Over the past several months, shares of T2 Biosystems (TTOO) have plummeted. While everything about the business model of the company is known for a long time, nevertheless shares have been hammered. It seems to me investors still do not understand that the company represents a game changer in how blood cultures might be done in the future.
In addition, there were several articles published here on Seeking Alpha in reference to T2 Biosystems with a very bearish tone. While all is fair in love and war, and every author has the right to publish a bear article to benefit himself, it is unethical to present the arguments in such a way as to leave unanswered questions, or to imply without actually saying so something else.
In my opinion, those articles either presented part of the story, or omitted important facts, and in many cases are outright incorrect. So in this article, besides setting the record strain, I will also answer with counterarguments to several of the issues raised in these 3 articles.
Let’s start with MR (Mako Research) with their article on March 23. The author states:
"Sepsis, a leading cause of death in hospitals, can be caused by several organisms, including bacteria, fungi, and viruses. The current standard of care is to immediately administer broad-spectrum antibiotics to a patient suspected of having sepsis and simultaneously take a blood culture to identify specific pathogens and determine whether they are resistant to antibiotics. With the results, doctors can tailor the antibiotics to combat the specific cause of sepsis. T2 Biosystems aims to speed up the process of testing blood samples with its T2Dx device and related panels, but it lacks the ability to detect all relevant bacteria species as well as their susceptibility for specific antibiotics."
"According to the 2018 10-K, T2Bacteria tests for just 5 pathogens, which only account for half of all sepsis infections:
The bacteria species included in T2Bacteria are Staphylococcus aureus, Enterococcus faecium, Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa. The five bacteria species in our T2Bacteria Panel are responsible for about half of all septic infections.
The bull case around the T2DX is that it gets to a result faster than a blood culture or competing devices, potentially saving lives in the process. I agree with this view point. However, what bulls miss and why the T2DX is failing in commercial launch is that this speed advantage doesn't matter in practice.
Importantly, the T2DX is unable to detect half the species responsible for sepsis, so the T2Bacteria panel provides inadequate information to doctors attempting to narrow down the administered antibiotics."
While it is true that today T2 only tests for 5 bacterial pathogens that cause sepsis, what the author omitted is that the T2Bacteria Panel covers 90% of the deadly ESKAPE bacteria, those commonly not covered by empiric antibiotic therapy (e.g. broad spectrum therapy). So when broad spectrum antibiotics are administered concurrently with a T2 test, then 90% of all patients may be delivered effective therapy in 3-5 hours.
This is because if T2 picks up an infection, then immediately the pathogen is identified and the correct therapy is applied or adjusted. If the T2 does not pick up anything, then broad spectrum antibiotics will mostly cover the remaining pathogens, even if we have to wait 3 days on average for a blood culture to tell us what specific treatment to administer.
In 90% of the cases with an infection, patients would be on their way to better health when the T2 is used with broad spectrum antibiotics concurrently.
Please remember that the sooner sepsis pathogens are identified, the sooner patients can get the correct care. Sepsis is very dangerous. According to the CDC (Center for Disease Control), at least 1.7M adults in the U.S. develop sepsis, with nearly 270,000 deaths every year. In addition, 1 in 3 patients who die in hospitals have sepsis. So detecting sepsis early and getting patients on the correct medication is both a matter of saving lives and money. The cost of hospitalization for sepsis in the U.S. was $27B, more than any other disease. Sepsis kills more Americans annually than breast cancer, prostate cancer and aids combined.
Then the author says (same article):
"This lack of specificity in the test results means that the T2DX will never be the standard of care!"
This is simply not true. The T2Bacteria Panel has an overall average sensitivity of 90% and an overall average specificity of 98%, while demonstrating no interference from the presence of antibiotics in the bloodstream.
MR continues saying:
"Again, T2Candida detected ">50%" of the relevant species, but still misses a lot. Our checks suggest that the inability to detect many pathogens is the reason why TTOO has been unable to generate meaningful product revenue."
Again, while it is true T2Candida detects five clinically relevant species of Candida, the author is not telling all the facts. The species currently identified by the T2Candida panel account for >95% of all invasive candidiasis. Invasive candidiasis means a Candida infection in the bloodstream or in a deep-seated tissue such as in the abdomen. And why is detecting invasive candidiasis important? Because blood cultures fail to detect 50% of them.
