In post-operative cancer care, clinicians face a delicate balancing act. After a primary tumor is surgically removed, microscopic residual disease may remain, threatening a future recurrence. To mitigate this risk, patients are frequently prescribed adjuvant therapies—such as chemotherapy, radiotherapy, or immunotherapy.
Historically, these treatments were allocated using broad, non-selective clinical staging. However, a one-size-fits-all approach creates a severe double-edged sword:
- Over-treatment: Low-risk patients or deep responders who are already cured derive zero therapeutic benefit from additional treatment but must still absorb its full physiological toxicity and financial burden.
- Under-treatment: High-risk patients with aggressive tumor biology receive standard interventions that are insufficient to prevent recurrence.
The core insight of risk-stratified adjuvant therapy is to shift the clinical paradigm from broad staging to precise, individualized risk tiers. By dividing patient populations into distinct subgroups based on pathological, mathematical, or molecular risk profiles, clinicians can systematically de-escalate treatment for low-risk cohorts and escalate treatment for high-risk cohorts. This approach optimizes the therapeutic ratio, ensuring that aggressive interventions are reserved exclusively for patients where the absolute survival benefit outweighs the risk of toxicity.
Pathological and Clinical Risk Stratification
The first layer of risk stratification relies on granular anatomical and pathological features observed during surgery. Rather than treating an entire stage of cancer uniformly, clinicians isolate specific high-risk indicators to determine treatment necessity.
For example, postoperative radiotherapy (PORT) is highly controversial in N2-stage non-small cell lung cancer due to its potential toxicity. However, Yujin Xu et al. demonstrated in Association between clinicopathological factors and postoperative radiotherapy in patients with completely resected pathological N2 non-small cell lung cancer that PORT's survival benefits are not universal; instead, they are heavily concentrated in a specific high-risk subgroup characterized by bronchial involvement or multiple positive lymph nodes. Similarly, in esophageal cancer, Qian Zhang et al. showed in Survival benefits of postoperative radiotherapy in esophageal cancer during the immunotherapy era:a retrospective cohort study based on the SEER database and a single-center registry in China that advanced tumor stages and positive lymph nodes serve as the critical thresholds for justifying intensified postoperative care.
This selective logic applies to surgical and supportive interventions as well. In advanced ovarian cancer surgeries, Shuang Ye et al. evaluated the necessity of prophylactic chest tubes in Diaphragmatic Surgery and Related Complications In Primary Cytoreduction for Advanced Ovarian, Tubal, and Peritoneal Carcinoma. The authors found that blanket usage is unjustified; instead, interventions should be strictly restricted to the high-risk subgroup undergoing full-thickness diaphragmatic resections, where symptomatic fluid accumulation is significantly more common.
Multi-Variable Nomograms for Survival Prediction
While individual pathological markers are useful, tumor behavior is inherently multivariate. To capture this complexity, researchers construct nomograms—graphical statistical models that aggregate multiple clinical variables (e.g., age, tumor differentiation, biomarker levels) into a single, unified recurrence probability score.
In esophageal squamous cell carcinoma (ESCC), two independent studies by Yumeng Zhang et al. (Survival risk stratification based on prognosis nomogram to identify patients with esophageal squamous cell carcinoma who may benefit from postoperative adjuvant therapy) and Qiuying An et al. (Nomogram-based risk stratification to analyze the value of receiving postoperative adjuvant therapy after neoadjuvant immunochemotherapy for patients with locally advanced esophageal squamous carcinoma) demonstrated the power of this approach. Both research teams discovered that when patients are stratified into low-, medium-, and high-risk tiers via nomograms, only the high-risk cohorts experience significant improvements in progression-free and overall survival when given postoperative adjuvant therapy. For the low-risk cohorts, the additional therapy offered no survival divergence from observation alone, meaning the treatment was entirely net-negative due to toxicity.
[Patient Cohort] ──> [Nomogram Evaluation] ──> Low-Risk Tier ──> Surgery Alone (Avoids Toxicity)
──> High-Risk Tier ──> Adjuvant Therapy (Survival Gain)
A parallel phenomenon was observed by Jian Yang et al. in Prognostic prediction and treatment options for gastric signet ring cell carcinoma: a SEER database analysis. Their nomogram successfully isolated a low-risk sub-population of gastric cancer patients who achieved equivalent survival outcomes through surgery alone, establishing a clear pathway for omitting aggressive systemic chemotherapy.
Molecular and Genomic Risk Stratification
The most granular resolution of risk stratification occurs at the molecular level. By analyzing genomic signatures or circulating tumor biomarkers, clinicians can detect aggressive tumor biology or minimal residual disease that is completely invisible to traditional imaging and microscopy.
Genomic Signatures and MicroRNAs
Tumor gene expression profiles can directly quantify metastatic potential. In nasopharyngeal carcinoma, Jeff Bruce et al. developed a robust four-microRNA signature in Identification of a microRNA signature associated with risk of distant metastasis in nasopharyngeal carcinoma that outperforms traditional clinical staging, allowing low-risk patients to safely bypass intensive chemotherapy.
