Bavarian nordic: debunking the management narrative

August 2015
Bavarian Nordic (BAVA DC/BVNRY)
Debunking the Management Narrative
During Bavarian Nordic's second-quarter earnings call, the company's management sought to
downplay our critique of Prostvac's clinical data. But its arguments rely on cherry-picked studies
and inappropriate comparisons; in many cases, moreover, the presentation simply ignored our
detailed analysis in favor of repeating the same old claims as if we had not already addressed
them. Below we highlight some of the problems with Bavarian Nordic's case. In attempting to
characterize placebo-group survival in the pivotal Prostvac Phase 2 (Kantoff, et al., 2010) as
typical, the company uses misleading benchmarks, comparing the minimally symptomatic,
chemotherapy-naïve, relatively healthy men in that trial to patients in other trials who are highly
symptomatic, have sometimes progressed after chemotherapy, and often have less favorable
disease characteristics, like visceral metastases and high Gleason scores – all issues we
discussed in our original report.
Regarding immune responses, Bavarian Nordic re-frames data that fellow cancer-vaccine
advocates have described as "somewhat disheartening" (Kissick & Sanda, 2015) as if it were
highly impressive, ignoring multiple studies that found no significant responses at all. On the
bigger-picture question – what would allow Prostvac to succeed where so many other highly
similar agents have failed? – the company remains silent, just as it continues to remain silent on
the risk that future smallpox-vaccine orders will be slashed based on its own data showing that
a far smaller dose can achieve the same benefit.
Survival data
Bavarian Nordic's citations
Bavarian Nordic argues that "[t]he median survival of placebo mCRPC patients from several
Phase III trials reported in the same period [as the large Prostvac Phase 2] ranged from 12.2 to
21.7 [months]," citing eight different papers to make its case. But what trials "reported in the
same period" show is not actually what matters in assessing Prostvac. Several studies
published after the Phase 2 but drawing on patient experiences from around the same time
supply highly informative survival curves that reveal the mediocrity of the Prostvac data.
In some cases, there's nothing necessarily wrong with Bavarian Nordic's citations – in some
cases, we highlighted several of the same studies in our ow For example, in a
study of the GVAX cancer vaccine compared to docetaxel in asymptomatic men – roughly
Disclaimer: As of the publication date of this report, Kerrisdale Capital Management, LLC
("Kerrisdale"), other research contributors, and others with whom we have shared our research
(the "Authors") have short positions in and may own option interests on the stock of the
Company covered herein (Bavarian Nordic) and stand to realize gains in the event that the price
of the stock decreases. Following publication, the Authors may transact in the securities of the
Company. The Authors have obtained all information herein from sources they believe to be
accurate and reliable. However, such information is presented "as is", without warranty of any
kind – whether express or implied – and without any representation as to the results obtained
from its use. All expressions of opinion are subject to change without notice, and the Authors do
not undertake to update this report or any information contained herein. Please read our full legal
disclaimer at the end of this report.

