0.68 |
|
$ |
0.68 |
|
$ |
0.56 |
|
|
April 19, 2012 |
|
125,805 |
|
0.68 |
|
0.68 |
|
0.56 |
| |||
|
October 17, 2012 |
|
32,258 |
|
0.68 |
|
0.68 |
|
0.56 |
| |||
|
June 17, 2013 |
|
1,057,247 |
|
2.23 |
|
2.23 |
|
1.55 |
| |||
|
June 19, 2013 |
|
26,655 |
|
2.23 |
|
2.23 |
|
1.55 |
| |||
|
August 6, 2013 |
|
39,490 |
|
2.23 |
|
2.23 |
|
1.55 |
| |||
|
August 12, 2013 |
|
98,724 |
|
2.23 |
|
2.23 |
|
1.55 |
| |||
|
August 27, 2013 |
|
69,286 |
|
7.44 |
|
7.44 |
|
5.21 |
| |||
|
September 3, 2013 |
|
197,449 |
|
7.44 |
|
7.44 |
|
5.27 |
| |||
|
September 26, 2013 |
|
122,579 |
|
7.44 |
|
7.44 |
|
5.21 |
| |||
Our options granted in 2013 generally vest over four years with one-sixteenth of the shares subject to the applicable option vesting every three months, with the following exceptions: four option grants, for an aggregate total of 184,616 shares, vest over four years with one-eighth of the shares subject to the applicable option vesting six months following the applicable vesting commencement date and one-twelfth of the remaining shares vesting every three months thereafter; two option grants, for an aggregate total of 3,226 shares, vest over four years with one-fourth of the shares subject to the applicable option vesting one year following the applicable vesting commencement date and one-twelfth of the remaining shares vesting every three months thereafter; one option for 197,450 shares vests over four years with one-fourth of the shares vesting one year after the vesting commencement date and one-thirty-sixth of the remaining shares vesting each month thereafter; and one option grant for 17,742 shares vests over three years with one-twelfth of the shares vesting every three months.
In determining the exercise prices of the options set forth in the table above granted since January 1, 2012, our board of directors considered the most recent available independent third-party valuations of our common stock, which were prepared as of July 8, 2010, April 30, 2013 and August 15, 2013, and based its determination in part on such valuations, with the analyses summarized below. The intrinsic value of all outstanding vested and unvested options of $30.2 million is based on a per share price of $13.00, the midpoint of the price range set forth on the cover page of this prospectus, 2,804,264 shares of common stock issuable upon the exercise of options outstanding as of September 30, 2013 and a weighted average exercise price of $2.23 per share.
Stock option grants from January 1, 2012 to October 17, 2012
Our board of directors granted stock options from January 1, 2012 through October 17, 2012, with each having an exercise price of $0.68 per share. The exercise price per share was supported by an independent third-party valuation of $0.68 per common share as of July 8, 2010 in connection with our initial Series B and Series B-1 preferred stock issuances. In conducting this valuation, we estimated the value of our common stock using the option pricing method. The option pricing method treats common stock as options on the enterprise’s value, with exercise prices based on liquidation preferences set forth in the terms of the preferred stockholders agreements. The enterprise value was determined based upon the Series B preferred stock pre-money valuation and this value was employed in the option pricing method for valuing the common stock. We completed the Series B and Series B-1 preferred stock issuances, at $1.00 per share, which is effectively $6.20 per share of common stock on an as-converted basis, for up to $39.3 million in July 2010. The common stock is treated as a call option that gives its owner the right to buy the underlying net assets at a predetermined or “strike” price at a liquidity event, such as an IPO, merger or sale. The option pricing method considers the various terms of the preferred stock, including the level of seniority among the securities, dividend policy, conversion ratios, and cash allocations upon liquidation of the enterprise. In addition, the option pricing method implicitly considers the effect of the liquidation preference as of the appropriate date in the future, not as of the valuation date. Under the option pricing method, value is allocated to the common stock only if the net assets of the enterprise exceed the liquidation preference at the time of the liquidity event. The option pricing method commonly uses the Black-Scholes formula to price the common stock as a call option. The Black-Scholes assumptions used in the July 8, 2010 valuation were determined as follows:
· The current value per share of common stock was management’s estimate of fair value based on the pre-money Series B financing value.