In a paper by Clancy and Nguyen titled, Finding the "missing 50%" of invasive candidiasis, we read from the abstract:
"Blood cultures are limited for diagnosing invasive candidiasis by poor sensitivity and slow turn-around time. New diagnostics are needed to complement cultures, in particular to identify the "missing 50%" of patients who are blood culture-negative."
Please note that the overall mortality rate for Candida infections is about 40%. But if patients receive targeted treatment, the mortality rate can be reduced to 11%. Rapid detection of Candida enables physicians to provide targeted treatment quickly, and research has shown this can reduce a positive sepsis patient's length of stay in the hospital. In addition, a rapid negative result can prevent unnecessary hospital stay and antimicrobials, further reducing costs and preventing or reducing antimicrobial resistance.
In an article published on May 9 by WDR (White Diamond Research), the author states:
"In our opinion, the presented data was weak, like the slide shown above. It discussed a statistically insignificant difference between proportions of a positive T2 Bacteria test and a standard BC test. Again, inclusion of the presentation entirely focused on immune- compromised patients is evidence of the narrowness of the preferred T2 patient population. Many listeners left the audience before Dr. Walsh's presentations ended.
Overall, there was little energy in the audience. Only one question was asked: How do you select a proper patient for T2Bacteria testing?
The speaker panel did not have a clear answer to this fundamental challenge to T2 panel utilization. There was one comment from an EU participant, who complained about personal experience with poor clinical adoption of the T2 test. That was it from the audience."
Patient selection strategies apply to every diagnostic technology, in order to be able to effectively diagnose and correctly treat patients. I invite readers to see the video below from the MAD-ID 2019, The Antimicrobial Stewardship Meeting.
Debbie Goff, PharmD, FCCP, Ohio State University clinical associate professor and infectious diseases specialist, says that the University of Ohio was one of the first to introduce rapid diagnostic tests in 2009, and it was a "game-changer" compared to blood cultures. Then she says today we have T2, which is "the Star Wars of game changers".
Debbie Goff in her initial remarks on how to choose patients:
"The challenge for stewardship is how do you determine whose blood goes into this machine, and that’s skill of the stewardship team. It’s not up to the microbiologist to figure that out, they are not taking care of the patient, we are. And you work collaboratively to figure out this process. And that’s what I’m excited about. We have 2 clinicians who have figured it out and are using this in their hospital."
The next speaker in the video is Ryan Shields, PharmD, MS. He provides real world examples on how he uses T2Candida and also explains the complications of finding and curing invasive candidiasis. It’s not as simple as it sounds, nor is it just a function of getting a positive result. So targeting patients is actually part of the process for providing best of care.
Also, a very good comment from WDR's article is from the reader GlibSonoran:
"The issue is that there is a clear understanding that this can reduce hospital costs, reduce patient bed time and improve patient care and outcomes. But it's not as easy as run the test and reap the rewards, it requires a lot more from the hospital staff and antibiotic stewardship committee.
The hospital has to provide a multi-disciplinary approach based on solid guidelines and clinical algorithms (which are just decision trees/flowcharts) developed by a diagnostic and stewardship group. These algorithms are going to be different for different areas based on the local bacterial epidemiology. It creates more complexity for the physician.
An example hypothetical: Rather than just leave the patient on broad-spectrum antibiotics until blood culture comes back in 2 days (as they used to do) the physician now decides to eliminate two antibiotics and add another based on species identification from T2dx's results available 4 hours after blood draw. They determined to use T2dx on this bacteremia patient by comparing the patient's symptoms and any bio-markers to their clinical algorithm to find out if this was a good candidate. But the physician now has to consult with the pharmacist and clinical microbiologist in following the clinical algorithm to monitor the patient to make sure they aren't dealing with a resistant strain or a multi strain bacteremia. So they test for a couple of additional bio-markers and carefully monitor the patient's progress. If the bio-markers come back as expected and the patient improves, they've bought valuable time and provided better care than simply leaving them on the less effective broad-spectrum antibiotic, exposing them to liver damage. Plus using broad-spectrum antibiotics with poor effectiveness promotes the development of antibiotic resistance, which is a threat to every patient in the hospital. If the patient takes a turn for the worse they'll have to modify antibiotic treatment again per their algorithm. However, by carefully screening the patient before deciding to use the T2dx test, they find that the vast majority of patients get much improved care. Now when the blood culture finally arrives a couple of days later, they might tailor the antibiotics further by looking at the susceptibility results it provides.