comparable to the population in the pivotal Prostvac Phase 2 – median overall survival was 20.7 months in the vaccine arm and 21.7 months in the docetaxel arm, outperforming the Halabi model by approximately 5 and 6 months, respectively (Higano, et al., 2009). While those figures may look slightly worse than the survival posted by the Prostvac arm in the Phase 2 (though likely not to a statistically significant degree), the Prostvac group was unusually healthy even relative to other asymptomatic patients. For example, while 13% of the GVAX trial participants had visceral metastases, none of the Prostvac trial participants did; while almost half of GVAX trial participants had Gleason scores of 8 or higher, an indication of aggressive disease, none of the Prostvac trial participants did. Indeed, as we pointed out in our original report, when looking just at the GVAX trial participants with Halabi-predicted survival greater than 18 months – a proxy for patients with good prognoses – median overall survival was actually 29.7 months in the GVAX arm and 27.1 months in the docetaxel arm. Within the Prostvac treatment arm, at least 75% of the patients did indeed have Halabi-predicted survival of 18 months or higher, making this a fair comparison with a trial that Bavarian Nordic itself called attention to.* Similarly, the company cites the original publication of survival data for the TAX 327 docetaxel trial (Tannock, et al., 2004). But that trial, which ran from March 2000 to June 2002, enrolled a mixed group of patients, including many with high PSA levels, high Gleason scores, and visceral metastases. The pivotal Prostvac Phase 2, by contrast, explicitly included only "minimally symptomatic" men generally free from these negative factors. Fortunately, a follow-up study (discussed in our original report) specifically examined the TAX 327 results for minimally symptomatic patients and found that they experienced median OS of 25.6 months, slightly higher than the Prostvac treatment arm in an older study (Berthold, et al., 2008). Thus we see that Prostvac's mediocre results are not rescued by the mere fact that the study took place in the early 2000s. Contrary to Bavarian Nordic's claims, comparable studies from the same period or earlier delivered similar or better results. TAX 327, which began in 2000, showed 26-month median OS for minimally symptomatic patients, only some of whom received life-extending chemotherapy, which in turn has only a small impact on survival; the GVAX trial in asymptomatic men, beginning in 2004, showed 28-month median OS for similar patients, half of whom received an ineffective cancer vaccine. The pivotal Prostvac Phase 2, initiated in 2003 and overlapping with the GVAX trial, showed 25.1-month median OS in the treatment arm – clearly not a standout result relative to the two just discussed – and 16.6-month median OS in the placebo arm – again, a strikingly bad outcome given how favorable the patients' disease characteristics were. The fact that Prostvac actual y looks worse than GVAX in a similar patient population is hardly a good sign, given that GVAX failed in Phase 3. One of the other studies Bavarian Nordic cites specifically examines men who progressed after chemotherapy (Sternberg, et al., 2009). Naturally, such patients, having already tried a harsh treatment and gotten worse, tend to have fewer years left than their chemotherapy-naïve peers, making this comparison uninformative. Similarly, the top-line results for chemotherapy trials in prostate cancer in general will be fairly poor, since patients tend to resist chemotherapy except as a last resort. Chemotherapy-focused studies thus tend to understate survival for healthier men with more options, like those enrolled in the Prostvac Phase 2. In the table below, we summarize the differences in baseline characteristics between the Prostvac Phase 2 and the studies that Bavarian Nordic cites. Note also that, as usual, median ages in the other studies are approximately 70, substantially lower than the 79-year median in the Prostvac placebo arm. * While the Halabi 2003 nomogram is not especially accurate in terms of its absolute predictions, it does tend to distinguish patients with good and bad prognoses. Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153

Receiving
Reference
age (yrs)
visceral
progressing
metastases
after chemo
(0 = best)
(Higano, et al., 2009) [Prostvac]
Source: cited studies (see References), Kerrisdale analysis Note: Prostvac row shows figures for treatment arm; other rows show figures for control arms. NR = not reported *Sum of lung and liver, but the categories may not be mutually exclusive. **While the control group received placebo, the trial was double-blind, so patients had to have been prepared to receive chemotherapy had they been randomized to the treatment arm. †The median score is 7; the upper end of the range goes to 10. Relative to the placebo groups in the studies that Bavarian Nordic cites, the Prostvac treatment arm has the lowest median PSA level, among the most favorable distributions of performance status, and the lowest (0%) level of patients with high Gleason scores and visceral metastases. In short, the Prostvac treatment arm was unusually healthy and thus had a meaningful head start relative to many other trials. The lower survival numbers in those trials are primarily a Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153

function of the less healthy patients they enrolled. Again, Bavarian Nordic relies on misleading benchmarks to make Prostvac look better. Studies that Bavarian Nordic prefers not to mention (and the curious case of
CureVac)
While management would like to focus on trials involving relatively sick patients, a number of
retrospective studies looking at broad populations show that typical survival can be much better.