· The exercise price was calculated based on the aggregate liquidation preferences of the outstanding Series A and B preferred stock.
· The time until expiration was based on the estimated time horizon for common stock value. A four-year time horizon was used based on our early stage of development and strategic plans. It was assumed that if our research and development plans progress as planned, that a liquidation event would occur within four years; if the development plans fail or have limited success, it was assumed that we would liquidate for less than the preferred stock preference or that we would recapitalize, in either case leaving the common stock virtually worthless.
· The volatility factor of 100% was based on comparable companies in the U.S. biotechnology market with market capitalizations less than $100 million.
A discount for lack of marketability of 50% was then applied to the resulting Black-Scholes value. A discount for lack of marketability was applied to reach the final valuation of the common stock because, as we are a private company, there are impediments to liquidity, including lack of publicly available information and the lack of a trading market. Our determination of the discount included factors such as our proximity to an IPO, reduced funding risk and our progress made
on our clinical development program. The discount for marketability decreases as we move closer to marketability of common shares through an event, such as an IPO, and as the risk declines for our company as milestones are achieved.
We concluded that the value of our company remained relatively unchanged from July 8, 2010 through October 17, 2012. This was primarily attributable to the absence of a significant product inflection point, insofar as TRV027, our lead asset, was in a Phase 2a clinical trial from the end of 2010 into the first half of 2012, and our continued efforts to obtain financing to support our liquidity needs and funding of operating expenses. The specific facts and circumstances considered by our board of directors included the following:
· We had principally financed our operations through private placements of preferred stock and debt. In 2010 and 2011, we successfully closed and received $17.5 million in each year. The original issuance per share price of $1.00, which is effectively $6.20 per share of common stock on an as-converted basis, for the preferred stock remained unchanged in 2010 and 2011.
· There were no preferred stock issuances in 2012 despite fund raising efforts and our cash and cash equivalents at the end of 2012 were $6.7 million. At December 31, 2011, we had cash and cash equivalents of $17.1 million.
· In anticipation of completing a financing in 2012, we had a valuation conducted in October 2012 that indicated that the value of our common stock was less than $0.01, and as a result, we ceased making additional option grants until after completion of a financing.
· We had no completed clinical trial data for any of our programs from the end of 2010 through the end of 2011.
Stock options granted on June 17, 2013 and July 29, 2013
Our board of directors granted stock options on June 17, 2013 and July 29, 2013, with each having an exercise price of $2.23 per share. Three of these grants became effective upon later dates (June 19, 2013, August 6, 2013 and August 12, 2013) when the respective recipient initially became an employee of our company. The exercise price per share was supported by the most recent independent third-party valuation of $2.23 per common share as of April 30, 2013. In conducting this valuation, we utilized the option pricing model backsolve method to calculate our enterprise value utilizing as a starting point the May 2013 Series C financing at $1.632 per share of preferred stock, which is effectively $10.12 per share of common stock on an as-converted basis. The price per share of preferred stock in the Series C financing was deemed to be fair value because the price was set by willing and informed unrelated parties (Forest and our company), neither of whom was forced to transact. The Series C financing was concluded after extensive negotiations in which each party sought to obtain the best price. Although we also transferred an option to Forest concurrent with the stock purchase, the value of the option was negligible, so the inclusion of the option in the transaction did not alter the fair value assigned to our equity.
We estimated the value of our common stock versus the other share classes using the option pricing method, consistent with the methodology noted above in the July 8, 2010 valuation. Changes in assumptions since the July 8, 2010 valuation included adjusting the enterprise value based upon the Series C financing raise at $1.632 per share, changing the expected term to 2.5 years based on updated management estimates, utilizing volatility of 80.5% based on the median of comparable companies and reducing the discount for lack of marketability to 30%. The comparable companies we used were publicly traded companies selected primarily on the basis of the lead indications they have under development. These companies consisted of Pain Therapeutics, Acura Pharmaceuticals, Horizon Pharma, Zogenix and Neurocrine Biosciences, each of which have lead indications focused on pain/neurological disorders, and Aastrom Biosciences, Pozen and Cytokinetics, each of which specializes in cardiovascular indications. All of the selected companies have market capitalizations of less than a billion dollars and low or no product revenue, which we believe makes them representative of our size and stage of development.
Stock options granted on August 27, 2013 and September 26, 2013
Our board of directors granted stock options on August 27, 2013 and September 26, 2013 with an exercise price of $7.44 per share. One of these grants became effective upon a later date (September 3, 2013) when the recipient initially became an employee of our company. The exercise price per share was supported by the most recent third-party valuation of $7.44 per common share as of August 15, 2013. In conducting this valuation, we used the probability weighted expected return method, or PWERM, with two main categories of discrete scenarios—a non-IPO scenario and multiple IPO scenarios—to take into account the decision by our board of directors in August 2013 to proceed with preparations for a potential IPO. We applied a weighting of 50% to the non-IPO scenario, which was valued using the market adjusted option
pricing method, which captures the distribution of exits appropriate for a private company where there are numerous potential future pathways as yet poorly defined, and a weighting of 50% to the IPO scenarios included under the PWERM approach, which assumed that we would achieve an IPO exit in the near-term. This 50% probability of an IPO exit in the near term was deemed appropriate because a successful IPO could only be achieved if the ongoing Phase 1b trial for TRV130 yielded positive data, and the results of this trial were unknown as of the August and September option grant dates. In addition, we had not yet obtained investor feedback that might have increased our assessment of the likelihood of a successful IPO and we were concerned about subsequent changes in the market’s receptivity for biotech IPOs based on the early stage of our IPO preparations.
To calculate our enterprise value under the market adjusted option pricing method for the non-IPO scenario, we started with the enterprise value calculated in the April 30, 2013 valuation described above. We first adjusted that value, increasing it by 9.7%, to reflect changes in the market conditions for the biotechnology industry generally between April 30, 2013 and August 15, 2013, as evidenced by the growth in enterprise value of three relevant market indices (the NASDAQ Biotechnology Index, iShares NASDAQ Biotechnology Index and SPDR S&P Biotech). We chose not to use the mean change in the enterprise values of our comparable companies because such values did not indicate a trend in biotech company valuations, but rather were significantly influenced by company-specific events. We then further adjusted the enterprise value, increasing it by 12.9% to reflect value assumed to have been created with the cash we utilized between April 30, 2013 and August 15, 2013 and an additional 0.4% based upon an assumed 20% rate of return from this investment of our cash over this time. These further adjustments reflect the interim incremental value accretion that bridges the gap between major value inflection points. From this enterprise value we then estimated the value of our common stock using the option pricing method, consistent with the methodology noted above in the July 8, 2010 and April 30, 2013 valuations. We used a discount for lack of marketability of 30% in line with the April 30, 2013 valuation to reflect the absence of a near-term exit in the non-IPO scenario. The implied value per share of common stock in the option pricing method as of August 15, 2013 was $2.48 per share.
