In Episode 176, Kelly Twigger discusses how the rules of self-collection apply to experts, and the failure to supervise the collection of documents from an expert led to misrepresentations to the court, the exclusion of key evidence at trial and the sanctions of more than $3 million for discovery violations in the October 23, 2024, and July 9, 2025 Orders in Guardant Health, Inc. v. Natera, Inc.
Introduction
Welcome to this week’s episode of the Meet and Confer podcast. My name is Kelly Twigger. I am the Principal at ESI Attorneys, a law firm for ediscovery and information law, and the CEO and found at Minerva26, where I take the insights from my 28 years as a discovery strategist and, together with my team, provide a strategic command center for litigators to leverage the power of ESI. Thanks so much for joining me today.
Our Case of the Week segment is brought to you by Minerva26 in partnership with ACEDS. On this segment, I analyze a recent decision or, in today’s case, two decisions, on discovery issues involving ESI and talk about why they matter and what you should take away from the court’s decision.
It’s no secret that our rules governing discovery are not keeping pace with technology. That’s why judicial decisions interpreting existing rules in the context of today’s technology, and the electronic evidence that we create using them, act as our guide to how to advocate for our clients. Whether those decisions are good or bad, they are the roadmap for lawyers to use to argue about discovery now.
As always, judicial decisions are as good as the facts and arguments presented to the court. So part of what we talk about here is the lawyering — the good, the bad and the very ugly. Unfortunately, today our decision is a $300 million example of why all attorneys involved in litigation need to understand and implement the principles of ediscovery — specifically, self-collection, and tracking of issues — knowing what’s happening at all times in terms of managing the complexity of high-stakes litigation. In this case, it’s self-collection as it pertains to experts.
Let’s dive into these decisions for this week.
This week’s episode covers two related decisions from the case titled Guardant Health, Inc. v. Natera, Inc. The first is from October 23, 2024, and then a follow-up decision from a couple of weeks ago, on July 9, 2025. The Court also issued a subsequent decision just last week, on July 23rd, regarding the Special Master that we’ll get to as well. Both of the decisions are from United States District Judge Edward Chen.
Background
All right, let’s take a look at the background for this case.
Guardant and Natera both make competing detection products for colorectal cancer cells. After a patient is diagnosed with colorectal cancer, a patient’s medical team comes up with a plan, and that plan can be surgery or it can be radiation prior to surgery. Even after initial treatment for the disease, which again can be either that surgery or that radiation, some colorectal cancer patients still have a small number of cells remaining in their body — cancer cells that can later multiply and cause recurrence of the disease. That small number of remaining CRC cells is termed Minimum Residual Disease, or MRD. MRD is not detectable with current technologies, so there’s often a question of whether or not the patient needs chemotherapy after that initial treatment to try and kill any remaining cells. Chemo is, to put it bluntly from someone who’s been through it, the equivalent of putting poison in your body and a shotgun approach, as it cannot be targeted to the remaining cancer cells.
The parties here are trying to fix whether a patient can make a more educated decision about whether they need chemotherapy following initial treatment.
Natera, the defendant, developed a product called Signatera that uses tumor tissue to detect MRD or tumor-informed assays. I’m very familiar with the Signatera product, because I was part of a trial when I was diagnosed with colorectal cancer in 2021. My oncologist enrolled me in a trial with Natera. I didn’t get an option to use the Guardant, I just was told that Natera had a trial. Following surgery to remove my cancer, the initial treatment I mentioned, a tissue sample was removed from my tumor and then used to create a unique set of tumor mutations for me. Then that information was used to create a custom circulating tumor DNA, or a ctDNA test. That test allowed for highly sensitive and specific detection of ctDNA in subsequent blood draws, which could have potentially detected any cancer recurrence or response to treatment earlier than traditional methods of wait and see. Thankfully, my Natera results were always zero post-surgery. But, as my oncologist stated, even if there were values shown indicating recurrence, it didn’t help with a treatment plan until they were detectable in a way that helped us identify the location. So, if you can’t see them, it’s hard to do anything about them.
Guardant, the plaintiff here, developed a competing product to Signatera called Reveal that uses blood samples to detect ctDNA in the bloodstream, or tumor-naive assays. I don’t have any familiarity with the Guardant method and, frankly, didn’t even know that it existed until I was reading these opinions. It’s a different method for trying to identify any minimal residual disease using blood draws instead of the tumor-specific version that Signatera uses.
When cancer is identified in the body, and in my case through a colonoscopy, doctors want to know as much as they can about it to create a treatment plan. Part of how they do that is by taking CT scans or an MRI to detect the size of the tumor and the staging. What you don’t learn until you go through the process is that scans are very flat images and they often do not allow for enough visualization to be accurate in terms of staging. That was certainly the case for me. They also cannot detect any minimal growth or stray cancer cells. They can only identify growths that are large enough to be seen on a scan.
Once the doctors develop a treatment plan and execute on it, the question then becomes whether the patient may have residual cancer cells that can lead to a recurrence but are not yet detectable on scans. That’s the minimal recurrence. It’s a scary time for a patient because you have to make a decision about whether or not to do chemotherapy or other highly impactful treatments with little to no information. The goal of these products from Guardant and Natera is to provide more information for patients to make a decision. Now keep in mind that this is a very significant problem — the ability to detect whether MRD exists after initial treatment affects more than 1.9 million people who are diagnosed with cancer each year, and I can tell you from personal experience that chemotherapy is brutal.
When this lawsuit was filed in May 2021, five months before I started the Natera trial, the parties accused one another of making false or misleading statements in their advertisements regarding their competing products. Generally speaking here, on Case of the Week, we stay within the four corners of the decision that we’re evaluating, but here it was really necessary for me to look back a little bit more at what the allegations were of the parties, and so we’ll talk about some of those as we go through it.
Essentially, I’ll give you one that’s important to lay the groundwork, and that is that, in part, Natera was providing evidence to oncologists that was comparing the Signatera product to Guardant’s product, Reveal, using two separate studies without informing those oncologists of the differences between the studies that might provide evidence to why the comparisons were different and why they could not necessarily be considered in the flat way in which Natera presented them. That was largely what was considered the false and misleading advertisements that were given to oncologists in order to have them choose the Signatera product over the Reveal product.
All right, fast forward to the first of two decisions in this matter from October 23, 2024. The parties were before the Court on Guardant’s motion for evidentiary and monetary sanctions against Natera for material misrepresentations regarding Natera’s expert witness, Dr. Hochster. Dr Hochster was, and is still listed as of last week when I did this research on the Rutgers University website, as a distinguished Professor of Medicine, the Associate Director for Clinical Research and Director of GI Oncology, as well as the Director of Oncology Research at Rutgers Cancer Institute of New Jersey.
Now, Dr. Hochster was retained as an expert for Natera and filed his original expert report in support of Natera in October 2022. In January 2020, so two years before Dr. Hochster’s report and a year before this case was filed, Guardant undertook a clinical trial sponsored by the NRG oncology branch of the National Cancer Institute and conducted, in partnership with — wait for it — the Rutgers Cancer Institute of New Jersey, where Dr. Hochster worked. The clinical trial was called the COBRA trial and, to be clear, the COBRA trial was being run by Guardant, not Natera, who had retained Dr. Hochster as their expert. The COBRA trial used a version of Guardant’s Reveal and aimed to assess the impact of the Reveal blood test on enhancing clinical outcomes for patients diagnosed with Stage 2 colon cancer.
The COBRA study assessed whether the ctDNA is a reliable marker for cancer prognosis and whether the ctDNA testing offered a more reliable method for early detection of cancer recurrence, as opposed to the current method of cancer surveillance, which was observation scans and blood tests. The goal of the COBRA study was to use an MRD test, specifically one of Guardant’s ctDNA assay tests, to identify which patients among a particular cohort of early stage colorectal cancer patients would benefit from chemotherapy. Now, this is a very critical decision. When I went through this process in 2021, the generally accepted practice was that anything that was Stage 3 required chemo post-surgery, regardless of whether there was any evidence of remaining cancer cells. The reality is that we don’t yet have accurate science to know whether there is minimal residual cancer left in a patient following treatment.
During the summer of 2023, the COBRA trial was suspended and then terminated after Guardant had learned that a greater than anticipated number of participants may have seen false positives for colorectal cancer, meaning that those patients were told that they did have minimal residual cancer, and those patients received chemotherapy as a result of those false positives. Remember that Dr. Hochster is the head of research at Rutgers that is partnering on the Guardant study, but he’s the expert for Natera.
Here’s the important part for sanctions — the timing. Guardant sent a letter to oncologists participating in the trial on August 30 2023, and the closure of the study was made available on September 4, 2023. At the time it closed the study, Guardant only told the public that the study was closing, but not why. The “why” was not available for several months. Neither Natera nor Dr. Hochster informed the Court about the existence of the study or its closure. Trial was set for the matter in March of 2024, so roughly seven months after this study was closed. On January 16, 2024, the study was published and discussed in an abstract and slide deck given out at the American Society of Clinical Oncology Gastrointestinal Cancer Symposium. Say that five times fast.
Two weeks after that abstract was delivered — on January 31, 2024 — Dr. Hochster submitted a supplemental expert report explaining the findings of the COBRA trial and using it as further support for his reluctance to use tumor-naive diagnostics in his practice, meaning he was advocating for the tumor assay usage that Natera leverages versus the blood test from Guardant.
Expert discovery had closed at this point and Natera sought to reopen it to have the report admitted. Guardant moved to strike the report as untimely and prejudicial. The Court denied Guardant’s motion, allowed the new report and the COBRA trial to be admissible, and continued the trial by four months to allow Guardant to conduct discovery and respond to the report.
In making its decision, the Court pointed to multiple statements by Natera and its counsel, both in writing and to the Court directly. Specifically, the Court noted that counsel advised that it was not aware that Dr. Hochster had any early access to the findings of the trial prior to the abstract in January 2024. Further, counsel noted that Dr. Hochster knew about the trial when his original report was filed in October 2022, but had no knowledge about the closure until the abstract was published. You can see by my emphasis where we’re going here.
Guardant asked for discovery from Dr. Hochster, including email. Dr. Hochster was apparently left to search his email himself, and told counsel that he had no responsive documents that included communications from him, COBRA investigators, NRG or Natera regarding the COBRA study. Guardant went to Magistrate Judge Kim with the lack of production, finding that it was inconsistent with his statements in his supplemental report about his familiarity with the study from the outset — meaning, if you knew about it, you had to have some communications about it.
Natera’s counsel again responded that Dr. Hochster had a limited role in the study, had searched his email again using the requested search terms, and had no responsive documents. Counsel again reiterated the same position at a hearing with Magistrate Judge Kim, saying that he did the exact search Guardant asked him to do and that “he’s not withholding anything, there’s just nothing to compel.” I have to stop for a second and note that all of these specific statements that counsel made are called out by the Court in its opinion in bold text. Bold text. That’s not something I’ve seen before in an opinion. The District Court found that Magistrate Judge Kim relied on those representations and denied Guardant’s motion to compel based on them.
Guardant then went to Rutgers, Dr. Hochster’s employer and home to the investigators of the COBRA study, with a third party subpoena, and — you guessed it — Rutgers produced dozens of emails between Dr. Hochster, COBRA investigators and NRG which showed that Dr. Hochster knew the study would be discontinued on August 30, 2023, when he received the letter from NRG, that he specifically asked for and received a non-public, embargoed version of the abstract summarizing the study’s results and data on September 13, 2023 — four months before the abstract was presented — and, after receiving the shutdown notice, Dr. Hochster started emailing about the study, making disparaging remarks and also trying to encourage the investigators to substitute Natera’s product into the study.