A 2005 paper (whose authors include one of the key Prostvac researchers, James Gulley)
analyzed patients treated at the National Cancer Institute in two chemotherapy trials. For
patients with radiographic evidence of metastatic disease at the time of castration resistance,
median overall survival from the inception of castration resistance was 44.2 months (Sharifi, et
al., 2005). Depending on the lag from castration resistance to clinical-trial enrollment, lower
reported survival figures are of course possible, but this NCI study offers a relevant starting
point and does not rely on especially recent data. (One underlying trial was initiated in 1999, the
other in 2001.†)
A similar figure comes from a recently published study examining men treated for mCRPC in
2006-07 at Queen Elizabeth Hospital, Birmingham, UK (Afshar, Evison, James, & Patel, 2015).
Median overall survival for men with CRPC who presented as metastatic at diagnosis was 38.7
months. The authors explicitly note that their data reveal how unrealistic older estimates of
mCPRC life expectancy are (emphasis added):
Some current international guidelines quote ≤19 months as a survival figure for patients
with metastatic CRPC. In our study, median survival is more than double this. We have
shown survival more than previously reported figures and believe that these data benefit
clinicians and patients in understanding prognosis and treatment choices. Importantly,
our patients were diagnosed before the current wave of novel therapeutics for
CRPC, so survival for men diagnosed today may be more than our findings.

In other words, the excuse that patients only live longer because of new drugs like Xtandi and Zytiga doesn't apply to this data set, which, as already mentioned, goes back to 2006. Another retrospective study, which we highlighted in our original report, looks at chemotherapy-naïve clinical-trial participants with mCRPC treated at the Royal Marsden Hospital in the UK from 2003 to 2011 (a period overlapping with the pivotal Prostvac Phase 2) (Omlin, et al., 2013). Those men experienced median OS of 30.6 months from time of referral – again, superior to the Prostvac Phase 2 and comparable to the Phase 1 combination trial with Yervoy. Importantly, the patients outperformed both the Halabi 2003 and the Smaletz models (by 10 and 13 months, respectively), notwithstanding Bavarian Nordic's contention that those models generate extremely accurate point estimates. While the company would like to believe that outperforming outdated prognostic models is just as good as demonstrating a real survival advantage in a randomized, controlled, well-balanced clinical trial, if that were true then GVAX, which also † Trials(ketoconazole/alendronate sodium) and(docetaxel/thalidomide). Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 handily beat the Halabi model, would be on the market today instead of standing as one of many examples of cancer vaccines that failed in Phase 3. Bavarian Nordic did not acknowledge or discuss the fact that, even in the more recent Prostvac/Yervoy Phase 1, median OS looks unimpressive relative to the results of similar clinical trials. A recent example is especially intriguing. CureVac, a small, private German firm, reported Phase 2 OS data from a single-arm, 36-patient trial of its own prostate-cancer vaccine candidate, which consists of mRNA encoding the genes for PSA and three other prostate-related antigens (Kübler, et al., 2015). Like the Prostvac trials, the CureVac trial enrolled relatively healthy patients, and median OS was 31.4 months, almost identical to the Prostvac/Yervoy result but with a much simpler treatment and fewer severe side effects. Given the lack of a control group, these data are difficult to interpret, but, using Bavarian Nordic's pseudo-logic, they look impressive, with median OS exceeding the Halabi prediction by 15 months,‡ higher than any of the deltas that Bavarian Nordic touted for Prostvac during the second-quarter call. However, we believe that CureVac, like Prostvac, is likely targeting the wrong antigens and, in reality, is not clinically beneficial. Instead, 31.4-month median OS is simply a normal outcome for a placebo-like treatment. But even if we were wrong, CureVac's results are stil bad news for Prostvac, because they raise the specter of a lower-cost, equally effective competitor. We do not believe that any other analysts have acknowledged the challenges that the CureVac data pose. Finally, while Bavarian