So it takes a well-organized hospital that isn't broken up into insulated silos to make this work well. It takes a good stewardship committee that keeps up with the field and is aware of the local bacterial epidemiology. As a result, some institutions see better results than others."
So picking whose blood goes in a T2 is not a black and white issue. There are many things to consider and each decision flowchart is different for every stewardship team. And unless you are in such a team and have real life experience using a T2, I just don't see how one can formulate an expert opinion.
WDR also makes a point that most people do not have bloodstream infections (article here):
"On February 13, 2019, Diagnosticsworldnews.com published an article interviewing Inflammatix CEO, Tim Sweeney. The first sentence of the article says:
Diagnostics for acute infections and sepsis typically focus on "finding the bug," but most patients with infections do not have pathogens in their bloodstream.
A physician we interviewed told us:
Most patients with bacterial infections don't have bloodstream infections. In an emergency department population that is judged to have bacterial infections, only about 10-20% have blood stream infections. For the rest, the infection is walled off by the immune system from the rest of the body."
This might be true; however, the percentage of the patient population is not the issue. The issue is that these infections are a huge burden for the entire healthcare system. As the CDC says, at least 1.7M adults in the U.S. develop sepsis, with nearly 270,000 deaths every year. If the CDC is correct, 1.7M people (in the US alone) is not a small number.
Two recent studies confirm the cost effectiveness potential of using T2, here and here. So one can take the word of someone who never used the technology, or doctors who have studied it and say it saves money.
And as shown in the picture above, major hospitals are already saving the entire healthcare system millions with T2.
WDR also says that the FDA's "Breakthrough Device Designation" for the company's T2Resistance panel isn’t an acknowledgement that the device has value. Well, no one claimed that such a designation says anything about "value". However, as WDR writes, it does mean the company will get more attention for the approval of the device, assuming it proves its effectiveness. And in Europe, the T2Resistance panel is expected to be commercially available before the end of the year, with research-use-only version available in the U.S. on the same timeline.
T2's T2Lyme Panel
While WDR is right in that the T2Resistance panel will probably not be available commercially in the U.S. soon, I want to bring to your attention of another product that could be submitted for FDA approval around the end of the year, and that is the company's T2Lyme Panel.
Clinical trials for T2Lyme are currently being conducted and could be finished by Q4'19. About 3.4 million Lyme tests are conducted every year (link here) with an estimated cost of about $500M.
"Seven participating laboratories performed approximately 3.4 million LD tests on approximately 2.4 million specimens nationwide at an estimated cost of $492 million. Two-tiered testing accounted for at least 62% of assays performed; alternative testing accounted for <3% of assays. The estimated frequency of infection among patients from whom specimens were submitted ranged from 10% to 18.5%. Applied to the total numbers of specimens, this yielded an estimated 240 000 to 444 000 infected source patients in 2008."
Please note that while the report is dated 2014, the data in the report is from 2008. The reality is that even the CDC does not have recent data as to what is going on with Lyme.
My second observation is that the 3.4M tests were conducted on 2.4M specimens. Meaning, there are repeat tests because the results are very inaccurate or non-conclusive. In fact, during the early stages of the disease, the sensitivity of 2-tiered testing is low (30%–40%). Only in late stages of the disease (meaning weeks or months) does the sensitivity reach 70-100%.
Patricia Smith, President of the Lyme disease Association, has a 1 min video on the problem.
Another observation is that the 3.4M tests are far more than the 300,000 or so infections that accrue every year according to the FDA. So in most cases, these test are done to rule out Lyme disease. If that's the case, then T2Lyme is the perfect diagnostic tool, because in a preclinical trial, it proved to be better than any other method of detection. From T2's September 26 press release:
"The data from a 2017 pre-clinical study evaluated 21 patients enrolled with an erythema migrans (EM) rash with suspected early Lyme disease and assessed with multiple Lyme diagnostics. The study compares various diagnostic methods with the T2Lyme Panel, in determining a definitive Borrelia infection, using tissue biopsy culture from the EM rash. Of all the diagnostics tested, the T2Lyme Panel blood test was the most accurate compared to tissue culture with a 78% positive percent agreement (PPA) and 100% negative percent agreement (NPA). Two-tier serology, the currently recommended diagnostic, had a 56% PPA and 92% NPA compared to tissue culture, while blood polymerase chain reaction (PCR) did not correctly identify any positives. Eight (38%) patients were negative for all diagnostics. The 100% NPA of the T2Lyme Panel indicates greater specificity over serology resulting in less incidence of false positive results. Overall, the T2Lyme Panel has been evaluated against 558 negative samples with no false-positive results."
Please note that when evaluated against 558 negative samples, there were no false positives. And all that in 3-5 hours as opposed to weeks. The way I see it, it would be irresponsible for the medical community not to incorporate T2's Lyme panel if it gets approved in Q1'20.
While officially Lyme has been around for about 40 years, it is considered a new generational disease. Many call it the first climate change disease, because as the climate becomes warmer, the number of insect population increases around the world.
In addition to the fact that Lyme cases are increasing every year (especially in Europe), the report to Congress notes:
"Most Lyme disease patients diagnosed and treated early can fully recover, yet an estimated 10 to 20% of patients suffer from persistent symptoms that are potentially chronic and disabling. Using a research definition of and data on post-treatment Lyme disease syndrome (PTLDS), the number of PTLDS cases may approach 30,000-60,000 each year in the United States. A precise definition does not yet exist for chronic Lyme disease, so uncertainty is extremely large. The number of U.S. patients with a clinical diagnosis of chronic Lyme disease may be larger, but is unknown."
And the problem is not just in the U.S. but all over the world. The European parliament had this to say on the matter:
"MEPs have called for the EU to set out guidelines on Lyme disease, a condition transmitted by ticks affecting between 650,000 and 850,000 Europeans every year and whose diagnosis remains unclear.
Lyme disease is difficult to diagnose, particularly because it often appears in the form of subjective side effects, such as headaches or forgetfulness. Moreover, the tests to aid diagnosis are still far from operational.
As a result, making a diagnosis is often difficult and the estimates of the number of people with Lyme disease – between 650,000 and 850,000 Europeans – are approximate."
So while WDR is right in that the T2Resistance panel will not be available for commercial sale in the U.S. soon, they forgot to mention T2's Lyme panel that will probably be submitted for FDA approval around the end of the year.
The Minimum liquidity covenant issue
WDR says that while there is minimum liquidity covenant in the Term Loan Agreement with CRG, it is not disclosed in the credit agreement.
"There is also a minimum liquidity covenant. The amount of cash required is not disclosed in the credit agreement, but the 10-K says ..."
Actually it is disclosed but WDR missed it. Please read the loan agreement (link here) and please go down to exhibit A. The answer is $5 million.
In another instant, WDR states (same article fro previous link):
"In my opinion, given the current run rate for product revenue and the shortcomings of the product, T2 is likely to miss its revenue target again this year. CRG Servicing appears to have the option to simply accelerate the debt at any time, which would send the company into bankruptcy with its current cash balance."
No, I do not think CRG can do that, and I doubt they would want to. The company’s 10-Q provides some info on "potential sources of financing":
"Until such time as we can generate substantial product revenue, we expect to finance our cash needs, beyond what is currently available or on hand, through a combination of equity offerings, debt financings and revenue from existing and potential research and development and other collaboration agreements."
CRG understands the terms of its loan, and in addition to getting its money back, it also expects to make money on the warrants is has. Finally, on the Singular Web Call, CEO John McDonough said this (around 7:08 – 7:11 time):
"We’re in total compliance with all our loan covenants, by the way, with an extremely supportive partner."
Also noteworthy from the Singular Web Call is that T2 has 60 patents and 40 pending.
The non-recurring revenue issue
"The majority of 2018 revenue was non-recurring research revenues. Therefore, we expect the company to lower 2019 guidance, which could cause the stock to fall to new 52-week lows."
