NBCC Project LEAD Institute 2026, Day 1: The Double Coincidence That Causes Cancer

I'm spending this week at the Project LEAD Institute, the science training program run by the National Breast Cancer Coalition. It runs July 26 through 31 in La Jolla, California, and it is exactly what it sounds like: a first-year biology curriculum compressed into five lectures, followed by epidemiology, clinical trial design, and critical appraisal of the literature — taught to patient advocates, so that advocates can sit on the committees where breast cancer research decisions actually get made.

I came to Project LEAD because I was diagnosed with breast cancer at 37. No family history. No genetic mutation. What I was told, essentially, was that it was bad luck, and that environmental factors were the likely explanation. That is not an answer. It's the absence of one. I have two daughters and three nieces, and I would like the science to have moved far enough by the time they're my age that no one hands them the same shrug. I'm here to turn what happened to me into something more useful than a story.

I'm going to write one of these every day. Partly so I remember it. Mostly because I think the way this material gets taught here — to advocates, by researchers, with the explicit expectation that we will go argue with people — is worth more people seeing.

Here is what stayed with me from Day 1.

Fran Visco, JD, president of the National Breast Cancer Coalition, opened the morning. NBCC formed in 1991 as a coalition of organizations with one mission: to end breast cancer. Three goals sit underneath it — access, research, and influence — and the third one is the one that gives the organization its character. The point isn't to help researchers get grants. The point is to have trained advocates in the room when the questions get chosen, when the trials get designed, when the drugs get approved, and when the prices get set.

The numbers she put up are the reason for the urgency.

An estimated 694,000 women worldwide died of breast cancer in 2024. By 2050, that figure is projected to exceed 1,100,000.

And in the United States, the number that stopped me: breast cancer took the lives of 43,063 women in 1992. In 2026, it is projected to take approximately 42,670 women and 530 men. Thirty-four years, and the absolute number has barely moved.

Mortality rates did fall over that period — but the rate of improvement is slowing. Age-adjusted breast cancer mortality dropped 2–3% annually through the 1990s and 2000s. From 2010 to 2022, it fell about 1% per year. And the incidence of distant-stage disease at diagnosis — the kind that kills people — has been essentially flat since 1975, straight through the entire rise of screening mammography.

That last chart is the one worth sitting with. We got much better at finding early breast cancer. We did not get better at preventing the lethal kind.

NBCC's answer to that has been to build what the existing system wasn't producing: the Department of Defense Breast Cancer Research Program, which advocates fought into existence and which has directed roughly $4.6 billion to breast cancer research since 1993, with trained advocates embedded in peer review, programmatic review, and the vision-setting Integration Panel. And the Artemis Project, launched in 2010 out of frustration with incrementalism — summits that convene visionary researchers and advocates to define what would actually stop metastasis, then fund seed grants against those questions rather than against whatever happened to get proposed.

Fran quoted Shirley Chisholm: if they don't give you a seat at the table, bring a folding chair. That is a fair description of the whole organization.

The afternoon belonged to Stephan Woditschka, PhD, MSc, who taught the introduction to basic science and the biology of cancer.

He started upstream of cancer entirely, with the thing I keep coming back to as someone who formulates skincare: the human body is mostly protein. DNA is not the machinery. DNA is the recipe. Genes get transcribed to RNA, RNA gets translated to protein, and it is the proteins that do the work — enzymes, structure, transport, signaling, regulation. A mutation matters because it produces the wrong protein, and the wrong protein fails at its job.

Then the list of what causes cancer, which is long and frustrating in how broad it is: chemicals, ultraviolet light, ionizing radiation, viruses and bacteria, inherited mutations, dietary factors, lifestyle factors, aging, and ordinary physiological processes.

And then the part that reorganized how I think about all of it — the common link between every item on that list.

Cancer requires two things to go wrong at once.

• Oncogene activation. Proto-oncogenes are the genes that tell cells to grow and divide. When they mutate, they can no longer be controlled, and they signal excessive survival and proliferation. These behave dominantly — a single activated copy can be enough.
• Tumor suppressor gene deactivation. Tumor suppressors are the brakes. When they're mutated, they lose the ability to stop survival and cell division. These generally require both copies to be knocked out — the classic two-hit pattern.

Neither one alone produces cancer. A cell with a stuck accelerator and functional brakes gets stopped. A cell with failed brakes and a normal accelerator has nothing pushing it. Cancer is the double coincidence: the accelerator jammed on and the brakes cut, in the same cell, accumulating over time.

This is the framework I was missing. Not "what causes cancer" as a list of exposures, but why any of those exposures matter — each one is a way of buying another mutation, and enough mutations eventually land on both sides of that coincidence in the same cell.

Breast tissue makes this vivid, because the breast is built to do this on purpose, every month.

At the start of the menstrual cycle, estrogen signals ductal cells to grow, and proto-oncogenes drive that proliferation. After the cycle, tumor suppressor genes tell those cells to stop. Growth, then stop. A tissue that is deliberately, cyclically switching the accelerator on and the brakes on, for decades.

Break that cycle in the wrong place — oncogenes activated, suppressors deactivated — and the ductal cells simply don't stop growing.

Underneath it sit the basal cells, resting on the basement membrane, whose entire job is to divide and produce the differentiated cells above them. Dividing cells are where mutations get fixed into the lineage. This is also where the cancer stem cell theory lives: the still-unproven but persuasive idea that a small population of stem-like cells sustains some cancers, resists chemotherapy that targets rapidly dividing differentiated cells, and persists at low levels afterward — which would explain a great deal about relapse.

The last section reframed the entire disease for me.

Metastasis is not one event. It's a cascade, and the tumor cell has to survive every step: invade the surrounding tissue, cut through the extracellular matrix, get into a blood vessel, survive the circulation, arrest in a capillary bed somewhere distant, get back out of the vessel, survive after it lands, start growing, and keep growing — which requires recruiting an entirely new blood supply, because a tumor cannot exceed roughly one cubic millimeter without one.

Here's what makes it strategically interesting. Not all of those steps are equally hard for the cancer. Surviving in circulation, arresting in an organ, getting out of the vessel — cancer cells are actually efficient at those. The inefficient steps are the first one and the last ones: getting into circulation at all, and then surviving, growing, and persisting after arrival. Those inefficient steps are where metastatic ability is actually determined. They are the windows of opportunity.

And because a metastasis is a clonal expansion of one particular cell that made it through, metastases can be biologically distinct from the primary tumor that produced them. A primary tumor may be ER+ and respond to tamoxifen while its metastases don't express that target at all. That single fact explains an enormous amount of why treating metastatic disease is so much harder than treating the primary.

The framing that closed the session: the reductionist view looks at cancer cells in isolation. The systems biology view looks at cancer cells as part of a biosystem — surrounded by fibroblasts, immune cells, endothelial cells, and a microenvironment that either enables their growth or doesn't. Cancer cells are metabolically inefficient and structurally disorganized. The question that follows is: how do we make the environment harder for them?

Three things.

First, the reason to care about which questions get funded is that most trials ask incremental questions. Incremental questions produce incremental answers, and the distant-stage incidence curve has been flat since 1975. Someone has to argue for the non-incremental ones, and that someone has to actually understand the biology.

Second, the double coincidence is the most useful mental model I've been handed in a long time — accelerator jammed, brakes cut, same cell, accumulated over time.

Third: an advocate's job, as it was described this morning, is to have no agenda except ending breast cancer. Challenge. Question everything. Report back. Call it like it is.

Day 2 is DNA, RNA, proteins, translation, and mutations. Then NBCC's legislative and public policy priorities.