Nordic tries to cast our analysis of the available mCRPC survival data as biased and self-serving, it's worth pointing out that several highly regarded researchers at the National Cancer Institute made very similar points in their 2011 critical review of the clinical data on cancer-vaccine efficacy, albeit while discussing Provenge (sipuleucel-T), not Prostvac (Klebanoff, Acquavella, Yu, & Restifo, 2011): A recent subgroup analysis of the TAX327 study, one of the registration trials for docetaxel in metastatic CRPC patients, suggests that men who were asymptomatic or minimally symptomatic had a median overall survival of 25.6 months. This value is very close to the 25.8-month survival observed in patients randomized to the sipuleucel-T arm of the study. These scientists had no financial motivations yet came to the same conclusion: 25-month survival for minimally symptomatic mCRPC, now or a decade ago, is simply not an impressive result. Immune responses
Previously, we highlighted the weakness of the data on immune responses to Prostvac; the
vaccine fails to elicit measurable levels of PSA-specific T cells in a large fraction of patients,
‡ See the March 201by CureVac CEO Ingmar Hoerr, slide 34. Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 and, even in so-called responders, the response is tiny relative to the effects of ordinary infectious-disease vaccines and even other unsuccessful cancer vaccines. One concise summary of this point comes from a recent review paper by two cancer-vaccine researchers (emphasis added) (Kissick & Sanda, 2015): Two of the most effective human viral vaccines, YF-vax for yellow fever and Dryvax for
small pox induce a CD8 T-cell population of around 5-30% of the entire CD8 population
in the peripheral blood. As there are approximately 4 x 1010 total CD8 T-cells in the
human body, this translates to approximately 2-12 x 109 cells induced by these vaccines.
In comparison to these viral vaccine regimes in healthy host [sic], current cancer
vaccines only induce around a 2-fold to 10-fold increase in antigen specific T-cells.
Prostvac, a poxvirus modified to express Prostate Specific Antigen (PSA), increase the
number of PSA specific T-cells in patients by on average 5-fold to produce 30 vaccine
specific cells per million PBMCs. This number is around 0.03% of the total CD8 T-cell
population.
The gulf between Prostvac and an effective vaccine is enormous. Moreover, the main Prostvac immunological review paper (cited in the Bavarian Nordic presentation) includes several studies in which the vaccine arm also received immunostimulants like interleukin-2 and Yervoy, which might be expected to boost all T-cell levels irrespective of antigen specificity; standalone Prostvac trials have had even lower response rates. Nor does the review paper grapple with the fact that several Prostvac patients see their measured PSA-specific T-cell levels decline post-vaccination. Nor does it incorporate the results of the following studies:  Kaufman et al. 2004: "Figure 5 demonstrates a mixed response, with some patients exhibiting a significant increase in the number of PSA-specific precursors after [in vitro stimulation] in samples obtained following vaccination and some patients showing a decrease."  McNeel et al. 2015: "Significant PSA-specific CD4+ and CD8+ T-cell responses and IgG antibody responses specific for PSA were not detected."  DiPaola et al. 2015: "In this small cohort of patients, we did not find a clear and significant T-cell or PSA antibody immune effect." Not directly stated but visible in Figure 2 is the fact that only three out of eight evaluable patients had increases in PSA-specific T cells, while the other five had decreases, several much larger than any of the increases. In short, T-cell responses are small and inconsistent, sometimes absent altogether. While Bavarian Nordic tries to hang its hat on the fact that, among responders, PSA-specific T-cell levels approach levels for the influenza matrix protein used as a positive control, this comparison is nonsense. Influenza-specific memory cells persist at a low level, lying in wait to respond with a large, rapid burst to an acute influenza infection. During infections, though, influenza-specific T cells rise to several thousands of cells per million PBMCs within a week ( 100x the Prostvac level for "responders"), then subside back to lower levels (Wilkinson, et al., 2012). By contrast, Prostvac patients already have an acute ailment; if their PSA-specific T cells were truly functional in a clinically relevant way, they ought to expand dramatically in response to the cancer, not peak at just 0.03% of all CD8+ T cells. Moreover, as we explained during in earlier trials that included patient-level immune and clinical-outcome data, there was no correlation – indeed, sometimes an apparent inverse correlation – between immune responses and disease progression. Indeed, in one case, the most dramatic vaccine responder experienced the most rapid deterioration. Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 Bavarian Nordic cites a single study to argue that PSA-specific T-cell increases of 6x or higher are associated with higher overall survival, but, even if true, this result suffers from obvious confounding. Patients with stronger or more sensitive immune systems may be better able to mount immune responses to both Prostvac and their cancer, but that association may not be causal; the vaccine is not actually making them live longer. Moreover, the 6x threshold is arbitrary and smacks of data mining. Given that the researchers' usual threshold for marking a response as positive is a 2x increase, not a 6x increase, why don't they bifurcate survival at that lower point? If 6x is the magic level, then why regard lower levels as positive responses? One story the company has told to explain away the lack of strong immune responses is the notion of "antigen cascade" – perhaps a few PSA-specific T cells kill a few prostate-cancer cells, which release other antigens, which are presented to other T cells, which proliferate and kill other prostate-cancer cells, and so on. While the company presents this story as settled science, however, the evidence of it is meager and, once more, smacks of data mining. Across different studies, a varying, arbitrarily selected panel of antigens with some suspected link to prostate cancer are assessed along with PSA and, if measured levels of T cells specific for these other antigens increase sufficiently, this is treated as evidence for "antigen cascade." But, as with PSA, the actual magnitude of the responses to the other antigens is very small, often close to the limit of detectability. This creates a statistical problem: taking small samples to estimate tiny proportions will naturally generate noise, creating the appearance of fluctuations that may not amount to much. Meanwhile, in the event that T cells specific for some antigen decline, it's not counted against the vaccine; nor are responses to PSA alone treated as failures, even though they would seem not to have initiated any "cascades" In addition, as already mentioned, in several Prostvac studies that purportedly demonstrate "antigen cascade," patients also received IL-2, GM-CSF, or Yervoy; there's no good reason why all of the purported "cascade" should be attributed to the vaccine rather than these other, non-antigen-specific agents. There's a simpler problem for the antigen-cascade story as well. Fundamentally, the only proposed mechanism for Prostvac to help patients is to directly or indirectly cause the death of tumor cells. Without a cascade, PSA-specific T cells must be the killers; with a cascade, other T cells could, in theory, step in. But either way, if tumor cells actually die, then why is there no direct evidence of it? Prostvac has almost never been associated with objective responses (i.e. well-documented, significant tumor shrinkage or remission) or large (e.g. >50%) declines in serum PSA levels. But if prostate-cancer cells are in fact dying as part of a diverse T-cell "cascade," how is it that the death rate is sufficiently fine-tuned such that a) enough cells die to prolong patients' survival, yet b) not enough to ever be detected by conventional means that work for other cytotoxic therapies? The fact that prostate cancer is relatively slow-growing only exacerbates the difficulty of this question, because it should be easier for T-cell-mediated killing to overwhelm the underlying growth rate and result in net shrinkage. So where are all the objective responses? While observers have now learned not to expect such responses from cancer vaccines simply because they've almost never occurred, there is no compel ing explanation for how such vaccines could be clinically beneficial in the proposed manner without having such tangible effects. In closing, it's interesting to benchmark Prostvac against another type of immunotherapy: adoptive cell transfer. For Prostvac to work, its viral vectors need to infect dendritic cells, which must traffic to draining lymph nodes and locate there, among the ranks of T cells, the very small number of PSA-specific cells that survived central deletion in the thymus and weren't converted to regulatory T cells, and those T cells must receive the proper activation signals, migrate long distances to find and make direct contact with tumor cells, and lyse them before succumbing to Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 "exhaustion" or other immunosuppressive influences (including prostate-specific regulatory T cells). There are obviously many points at which the process could stop prematurely or fail entirely. With adoptive cell transfer, by contrast, clinicians can directly engineer billions of T cells targeting a single specific antigen and inject them directly into the patient, usually after conditioning regimens designed to kill off other T cells that might compete for resources, thus circumventing many of Prostvac's points of potential failure. Strikingly, though, despite its many advantages, adoptive cell transfer often still doesn't work. For example, a recently published study (Chodon, et al., 2015) showed that melanoma patients treated with billions or tens of billions of T cells specific for the melanoma-associated antigen MART-1, along with dendritic cells pre-loaded with MART-1 peptides, experienced only "transient" antitumor activity, with no true objective responses, despite the fact that MART-1-specific T cells rapidly rose to 30% of total T cells in the peripheral blood, a level 1,000 times higher than the peak PSA-specific T-cell response elicited among "responders" by Prostvac. Whatever the reasons for this particular trial's failure, it's difficult to believe, when a therapy 1,000 time stronger that Prostvac, targeting a more immunogenic type of cancer and routing around key Prostvac vulnerabilities like immunodominance and tolerance, nonetheless fails to have any clinical impact, that Prostvac itself has any chance of succeeding. References
Afshar, M., Evison, F., James, N. D., & Patel, P. (2015, August). Shifting paradigms in the
estimation of survival for castration-resistant prostate cancer: A tertiary academic center experience. Urologic Oncology, 33(8), 338.e1–338.e7. doi:10.1016/j.urolonc.2015.05.003 Berthold, D. R., Pond, G. R., Roessner, M., de Wit, R., Eisenberger, M., & Tannock, I. F. (2008, May 1). Treatment of Hormone-Refractory Prostate Cancer with Docetaxel or Mitoxantrone: Relationships between Prostate-Specific Antigen, Pain, and Quality of Life Response and Survival in the TAX-327 Study. Clinical Cancer Research, 14(9), 2763-2767. doi:10.1158/1078-0432.CCR-07-0944 Carducci, M. A., Saad, F., Abrahamsson, P.-A., Dearnaley, D. P., Schulman, C. C., Noth, S. A., . . Nelson, J. B. (2007, November 1). A Phase 3 Randomized Controlled Trial of the Efficacy and Safety of Atrasentan in Men With Metastatic Hormone-refractory Prostate Cancer. Cancer, 110(9), 1959-1966. doi:10.1002/cncr.22996 Chodon, T., Comin-Anduix, B., Chmielowski, B., Koya, R. C., Wu, Z., Auerbach, M., . . Radu, C. G. (2015, May 1). Adoptive Transfer of MART-1 T-Cell Receptor Transgenic Lymphocytes and Dendritic Cell Vaccination in Patients with Metastatic Melanoma. Clinical Cancer Research, 20(9), 2457-2465. doi:10.1158/1078-0432.CCR-13-3017 Higano, C. S., Schellhammer, P. F., Small, E. J., Burch, P. A., Nemunaitis, J., Yuh, L., . . Frohlich, M. W. (2009, August 15). Integrated Data From 2 Randomized, Double-Blind, Placebo-Controlled, Phase 3 Trials of Active Cellular Immunotherapy With Sipuleucel-T in Advanced Prostate Cancer. Cancer, 115, 3670–3679. doi:10.1002/cncr.24429 Higano, C., Saad, F., Somer, B., Curti, B., Petrylak, D., Drake, C., . . Sacks, N. (2009). A phase III trial of GVAX immunotherapy for prostate cancer versus docetaxel plus Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 prednisone in asymptomatic, castration-resistant prostate cancer (CRPC). Genitourinary Cancer Symposium: Proc Am Soc Clin Onco. Kantoff, P. W., Schuetz, T. J., Blumenstein, B. A., Glode, L. M., Bilhartz, D. L., Wyand, M., . . Gulley, J. L. (2010, March 1). Overall Survival Analysis of a Phase II Randomized Controlled Trial of a Poxviral-Based PSA-Targeted Immunotherapy in Metastatic Castration-Resistant Prostate Cancer. Journal of Clinical Oncology, 28(7), 1099-1105. doi:10.1200/JCO.2009.25.0597 Kissick, H. T., & Sanda, M. G. (2015). The role of active vaccination in cancer immunotherapy: lessons from clinical trials. Current Opinion in Immunology, 35, 15-22. doi:10.1016/j.coi/2015.05.004 Klebanoff, C., Acquavella, N., Yu, Z., & Restifo, N. (2011, January). Therapeutic cancer vaccines: are we there yet? Immunological Reviews, 239(1), 27-44. doi:10.1111/j.1600-065X.2010.00979.x Kübler, H., Scheel, B., Gnad-Vogt, U., Miller, K., Schultze-Seemann, W., vom Dorp, F., . . Fotin-Mleczek, M. (2015). Self-adjuvanted mRNA vaccination in advanced prostate cancer patients: a first-in-man phase I/IIa study. Journal for ImmunoTherapy of Cancer, 3(26), 1-14. doi: 10.1186/s40425-015-0068-y Omlin, A., Pezaro, C., Mukherji, D., Cassidy, A. M., Sandhu, S., Bianchini, D., . . de Bono, J. (2013, August). Improved Survival in a Cohort of Trial Participants with Metastatic Castration-resistant Prostate Cancer Demonstrates the Need for Updated Prognostic Nomograms. European Urology, 64(2), 300-306. doi:10.1016/j.eururo.2012.12.029 Petrylak, D. P., Tangen, C. M., Hussain, M. H., Lara, J. P., Jones, J. A., Taplin, M. E., . . Raghavan, D. (2004, October 7). Docetaxel and Estramustine Compared with Mitoxantrone and Prednisone for Advanced Refractory Prostate Cancer. New England Journal of Medicine, 351(15), 1513-1520. doi:10.1056/NEJMoa041318 Saad, F., Gleason, D. M., Murray, R., Tchekmedyian, S., Venner, P., Lacombe, L., . . Chen, B. (2002, October 2). A Randomized, Placebo-Controlled Trial of Zoledronic Acid in Patients With Hormone-Refractory Metastatic Prostate Carcinoma. Journal of the National Cancer Institute, 94(19), 1458-1468. doi:10.1093/jnci/94.19.1458 Sharifi, N., Dahut, W. L., Steinberg, S. M., Figg, W. D., Tarassoff, C., Arlen, P., & Gulley, J. L. (2005, November). A retrospective study of the time to clinical endpoints for advanced prostate cancer. BJU International, 96(7), 985-989. doi:10.1111/j.1464-410X.2005.05798.x Small, E., Demkow, T., Gerritsen, W., Rolland, F., Hoskin, P., Smith, D., . . Hege, K. (2009). A phase III trial of GVAX immunotherapy for prostate cancer in combination with docetaxel versus docetaxel plus prednisone in symptomatic, castration-resistant prostate cancer (CRPC). Genitourinary Cancer Symposium: Proc Am Soc Clin Onco. Retrieved from http://meetinglibrary.asco.org/content/20295-64 Sternberg, C. N., Petrylak, D. P., Sartor, O., Witjes, J. A., Demkow, T., Ferrero, J.-M., . . Berry, W. (2009, November 10). Multinational, Double-Blind, Phase III Study of Prednisone and Either Satraplatin or Placebo in Patients With Castrate-Refractory Prostate Cancer Progressing After Prior Chemotherapy: The SPARC Trial. Journal of Clinical Oncology, 27(32), 5431-5438. doi:10.1200/JCO.2008.20.1228 Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 Tannock, I. F., de Wit, R., Berry, W. R., Horti, J., Pluzanska, A., Chi, K. N., . . Eisenberger, M. A. (2004). Docetaxel plus Prednisone or Mitoxantrone plus Prednisone for Advanced Prostate Cancer. New England Journal of Medicine, 351(15), 1502-1512. Wilkinson, T. M., Li, C. K., Chui, C. S., Huang, A. K., Perkins, M., Liebner, J. C., . . Xu, X.-N. (2012, February). Preexisting influenza-specific CD4+ T cells correlate with disease protection against influenza challenge in humans. Nature Medicine, 18(2), 274-281. doi:10.1038/nm.2612 Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 Full Legal Disclaimer

As of the publication date of this report, Kerrisdale Capital Management LLC and its affiliates
(collectively "Kerrisdale"), others that contributed research to this report and others that we have
shared our research with (collectively, the "Authors") have short positions in and may own
options on the stock of Bavarian Nordic A/S (the "Company") and stand to realize gains in the
event that the price of the stock decreases. Following publication of the report, the Authors may
transact in the securities of the company covered herein. All content in this report represent the
opinions of Kerrisdale. The Authors have obtained all information herein from sources they
believe to be accurate and reliable. However, such information is presented "as is", without
warranty of any kind – whether express or implied. The Authors make no representation,
express or implied, as to the accuracy, timeliness, or completeness of any such information or
with regard to the results obtained from its use. All expressions of opinion are subject to change
without notice, and the Authors do not undertake to update or supplement this report or any
information contained herein.
