APPENDICURE: A CLEARER PATH – AI IN APPENDIX CANCER, PART 3
One of the most disorienting things about an appendix cancer diagnosis is how quickly you are asked to make decisions about a disease most people have never heard of. And somewhere in those early conversations, a doctor will use a word that might not mean much yet but turns out to matter enormously.
Subtype
Your subtype is not a detail. It is one of the most important pieces of information you have, and it shapes nearly everything that follows: how the disease tends to behave, what treatment is most likely to help, whether certain clinical trials are open to you, and how closely you need to be monitored over time.
For a long time, appendix cancer subtypes were lumped together in research, in clinical trials, and sometimes in treatment planning. That created real problems for patients, because a slow-growing mucinous tumor and an aggressive goblet cell adenocarcinoma are not the same disease in any meaningful sense. Treating them as though they were meant some patients were overtreated, others undertreated, and the research picture stayed murkier than it needed to be.
That is starting to change, and AI is part of why.
The Main Subtypes, in Plain Terms
Appendix cancer is not one disease. It is a family of diseases that happen to originate in the same small organ. Understanding where your diagnosis fits within that family is the starting point for almost every other conversation you will have.
| Subtype | Growth Pattern | Key Features |
| LAMN (Low-Grade Appendiceal Mucinous Neoplasm) | Slow | Produces mucin; can lead to PMP if it ruptures |
| HAMN (High-Grade Appendiceal Mucinous Neoplasm) | Moderate to fast | Higher risk of spread and recurrence than LAMN |
| Mucinous Adenocarcinoma | Moderate to fast | Most common appendiceal cancer; can be low or high grade |
| Goblet Cell Adenocarcinoma (GCA) | Often fast | Behaves differently from other subtypes; distinct biology |
| Signet Ring Cell Adenocarcinoma | Aggressive | Rare; associated with poorer prognosis |
| Well-Differentiated Neuroendocrine Tumor (NET) | Usually slow | Often found incidentally; smaller tumors typically curable with surgery |
A few things are worth noting about this table. First, grade matters as much as subtype. A low-grade mucinous adenocarcinoma and a high-grade one carry different prognoses and call for different treatment approaches, even though they share a name. Second, goblet cell adenocarcinoma deserves its own mention because it behaves differently from every other subtype on this list. It has features of both epithelial and neuroendocrine tumors, it tends to spread in patterns that other appendix cancers do not, and it requires a different clinical approach. If you have GCA, make sure your care team has experience with it specifically.
Why Subtype Has Been Hard to Define Consistently
Pathology is not always straightforward. Two experienced pathologists can look at the same tumor sample and reach different conclusions about subtype or grade, particularly in borderline cases or in tumors with mixed features. That kind of variability has real consequences for patients, because a different classification can mean different treatment recommendations.
There is also the problem of rarity. Because appendix cancer is uncommon, most pathologists see very few cases. The ones who see it most often are concentrated at high-volume centers, and access to those centers is not equal across the country.
This is one area where AI shows real promise. Machine learning models trained on large sets of pathology images can identify patterns in tumor tissue that are difficult for the human eye to detect consistently. They can flag features associated with specific subtypes, help distinguish low-grade from high-grade disease, and potentially reduce the variability that comes from limited individual experience. The goal is not to replace pathology review but to make that review more consistent, especially for patients whose initial diagnosis happens somewhere that does not see many appendix cancer cases.
| WHY A SECOND PATHOLOGY OPINION MATTERS If your initial diagnosis was made at a community hospital or a center without a dedicated appendix cancer program, it is worth asking whether your pathology can be reviewed at a high-volume center. Subtype and grade classifications can change on review, and those changes can affect treatment decisions. |
How Subtype Shapes Treatment
The clearest example is surgery. Cytoreductive surgery with HIPEC is an option for many patients with peritoneal spread, but how strongly it is indicated depends in part on subtype and grade. A patient with low-grade LAMN and PMP is in a very different situation from a patient with high-grade mucinous adenocarcinoma or signet ring cell disease. The surgery may be the same procedure, but the rationale, the expected benefit, and the post-surgical monitoring plan will differ.
Systemic chemotherapy is another area where subtype drives the conversation. Some appendix cancers respond to certain regimens. Others do not, and the reasons are tied to the underlying biology of the tumor. Goblet cell adenocarcinoma, for example, has historically been treated with regimens borrowed from colorectal cancer, but there is growing recognition that its distinct biology may warrant distinct approaches.
Clinical trial eligibility is also subtype-dependent. Many trials enroll specific subtypes, specific grades, or specific molecular profiles. Knowing your subtype precisely is the prerequisite for knowing which trials are actually open to you.
Where AI Comes In
Beyond pathology interpretation, AI is beginning to play a role in connecting subtype to outcome prediction. Models trained on outcomes data can ask: for patients with this subtype, this grade, and this pattern of spread, what have the results looked like across large datasets? That kind of analysis is not possible when appendix cancer cases are studied in small numbers at individual institutions.
AI can also help identify molecular features within a tumor that are relevant to treatment decisions but not visible on standard pathology. Mutations in genes like KRAS, PIK3CA, and others occur at different rates across subtypes and can influence both prognosis and treatment response. Connecting those molecular profiles to subtype classification is an area of active research.
For patients with GCA specifically, recent studies have looked at whether mismatch repair deficiency, a molecular feature that predicts response to immunotherapy in other cancers, appears in GCA tumors. The frequency is low, but for the patients who have it, the implications are significant. That kind of subtype-specific molecular work is exactly what AI-assisted analysis is designed to accelerate.
| MOLECULAR PROFILING AND YOUR SUBTYPE Comprehensive genomic profiling, sometimes called a tumor molecular profile or biomarker test, can identify mutations and other molecular features in your tumor. This is worth asking about regardless of subtype, but it is especially relevant if you are considering clinical trials or if standard treatment options have been exhausted. |
What This Means for You
If you are newly diagnosed, one of the most important things you can do early is make sure your subtype and grade are clearly established and documented. Ask your care team to spell it out. Write it down. Understand what it means for your treatment options and your monitoring plan.
If you are not sure whether your diagnosis was reviewed by someone with specific appendix cancer experience, it is reasonable to ask about a second opinion on pathology. This is not an unusual request, and at centers that see a lot of appendix cancer it is a standard part of the intake process.
Your subtype is not a sentence. But it is a starting point, and the more precisely it is defined, the more targeted everything that follows can be.
| Questions to Ask Your Doctor |
| • What is my specific subtype, and what grade is my tumor? |
| • Was my pathology reviewed by a pathologist with experience in appendix cancer specifically? |
| • Is there value in having my slides reviewed at a high-volume center before we finalize a treatment plan? |
| • Has my tumor been tested for molecular features, including mutations like KRAS or PIK3CA, or mismatch repair status? |
| • Given my subtype and grade, which clinical trials might be relevant to me now or in the future? |
| • How does my subtype affect what we watch for during follow-up, and how often should I be monitored? |
About This Series
This is the third post in Appendicure’s ongoing series on how artificial intelligence is beginning to intersect with appendix cancer. Future posts will explore AI’s role in personalizing treatment decisions, monitoring for recurrence, and expanding access to clinical trials.
Read Part 1: Appendix Cancer and Surgical Decisions: How AI May Help Guide the Hardest Choice
Read Part 2: Why Appendix Cancer Is So Often Missed, and What Could Change That
Leave a Reply