Similarly, in hormone receptor-positive breast cancer, the 21-gene Recurrence Score is widely used to estimate recurrence risk. Jonathan Chen et al. analyzed real-world outcomes for this paradigm in Practice patterns and outcomes for patients with node-negative hormone receptor-positive breast cancer and intermediate 21-gene Recurrence Scores. The authors found that for patients falling into the intermediate genomic risk category, adding chemotherapy to standard endocrine therapy yielded no significant 5-year overall survival advantage, validating the safety of genomic-driven treatment de-escalation.
Liquid Biopsies (ctDNA)
Circulating tumor DNA (ctDNA) acts as a real-time molecular indicator of remaining cancer fragments in the bloodstream. In small-cell lung cancer, Yao Fu et al. designed a Bayesian-based liquid biopsy algorithm in Integrated circulating tumor DNA-based prognostic algorithm for limited stage small-cell lung cancer under definitive chemoradiotherapy and utility in consolidation immunotherapy benefit prediction. Their model proved that consolidation immunotherapy selectively extends survival only for patients flagged as high-risk by the ctDNA algorithm, while low-risk patients achieve excellent outcomes with standard care alone.
The Mechanics of Escalation vs. De-escalation
Why does risk stratification work so reliably across different cancer types? It functions by accurately mapping the patient's biological reality to the appropriate direction of care: either de-escalation or escalation.
De-escalating Deep Responders
When patients receive neoadjuvant (pre-operative) therapy, some achieve a pathological complete response (pCR), meaning no viable tumor cells are found at the time of surgery. Giving these deep responders further post-operative treatment is often redundant. As Jifeng Feng et al. demonstrated in Adjuvant immunotherapy after neoadjuvant immunochemotherapy and esophagectomy for esophageal squamous cell carcinoma: a real-world study, a non-selective approach that administers post-operative immunotherapy to all patients results in widespread over-treatment. The survival benefit was strictly isolated to patients with residual disease (ypT+N+), proving that pCR status is an ideal clinical off-ramp for treatment de-escalation.
Escalating High-Risk Subgroups
Conversely, when high-risk markers are present, standard care is insufficient, and treatment must be aggressively escalated. This principle is demonstrated across several therapeutic modalities:
- Systemic Therapies: In a single-author study, Tianqiang Song demonstrated in Association Between Adjuvant Sorafenib and the Prognosis of Patients With Hepatocellular Carcinoma at a High Risk of Recurrence After Radical Resection that while adjuvant sorafenib does not prevent hepatocellular carcinoma from returning, it significantly extends overall survival specifically for patients with high-risk features like microvascular invasion or tumor thrombus. In breast cancer, Jing Peng et al. (Comparison of neoadjuvant chemotherapy response and prognosis between HR-low/HER2-negative BC and TNBC: an exploratory real-world multicentre cohort study) identified low hormone receptor expression as a high-risk marker that signals a need to abandon traditional hormone pathways in favor of aggressive post-surgical monitoring and escalated intervention. For colorectal cancer pulmonary metastases, Yuki Okazaki et al. (Efficacy of adjuvant chemotherapy after complete resection of pulmonary metastasis from colorectal cancer) showed that adjuvant oral chemotherapy should be focused primarily on patients with high-risk features like low tumor-infiltrating lymphocyte (TIL) density or multiple metastases.
- Local and Radiation Modalities: In pancreatic ductal adenocarcinoma, Tao Xia et al. (Factors predicting recurrence after left‑sided pancreatectomy for pancreatic ductal adenocarcinoma) showed that while adjuvant chemotherapy is universally vital, its survival impact is heavily modulated by post-operative CA19-9 biomarker levels. To optimize localized delivery, Shuo Li demonstrated in Safety and efficacy of intraoperative radiation therapy using a low-energy X-ray source for resectable pancreatic cancer: an interim evaluation of an ongoing prospective phase II study that tailoring intraoperative radiation doses based on preoperative biological markers correlates with prolonged survival in specific patient subgroups.
- Spatial and Precision Targeting: Risk stratification also dictates where and how treatments are delivered. For advanced vascular invasion in hepatocellular carcinoma, Ke Su et al. (Gamma knife radiosurgery versus transcatheter arterial chemoembolization for hepatocellular carcinoma with portal vein tumor thrombus: a propensity score matching study) proved that high-precision Gamma Knife Radiosurgery yields a distinct survival advantage over standard chemoembolization as vascular severity increases. Conversely, in brain radiation for primary central nervous system lymphoma, Hyejo Ryu et al. (Distribution and failure patterns of primary central nervous system lymphoma related to the hippocampus: implications for hippocampal avoidance irradiation) found a high rate of recurrence near the hippocampus, meaning that memory-sparing hippocampal avoidance protocols cannot be applied broadly; they must be strictly reserved for a highly selective low-risk subgroup with isolated tumors far from the hippocampal region.
Key Foundational Work
Association between clinicopathological factors and postoperative radiotherapy in patients with completely resected pathological N2 non-small cell lung cancer (2017) — contributed 2 concept elements to this direction.
Development of a ten-signature classifier using a support vector machine integrated approach to subdivide the M1 stage into M1a and M1b stages of nasopharyngeal carcinoma with synchronous metastases to better predict patients' survival (2015) — contributed 1 concept elements to this direction.
Identification of a microRNA signature associated with risk of distant metastasis in nasopharyngeal carcinoma (2015) — contributed 1 concept elements to this direction.