First of all, guidance has not been lowered. In T2's latest press release, the company once again reiterates guidance for 2019, saying it will be double that of 2018:
"In addition, the Company is reiterating its 2019 guidance of a doubling of revenue and securing 70-80 T2Dx instrument contracts."
Furthermore, in the conference call, CEO John McDonough said:
"So the product revenue guidance that we gave for Q2 is pretty much in line with the consensus that was out there and so the difference between that growth from Q2 that is, and I think it's totally consistent for the year, so the difference in the consensus that's out there relative to the guidance is research revenue. So we're expecting a bigger ramp of research revenue in the second half of 2019 and we're pretty confident about that. We will likely see that start to ramp in the third quarter, but it's difficult. Research revenue is tied to contracts closing and doing the work, et cetera. So we remain very confident that the research revenue is going to be there, it's just not in Q2 and we haven't provided any guidance on research revenue other than a number for the year. And I think it's just kind of flat lined in some of the models that were out there. But the product revenue number for Q2, I believe you will see is pretty consistent with what consensus is."
The data below depicts research revenue since Q1'17.
Source: T2 Biosystems quarterly reports
As you can see, research revenue is not spread out evenly every quarter. It comes when it comes. Neither WDR or myself are privileged to know the detail of the company's research contracts. Nor do we know with whom the company is talking to and what they are saying. So when the CEO says the bulk of the research revenue will come in the back-end of the year, I have to take his word for it.
As a reminder, current guidance for 2019 stands at about $20M and for 2020 the company says it will make at least $50M in revenue. That is actually more than 100% growth Y/Y from 2019.
Identifying more pathogens
MK said that the T2 does not identify enough pathogens. Please note that the value of T2 as stated above lies in diagnosing pathogens that lead to sepsis and other disease in a timely manner, and to provide targeted therapies while reducing unnecessary antibiotics and reducing patient expense.
Having said the above, T2 has demonstrated that all classes of pathogens can be detected such as bacteria, fungi and resistance genes. T2 Biosystems has indicated that it is continuing to expand these panels, and that in the future, more pathogens will be detected, but hasn’t indicated a timeline yet.
From the company's press release on March 06, 2018:
"The collaboration with CARB-X will be used to accelerate the development of new tests to identify bacterial pathogens and resistance markers directly in whole blood much more rapidly than is possible using today's diagnostic tools. The new tests aim to expand the T2Dx Instrument product line by detecting 20 additional bacterial species and resistance targets, with a focus on blood borne pathogens on the CDC antibiotic resistance threat list."
So while today a negative reading might still require a blood culture test because patients might be infected with something that the T2 was not originally designed to detect, but in the future, a negative reading might cover most pathogens, and a blood culture might not be needed as much.
And my question is, when the T2 reaches a point when it detects most pathogens, would there be any hospital small or large that would not want this diagnostic platform?
Please note that as adoption of T2's platform becomes more widespread, the cost of the tests will fall as well. Not that the current price of cartridges are expensive in comparison to the cost of hospitalization, antibiotics and other expenses, but the T2 will become extremely competitive, even compared to the very cheap test that exists (that need a blood culture anyway).
Around 1995 the internet became a reality. At the time, fax machines were the gold standard for sending information from one point to another. Today, rarely does one use one anymore. I for one have not used one for about 15 years.
Back in 1995, email was a bare-none service. There were PDF documents and even sending attachments was not an option. As the internet evolved, people simply stopped using the fax machine because using email was so much faster, cheaper, and more dependable. Yes, fax machines are still in use today, but very few people use them.
The way I see things, T2 is the Internet and blood cultures are the fax machines. Yes, blood cultures will be around for many years to come (perhaps will always be used), but the writing is on the wall for the future of diagnostics. And that is, technologies like T2 will eventually prevail and become the mainstream.
In 1995, no one could have predicted how the Internet would evolve and no one really believed that the fax machine would go away. At some point in the future, T2 might even become a standard procedure for every ER in most hospitals.
Also note that as time passes more and more pathogens will be detected by T2. It's not a matter of if, but when. Yes, blood cultures are still used today and will be used for many years to come; however, creative destruction will eventually have its way, and rapid diagnostic technologies like T2 will eventually prevail.
Disclosure: I am/we are long TTOO. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.