This document is for informational purposes only and it is not intended as an official
confirmation of any transaction. All market prices, data and other information are not warranted
as to completeness or accuracy and are subject to change without notice. The information
included in this document is based upon selected public market data and reflects prevailing
conditions and the Authors' views as of this date, all of which are accordingly subject to change.
The Authors' opinions and estimates constitute a best efforts judgment and should be regarded
as indicative, preliminary and for illustrative purposes only.
This document does not constitute a recommendation to transact in securities or derivatives of
the Company, and does not constitute a recommendation to buy, hold, sell or short shares of
the Company. Any investment involves substantial risks, including, but not limited to, pricing
volatility, inadequate liquidity, and the potential complete loss of principal. This report's
estimated fundamental value only represents a best efforts estimate of the potential
fundamental valuation of a specific security, and is not expressed as, or implied as,
assessments of the quality of a security, a summary of past performance, or an actionable
investment strategy for an investor.
This document does not in any way constitute an offer or solicitation of an offer to buy or sell
any investment, security, or commodity discussed herein or of any of the affiliates of the
Authors. Also, this document does not in any way constitute an offer or solicitation of an offer to
buy or sell any security in any jurisdiction in which such an offer would be unlawful under the
securities laws of such jurisdiction. To the best of the Authors' abilities and beliefs, all
information contained herein is accurate and reliable. The Authors reserve the rights for their
affiliates, officers, and employees to hold cash or derivative positions in any company discussed
in this document at any time. As of the original publication date of this document, investors
should assume that the Authors are long shares of WBS and have positions in financial
derivatives that reference this security and stand to potentially realize gains in the event that the
market valuation of the company's common equity is higher than prior to the original publication
Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153 date. These affiliates, officers, and individuals shall have no obligation to inform any investor about their historical, current, and future trading activities. In addition, the Authors may benefit from any change in the valuation of any other companies, securities, or commodities discussed in this document. Analysts who prepared this report are compensated based upon (among other factors) the overall profitability of the Authors' operations and their affiliates. The compensation structure for the Authors' analysts is generally a derivative of their effectiveness in generating and communicating new investment ideas and the performance of recommended strategies for the Authors. This could represent a potential conflict of interest in the statements and opinions in the Authors' documents. The information contained in this document may include, or incorporate by reference, forward-looking statements, which would include any statements that are not statements of historical fact. Any or all of the Authors' forward-looking assumptions, expectations, projections, intentions or beliefs about future events may turn out to be wrong. These forward-looking statements can be affected by inaccurate assumptions or by known or unknown risks, uncertainties and other factors, most of which are beyond the Authors' control. Investors should conduct independent due diligence, with assistance from professional financial, legal and tax experts, on all securities, companies, and commodities discussed in this document and develop a stand-alone judgment of the relevant markets prior to making any investment decision. Kerrisdale Capital Management, LLC 1212 Avenue of the Americas, 3rd Floor New York, NY 10036 Tel: 212.792.7999 Fax: 212.531.6153

Source: http://img.borsen.dk/img/cms/tuksi4/uploads/img_server_adm.files/2795_filename_6671.pdf