Our PWERM approach employed three IPO scenarios and weighted those as described below. We estimated our enterprise value under these IPO scenarios using the guideline public company method under the market approach. Under the guideline public company method, we considered an average of pre-money values for IPOs completed by biotechnology companies from the beginning of 2012 through the middle of 2013. We considered the value of cardiovascular therapeutic companies, which were typically at the low end of the comparable company range, and the value of platform-focused companies, which were typically at the high end of the comparable company range. The valuation range that we selected was between the valuation ranges for the two sets of companies because we work in multiple therapeutic areas, and employ a proprietary biased ligand research platform. In addition, we considered a medium multiple of invested capital as indicated by the IPOs. For the complete set of biotechnology companies that went public between the beginning of 2012 through the middle of 2013, the median step-up factor was 1.1x. However, for biotechnology companies with lead programs, partnerships with pharmaceutical companies and robust pipelines, the range of this multiple has been 1.2x to 1.4x. Given the general positive investor sentiment in the public markets for biotech IPOs, we used a factor of 1.3x as the upper end of the enterprise value range for our company, which is at a slight premium to the historical median.
For each of the various IPO scenarios, an equity value was estimated and the rights and preferences for each shareholder class were considered to determine what portion of the enterprise value to allocate to common shares. The common share value was then multiplied by a discount factor reflecting the calculated discount rate and the timing of the event. Lastly, the common share value in each scenario was multiplied by an estimated probability for that scenario. The probability and timing of each scenario were based on discussions between our board of directors and our management team.
We used the following three possible IPO scenarios under the PWERM, weighing them as indicated:
· an IPO at an assumed high valuation in the fourth quarter of 2013, weighted at 30%;
· an IPO at a lower assumed valuation in the fourth quarter of 2013, weighted at 12.5%, and
· an IPO at the higher assumed valuation in the first quarter of 2014, weighted at 7.5%.
A discount for lack of marketability of 10% was then applied to the resulting value for each IPO scenario. This discount was significantly less than the 30% applied to the April 30, 2013 option pricing model to reflect that the IPO scenarios assumed that we had moved closer to marketability of shares of common stock in anticipation of a successful IPO. The average implied value per share of common stock in the IPO scenarios was $12.40 per share.
The table below summarizes the significant assumptions utilized for each of the PWERM scenarios used in valuing the common stock and based upon which the fair value was determined to be $7.44:
|
|
|
PWERM Scenarios |
| ||||||
|
|
|
No IPO |
|
Early IPO |
|
Early IPO |
|
Late IPO |
|
|
Liquidity date |
|
2/15/2016 |
|
4Q13 |
|
4Q13 |
|
1Q14 |
|
|
Probability weighting |
|
50% |
|
30% |
|
12.5% |
|
7.5% |
|
|
Discount for lack of marketability, or DLOM |
|
30% |
|
10% |
|
10% |
|
10% |
|
|
Estimated per share value of common stock in each PWERM scenario after DLOM |
|
$2.48 |
|
$13.10 |
|
$10.91 |
|
$12.09 |
|
|
Probability weighted estimated per share value of common stock across IPO scenarios after DLOM |
|
|
|
|
|
$12.40 |
|
|
|
|
Probability weighted estimated per share value of common stock across all PWERM scenarios after DLOM |
|
$7.44 |
| ||||||
The primary drivers for the increased value per share of common stock between April 30, 2013 and August 15, 2013 were:
· The likelihood of a successful IPO increased. We believed a successful IPO would require good data from the then-ongoing Phase 1b trial for TRV130. In the beginning of August we received positive preliminary data from the first 10 subjects in the trial. Based on these data, our assessment of the likelihood that the final data from the study would be positive increased and so we initiated preparations for an IPO. The increased probability of a successful IPO was also linked to the advancement of TRV734, increased market receptivity to biotech IPOs, the decision by our board of directors in August 2013 to initiate the process for an initial public offering and the selection of co-lead underwriters.
· The increased valuations associated with early-stage biotech IPOs. The valuations associated with early-stage biotech IPOs were not incorporated into the April 30, 2013 valuation but these valuations were incorporated into the August 15, 2013 valuation methodology and increased the value of common stock as of that date thereby contributing to the change in value between April 30, 2013 and August 15, 2013.
Determination of estimated IPO offering price
In early November 2013, we determined the estimated initial public offering price per share of this offering, as set forth on the cover page of this prospectus, to be between $12.00 and $14.00 per share. We note that, as is typical in IPOs, the preliminary range was not derived using a formal determination of fair value, but was determined based upon discussions between us and the underwriters. Among the factors considered in setting the preliminary range were the following:
· our prospects and the history of and prospects for our industry;
· the general condition of the securities markets and the recent market prices of, and the demand for, publicly traded common stock of generally comparable companies;
· an assumption that there would continue to be a receptive public trading market for pre-commercial biotechnology companies such as us; and
· an assumption that there would be sufficient demand for our common stock to support an offering of the size contemplated by this prospectus.
The midpoint of the estimated price range for this offering reflects a significant increase over the estimated valuation as of August 15, 2013 of $7.44 per share. Investors should be aware of this difference and should recognize that the price range for this offering is in excess of our prior valuations. Further, investors are cautioned not to place undue reliance on the valuation methodologies discussed above as an indicator of future stock prices. We believe that the difference in value between the midpoint of the preliminary price range and management’s determination of the fair value of our common stock
of $7.44 per share as of August 15, 2013, the date of our last valuation, was primarily the result of a change in the probability of a near-term IPO from 50% to the 100% probability implicit in the preliminary price range. We note that the $12.40 average implied value per share of common stock in the IPO scenarios considered in the August 15, 2013 valuation is within the actual proposed offering range for the IPO. The primary factors that allowed us to progress towards an IPO from August 15, 2013, when we only attributed a 50% probability to being able to achieve a near-term IPO, to being in a position to determine the preliminary price range in early November 2013, were:
· We received positive final data from the 30 subjects in our Phase 1b clinical trial for TRV130 in early October.
· Subsequent to August 15, 2013, we completed several critical events necessary to proceed toward an IPO, including the confidential submission of a registration statement for an IPO in September, the public filing of the registration statement in October and testing-the-waters meetings with potential IPO investors in October that led us to believe an IPO was potentially feasible.
In addition, the estimated initial public offering price range necessarily assumes that the IPO has occurred, that a public market for our common stock has been created and that our preferred stock has been converted into common stock in connection with the IPO, and therefore excludes any discount for lack of marketability of our common stock, any discount to reflect the time value of money for the period from the assumed IPO dates back to the valuation date, any preferences of our preferred stock and any assumption of less than 100% probability of an IPO, all of which were factored into the August 15, 2013 valuation.
Investors should be cautioned that the midpoint of the price range set forth on the cover of this prospectus does not necessarily represent the fair value of our common stock, but rather reflects an estimate of the offer price determined in consultation with the underwriters. There are significant judgments and estimates inherent in the determination of these valuations. These judgments and estimates include assumptions regarding our future performance, including the successful enrollment and completion of our clinical studies as well as the determination of the appropriate valuation methods. If we had made different assumptions, our stock-based compensation expense could have been different. The foregoing valuation methodologies are not the only methodologies available and they will not be used to value our common stock once this offering is complete. We cannot make assurances as to any particular valuation for our common stock. Accordingly, investors are cautioned not to place undue reliance on the foregoing valuation methodologies as an indicator of future stock prices.
In the public markets, we believe there are investors who may apply more qualitative and subjective valuation criteria to certain of our clinical assets than the valuation methods applied in our valuations, although there can be no assurance that this will in fact be the case. As described above, as a private company we used a more quantitative methodology to determine the fair value of our common stock and this methodology differs from the methodology used to determine the estimated price range for this offering. The estimated price range for this offering was not derived using a formal determination of fair value, but rather was determined by negotiation between us and the underwriters. In particular, the estimate of fair value of our common stock as of August 15, 2013 was not a factor in setting the estimated price range for this offering. The price that investors are willing to pay in this offering, for which the price range is intended to serve as an estimate, may take into account other things that have not been expressly considered in our prior valuations, are not objectively determinable and that valuation models are not able to quantify.