The document production arrived on a Thursday evening: fourteen banker's boxes, a portable hard drive in a padded case, and a cover letter from Covington & Hargrove that was four pages long and contained seventeen separate objections to their own compliance.
Sarah Chen received the boxes in the conference room and sent me a text: They sent everything. Also I think they used the document production to tell us their objections to the document production.
I came down from my office and read the cover letter standing at the conference room table.
The objections were technical, precise, and mostly frivolous — the specific kind of frivolous that was designed to create paper trails that Covington would use later to argue Scott Roden had been warned about scope limitations. They'd anticipated we'd use everything in the boxes. The cover letter was pre-litigation for the litigation.
Good defense work. Expensive for their client. Standard.
"Pull three contract attorneys for the weekend," I told Sarah. "I want the hard drive reviewed first — the raw Phase III data. The physical documents can follow in sequence."
"What are we looking for?"
"Original adverse event classifications. Before the reclassification." I looked at the boxes. "Dr. Rowe said the original data was retained in an archive database. If Meridian produced the archive, those classifications will exist somewhere in 2.3 terabytes."
Sarah looked at the hard drive. "Should I be excited or worried?"
"Both," I said. "That's the correct response."
We found it on Saturday at 2 AM.
Marcus Williams had been running the electronic document review since Friday morning, tag-teaming with the two contract attorneys in six-hour shifts. I'd been in and out — reviewing batches they flagged, managing the document triage, checking the structure of the archived database against Dr. Rowe's original description.
Marcus came to my conference room doorway at 2:07 AM.
"I found the classification algorithm," he said.
I put down my coffee.
The original Phase III trial protocol had used a standard adverse event classification algorithm — industry-standard, FDA-accepted, the same system every clinical trial used to categorize and severity-rate reported events. The algorithm was documented in the trial's original regulatory submission.
What Marcus had found was a second algorithm.
A modified version. Deployed mid-trial, in month fourteen of a twenty-four-month trial. Same interface, different parameters. The cardiac event severity thresholds had been adjusted: what the original algorithm classified as a serious adverse event requiring FDA reporting, the modified version classified as minor or unrelated if the event fell below a new, higher severity threshold.
The modification was documented in the archive. Timestamped. Attributed to a specific user account.
I looked at the change log.
"This wasn't a judgment call," I said. "Someone wrote new code and deployed it mid-trial."
Marcus: "The modification was made by the system administrator account. But the change log shows the request was authorized by an executive user."
"Which executive?"
He pulled up the authorization record.
The chief medical officer.
I sat with that for a moment.
Not a mid-level data manager cutting corners. Not a research coordinator making a borderline judgment. The chief medical officer of Meridian Pharmaceuticals had authorized the modification of the adverse event classification algorithm during an active Phase III clinical trial, and the modification had systematically reclassified cardiac events to bring the reported adverse event rate below the FDA's approval threshold.
"Marcus," I said.
"Yeah."
"Good work."
He nodded. His face had the look that good lawyers' faces got when they'd found something significant and hadn't yet decided how to feel about it.
[ Win Rate Calculator: Meridian case — algorithmic modification evidence. Systematic data manipulation confirmed. Criminal fraud potential: elevated. Causal chain: CMO authorization → algorithm modification → adverse event suppression → FDA approval. Case probability revised from 44% → 58%. ]
Fifty-eight percent.
The case had changed shape. Not a whistleblower complaint about reclassification judgment. A documented, systematic, authorized modification of clinical trial data to manufacture an FDA approval.
I picked up my phone and called Dr. Rowe. It was 2 AM. She answered immediately, which meant she'd been awake waiting.
"We found the algorithm modification," I said.
A silence. Then: "That's what I suspected. I couldn't prove it without the archive."
"We have the archive. We have the change log. We have the authorization record." I paused. "Dr. Rowe, what Meridian did was not reclassification. It was programming."
Another silence.
"What happens now?" she said.
"Now I build a case around engineering, not interpretation. That's a different fight." I looked at the conference room window, the early morning Manhattan darkness. "A better one."
Dr. James Okafor came to Zane & Roden on a Wednesday in late March, two weeks after the document production review.
He was fifty-nine, biostatistician, Columbia faculty, the kind of expert witness who testified in about four cases per year and turned down thirty. He'd agreed to our engagement because Dr. Rowe's documentation was, in his words, "the most egregious pharmaceutical data manipulation I've seen in my career, and my career has included three congressional testimonies."
He spent two hours reviewing the algorithm modification evidence.
Then he sat back in the chair across from my desk and said: "The modification changed the adverse cardiac event rate from 4.7 percent to 1.2 percent."
"That's what the data shows."
"That's not reclassification. That's fabrication." He looked at the hard drive on my desk. "If the FDA had seen the real numbers, this drug would never have been approved. The 4.7 percent adverse event rate exceeds FDA's threshold for cardiac drugs by a significant margin. There is no scenario in which Corvalox passes review with the actual trial data."