Guardant moved to compel again, and sought a forensic examination of Dr. Hochster’s emails that had not been deleted, despite Dr. Hochster’s statements that they were. The Court granted the motion, but before the examination took place, Dr. Hochster suddenly found emails with the investigators.
And the hits just keep on coming.
During his deposition, Dr. Hochster testified that:
- he only searched for emails once, not twice, as counsel had represented to the court;
- that he performed the searches on his email himself and made decisions about whether emails were responsive to the subpoena; and
- that he had no explanation as to why the emails produced by Rutgers didn’t show up in his searches.
There is nothing in the decision about counsel providing any guidance to Dr. Hochster. The Court actually doesn’t address self-collection at all. That’s an issue that I identified and I’m raising.
Guardant then moved to compel the production of email communications between Dr. Hochster and Natera’s counsel that were listed on a privilege log, and Magistrate Judge Kim ordered an in camera review of the documents. Following that review, Magistrate Judge Kim ordered that the documents be produced to Guardant and found that the documents revealed that Natera’s counsel had misled both Magistrate Judge Kim and the District Court about the timing of the expert’s knowledge of the COBRA study results.
The Court also found that Natera had made those misrepresentations to defeat Guardant’s motion to compel and to obtain an order allowing Dr. Hochster’s supplemental report about the trial to be submitted. This is bad. The documents showed that two weeks after Dr. Hochster learned about the results of the trial, he emailed the study results to two of Natera’s counsel, marking it as confidential. He sent it directly to the law firm. Remember that the trial information here is key and may have been likely to have a huge impact on the jury in favor of Natera. But it’s impossible to know that.
The Court was furious, and went back and recounted all of the times when counsel had advised the Court that Natera did not know of Dr. Hochster’s early access to the study results. In one of those instances, the Court pulled a statement from Natera’s counsel, taken directly from the transcript, in which counsel specifically stated that “Natera and its counsel did not know Dr. Hochster received a draft, until receiving Rutgers’ document production”. Despite having emails sent directly to them four months earlier.
Based on all the examples, the Court found that counsel knowingly hid the fact that it knew Dr. Hochster had correspondence with the investigators and that he had a copy of the abstract, and that it knowingly and deliberately misled the Court. The Court also found the expert’s lack of memory to the communications implausible given the volume and extensiveness of the exchanges.
As sanctions, Guardant asked the Court to exclude the supplemental report and testimony of Natera’s expert witness, which included information on the COBRA study, and to exclude the study in its entirety. Guardant also asked for monetary sanctions in the form of payment of its attorneys’ fees flowing from the behavior issue.
Analysis
Let’s look at the Court’s analysis here.
That’s a lot of facts to take in and, quite frankly, it’s not all of them, but we’ll get back to those in a minute. The Court looked at the sanctions analysis and found that Dr. Hochster, Natera and its counsel made misleading and false statements to Magistrate Judge Kim, Gardant, and the District Court regarding Dr. Hochster’s email communications with COBRA investigators and the NRG and his access to the study results, including his receipt of the abstract months in advance of his supplemental declaration seeking to introduce the COBRA study.
The Court also found that the misleading statements were made with “full knowledge of the truth to the contrary” and used to gain a litigation advantage to have the Court reopen evidence to allow the introduction of the COBRA study long after discovery had closed and on “the eve of trial.” The Court even said that it was duped into believing that the COBRA evidence was literally late-breaking, when in reality Natera had known about it for months and failed to inform the Court of potential new evidence. In hindsight, the Court stated that had it known that reopening discovery would have opened Pandora’s box, it would not have done so and would have excluded the COBRA evidence. It’s weird to me that nobody ever had anything about the COBRA trial in discovery in this case at all.
As such, the Court granted evidentiary sanctions in the form of excluding any mention of or evidence about the COBRA trial and stated that “Should Dr. Hochster continue to testify regarding his previous reports, an adverse instruction will be given regarding his credibility.” Now that’s huge. I don’t know whether Dr. Hochster ended up being Natera’s expert at trial, whether he ended up testifying. Most of what I’ve read is in the post-judgment motions, which only quote information provided by Guardant’s expert.
In excluding all of that evidence, the Court noted that it was turning back the clock as if the trial was happening in March 2024, as planned. The Court also turned to whether Natera should be sanctioned for the actions of its lawyers and identified case law in which “Courts have granted evidentiary sanctions, and even dismissals with prejudice, where parties engaged in deliberate falsity to gain a litigation advantage, as is this case here.”
As to monetary sanctions, the Court stated that Guardant may be entitled to reimbursement for fees and costs related to the unwarranted extension of discovery and the misconduct, and set a schedule for after trial, which was to begin in two weeks.
Trial and the Court’s Decision on Cost Sanctions
Let’s fast forward to the end of the trial. On November 25, 2024, a unanimous jury awarded Guardant Health $292.5 million in a verdict, including $175.5 million in punitive damages, finding that Natera engaged in a deliberate campaign of misinformation, making false and misleading statements to convince doctors to choose Signatera over Reveal.
Natera is appealing the verdict. The Court also denied all of the post-trial motions just last week.
That takes us to the second decision in our saga: the ruling from the Court following the briefing on the parties for monetary sanctions that was left unresolved on the motion for sanctions. Guardant sought two things: an award of attorney’s fees and costs totaling just under $3 million and punitive sanctions to be decided at an additional hearing, and to refer four of Natera’s counsel, who are named by name in the decision to the state bar. Natera opposed the request, arguing that Natera had already been sanctioned enough, supposedly by the exclusion of the COBRA study, and that, in light of the jury’s verdict, additional sanctions would represent a windfall.
Natera also submitted declarations from all four attorneys swearing that the representations made to the Court were in good faith and in reliance on Dr. Hochster and that they were shocked upon the discovery that Dr. Hochster actually had emails and had received an initial draft of the COBRA study results. There’s nothing in the decision that speaks to what those declarations said about the fact that at least a couple of them received an actual email from Dr. Hochster with the study results attached.
The Court again rejected counsel’s representations, and here’s what’s interesting to me. Usually in these situations, counsel fall on their sword and beg for mercy. But not so in the four corners of this decision. Not here. Even though the Court found that the evidence was there that counsel knew about the expert’s knowledge from emails that counsel had physically added to a privilege log, the Court does not mention that counsel expressed remorse at all.
Natera argued that fees should be limited to the but-for standard in which the Court asks but-for the sanctionable conduct would there be any harm warranting compensatory relief. Here, the Court found the but-for test was satisfied as a result of the deliberate misrepresentations made to the Court and the Court’s reopening of discovery and delaying of trial as a result. The Court reviewed Guardant’s fee request, found that the time and hourly rates were reasonable and customary, and granted the entirety of the fee request for just under $3 million.
The court also stated that Guardant was entitled to any further amount of fees and costs incurred in connection with this motion for fees. Further, in a footnote, the Court noted that there’s an outstanding question about another $77,000 in fees that Guardant incurred as a result of canceling the March trial, and that if Guardant was required to pay them, the Court would grant the request for those fees as well.
As to appointments, or to who has to pay the sanctions order, here the Court elected to defer that question and appoint a Special Master to resolve that issue. Last week, the Court appointed a Special Master to be paid for by Natera. That ruling came out on July 23rd. That appointment requires the Special Master to hold a hearing and take testimony and documentary evidence to decide how to (1) apportion the compensatory sanctions imposed following Guardant’s sanctions motion among Natera, Natera’s law firm, and the individual attorneys from the firm, and (2) determine any appropriate punitive sanctions or disciplinary measures, including, but not limited to, referral to the State Bar of California with respect to the individual attorneys for their involvement with the issues addressed in the Court’s order.
Takeaways
Wow, this is bad. It’s almost hard to articulate the takeaways here.
The biggest one is this, and it seems very silly to have to say it out loud. Don’t lie to the court. And if you find out you did so unintentionally — and you better be the one to find out you did — tell the court and opposing counsel immediately and fall on your sword. Right the wrong. Acknowledge it. Here, Natera’s counsel did the opposite and it cost them and their clients millions of dollars, not to mention a stunning PR nightmare.
There are so many moving pieces to significant, complex litigation like this case and the sheer volume of electronic communications about the case —not even the documentary evidence, just the ones about the case — can be staggering, and so much to sort through, and so many things that happen at the same time. You have to have systems in place to manage all of that, especially with remote work. It’s critical for firms and it’s likely a little bit of what happened here.
Great job by Guardant’s counsel here to go the third party route when it kept getting stonewalled by the expert and for having the time to do that in the limited discovery period from the Court.
While not detailed in the facts of this case, it’s clear that Guardant’s counsel just kept digging for all of the pieces of the puzzle and putting them together for the Court. That is exactly what you have to do with ESI. It affords you a tremendous opportunity to do it, but it’s up to you to follow the trail, get the information and put it together. You have to understand all the places that data can be available and leverage them in your strategy to tell a story. Counsel here did a good job of amassing evidence to bring to the Court to get the order for the forensic examination and in asking the Court to look at the documents on the privilege log in camera. Before those privilege documents came to light, Guardant had shown a big cover-up by Dr. Hochster through the documents produced by Rutgers, but the privileged documents tied counsel to the cover-up as well.
Guardant’s counsel did exactly what I tell you to do regularly, and that is that you have to have evidence to support going to the court to challenge any non-production. Mere suspicion is never going to be enough.
Finally, that Natera’s counsel allowed Dr. Hochster to self-collect his documents here exacerbated the problem exponentially. Counsel needs to conduct collection of expert materials in a situation like this in the same way they do custodians from their own clients. The rule against self-collection applies to experts just as well as parties.
One thing that kind of sticks out to me, and obviously we’re Monday morning quarterbacking here on Case of the Week, is that it was really interesting that the Rutgers’ emails didn’t actually include the emails that were sent from Dr. Hochster to Natera’s counsel. Those emails were later found on that privilege log. Why weren’t they in the collection that Rutgers provided? If they still existed, why weren’t they in Dr. Hochster’s emails?
There’s no way to know whether the evidentiary sanctions, excluding the supplemental report about the COBRA trial, would have impacted the jury’s decision. The bulk of that seemed to focus on the improper juxtaposition of two different reports, comparing Signatera and Reveal in a way that misrepresented the findings. That’s what I mentioned to you at the outset. It could be that after summary judgment, which did not go Natera’s way at all, that was a signal that they needed to find something to tip the scales and the COBRA report was a last ditch effort. No one knows that other than Natera, their counsel and Dr. Hochster. But misrepresentations will always, always come back to haunt you.
We’ll keep an eye out on how the Special Master comes down and let you know about that when it happens, so we can see the full consequences of the conduct here by Natera.
Conclusion
That’s our Case of the Week for this week. Be sure to follow the Meet and Confer podcast on your favorite platform, or if you prefer, you can follow us on our blog at Minerva26.com/blog.
As always, if you have suggestions for a case to be covered on the Case of the Week, drop me a line. If you’d like to receive the Case of the Week delivered directly to your inbox via our weekly newsletter, you can sign up on our blog. If you’re interested in doing a free trial of our case law and resource database, you can sign up to get started.
Have a great week!