"You can testify to that."
"I can testify to it with the confidence of someone who has spent thirty years in pharmaceutical statistics and never seen anything quite this deliberate." He paused. "You should know: when I file my expert report, Meridian will go after my methodology. They'll bring their own experts. They'll question my credentials and my independence."
"I know. We'll prepare."
He looked at me. "You believe this case will go to trial?"
"Their defense is that the reclassification was methodologically sound. They'll fight because they have to — admitting the modification would expose the CMO to criminal liability and the company to punitive damages that dwarf the civil award." I looked at the case file. "We go to trial."
Okafor nodded. "Then I'll be thorough."
[ Win Rate Calculator: Meridian — expert analysis integrated. Favorable outcome probability with Okafor: 58%. Trial scenario probability: 64% favorable if case goes to jury. Settlement probability before trial: 37%. ]
Sixty-four percent in front of a jury.
Better than the Hessington trial had started. Better, in some respects, than the Forstman defense at its worst point.
Good odds for a case that mattered.
The New York Times called on a Tuesday.
The health reporter had been tracking the case since the court filing — FDA pharmaceutical litigation was her beat, and a whistleblower case with documented data manipulation was exactly the story she covered.
I spent twenty minutes on the phone with her.
I confirmed the existence of the lawsuit and the general nature of the allegations. I described Dr. Rowe's professional background and her relationship to the evidence. I said nothing about the algorithm modification — that was case strategy, not public record. I said nothing about Okafor's expert analysis, nothing about the trial timeline, nothing about probability assessments.
What I said was: "This case is about whether the clinical trial process that protects patients was honored. Dr. Rowe believed it wasn't. We believe the evidence supports her concern. That's what courts are for."
Simple. True. Specifically true in ways that would resonate with anyone reading a health section.
The article appeared on a Thursday. Whistleblower Suit Alleges Systematic Manipulation of Clinical Trial Data in Blood Pressure Drug Approval. Three columns, below the fold, the kind of placement that reached the people who mattered — physicians, hospital administrators, FDA staff.
Meridian's stock dropped 6% by close of business.
I sent Dr. Rowe a copy. Her response was two words: Thank you.
Zane came to my office that afternoon and looked at the article on my desk.
"You didn't give them anything about the algorithm," he said.
"That's for trial."
"The 6% drop is pressure."
"It's information. Shareholders reading the article are now asking their management questions that management has to answer." I paused. "The article doesn't win the case. It creates a context where winning is more likely."
Zane sat down across from me. "You're learning to think in multiple registers simultaneously. Case strategy, institutional strategy, information strategy." He looked at me. "That's Chess Master level thinking."
[ Chess Master Mode: 96% unlocked. ]
I noted the notification and moved past it.
"I had a good teacher," I said.
"Two," he said. "Me and Donna."
He was right. Donna had been teaching me information strategy since before I understood that was what she was doing — from the moment in the early days when she'd explained how she managed communication at PSL, what was said, when, to whom, with what framing. She'd been running information strategy her entire career. She'd just been running it for someone else's name.
Now she was building her own.
"The Meridian case," Zane said. "How long?"
"Six months to trial, if they don't settle. Another two to three months of litigation. September, October at the earliest."
"Do they settle?"
"Not yet. They need to evaluate the algorithm evidence first — their lawyers don't fully know what we have. Once Okafor's report is filed, they'll see the shape of what's coming." I paused. "Then it depends on whether the CMO cooperates with their defense or starts protecting himself."
Zane nodded. "When people face criminal exposure, the coalition fractures."
"Yes."
He stood. "Keep building. This case defines the firm."
He left.
I looked at the case file on my desk.
Dr. Rowe had walked in with a banker's box and six months of fear and the specific courage of a scientist who had decided that what she knew required her to do something about it regardless of cost.
Four to six hundred patients at elevated cardiac risk, taking a drug that had been approved on fabricated data.
The algorithm modification, sitting in an archive database that Covington & Hargrove had probably hoped we wouldn't find until it was too late to matter.
Okafor's expert report, not yet filed but already complete, describing the manipulation with the clinical precision of someone who had spent thirty years in pharmaceutical statistics and intended to spend every minute of his expert testimony ensuring the jury understood exactly what Meridian had done.
The case was real. The evidence was real. The patients at risk were real.
I opened the trial preparation folder.
Spring in New York — April light coming through the corner window at the angle it always came, telling me the morning was ahead and the afternoon would follow and the work was here.
The work was here.
I started building.
A LOT More chapters reign FREE upon unwrittenrealm.com.
• Normal | $8 — 15 chapters ahead
• Advanced | $11 — 20 chapters ahead
• Overlord | $19 — 25 chapters ahead, plus request a refinement (plot, character, system) on any of my novels
patreon.com/TheFinex5
▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰
― DECREE ―
More chapters reign FREE upon unwrittenrealm.com.
The throne acknowledges.
▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰▰