Questions to Ask About Cervical Cancer Treatment

The cervix is a cylinder of fibrous tissue. On average it is 3 to 4 cm long. It is lined by two kinds of cell. Squamous cells cover the outer part. Columnar gland cells line the inner canal. Where the two meet is the squamocolumnar junction. Almost all precancerous and cancerous change begins there.

The precursor lesion is dysplasia. It goes by two names. One is cervical intraepithelial neoplasia, or CIN. The other is adenocarcinoma in situ. Either can later become invasive cancer.

The pace varies enormously. NCI puts numbers on both ends. Take people with untreated in situ cervical cancer. Of those, 30% to 70% will develop invasive carcinoma over 10 to 12 years. But in about 10%, a lesion can move from in situ to invasive in under 1 year.

That range is why "how fast is this moving" is a real question. Cervical cancer is diagnosed most often between ages 35 and 44. For 2026 the American Cancer Society projects 13,490 new US cases and 4,200 deaths; NCI's PDQ summary carries the same series, still at its 2025 edition. Worldwide it is the fourth most common cancer in women. There were 661,021 new cases and 348,189 deaths in 2022. Most cases are preventable, through screening and treatment of precancerous lesions.

Clinical stage does most of the work. But NCI lists other factors that may affect outcome, and several are worth asking about by name.

• HIV status. Women with HIV have more aggressive and more advanced disease, and a poorer prognosis.
• HPV-18 DNA. This is an independent adverse molecular factor. Two studies found worse outcomes when HPV-18 turned up. Both covered patients having radical hysterectomy with pelvic lymphadenectomy.
• MYC overexpression. In patients with known invasive squamous carcinoma of the cervix, overexpression of the MYC oncogene was linked to a poorer prognosis.
• Number of cells in S phase. This may carry prognostic weight in early cervical carcinoma.
• A gene variant affecting cisplatin. Gamma-glutamyl hydrolase is an enzyme in folate metabolism. A variant in it has been shown to reduce response to cisplatin. Poorer outcomes follow.

That last one is unusually specific. It is a reason to ask whether anything in a molecular report speaks to how well cisplatin is likely to work.

Radical trachelectomy removes the cervix and the lateral parametrial tissues. It keeps the uterine body and the ovaries. It is the option for people with presumed early disease who want future pregnancy.

Selection differs between centers, but NCI lists general criteria:

• A desire for future pregnancy.
• Age younger than 40 years.
• Presumed stage IA2 to IB1 disease, with a lesion no larger than 2 cm.
• Preoperative MRI showing a margin from the most distal edge of the tumor to the lower uterine segment.
• Squamous, adenosquamous, or adenocarcinoma cell type.

Two things happen in the operating room. First, the patient is assessed as for a radical hysterectomy. The procedure is abandoned if more advanced disease than expected is found. Second, the specimen margins are checked during surgery. A radical hysterectomy follows if those margins are not clear enough.

So the honest question is not only "am I eligible" but "what happens if the assessment during surgery says no."

One more surgical fact belongs here. A randomized trial compared minimally invasive radical hysterectomy with open surgery in early cervical cancer. Disease-free and overall survival were worse with the small-cut route. Practice moved back toward an open incision at many centers. Our page on robotic and laparoscopic cancer surgery covers that evidence in detail.

For stage IIB, III, and IVA disease, primary tumor size matters when choosing therapy. Survival and local control are better with parametrial involvement on one side than on both. Patterns-of-care studies in stage IIIA and IIIB disease found the same gradient. One-sided pelvic wall involvement predicted better outcomes than both-sided. Both-sided in turn predicted better outcomes than involvement of the lower third of the vaginal wall.

Then comes a number that rarely reaches patients. Those same studies showed local control and survival rising in step with the paracentral radiation dose. That dose is known as the point A dose. Use of intracavitary treatment tracked the same way. The highest rate of central control came at point A doses above 85 Gy. That came from patterns-of-care observation rather than a randomised dose comparison, and it is one reason brachytherapy is a standard part of the course: external beam on its own does not reach that level without harming the bladder and bowel. Your own planned dose is a radiation oncology decision.

That is why brachytherapy is not an optional extra in this disease. Brachytherapy means radiation delivered from inside the body. External-beam radiation to the pelvis is combined with it, plus cisplatin or cisplatin with fluorouracil.

Here is the evidence base. Five randomized phase III trials showed an overall survival advantage for cisplatin-based therapy given at the same time as radiation. One trial of the same approach found no benefit. The populations included FIGO stage IB2 to IVA treated with primary radiation. They also included FIGO stage I to IIA disease found at surgery to have poor prognostic factors. Those were cancer in pelvic lymph nodes, parametrial disease, or positive surgical margins.

Para-aortic nodes have their own answer. Extended-field radiation with chemotherapy produces long-term disease control in low-volume nodal disease. Low-volume means under 2 cm, below the L3 vertebra. Toxicity is greater than with pelvic radiation alone. That falls mostly on people who have had previous abdominopelvic surgery. Intensity-modulated radiation therapy may reduce effects on the small bowel.

KEYNOTE-A18 is registered as NCT04221945. It was a phase III randomized trial in 1,060 women. All had newly diagnosed squamous cell, adenocarcinoma, or adenosquamous cervical cancer. Eligibility was stage IB2 or IIB node-positive disease, or stage III to IVA disease. None had been treated before.

Everyone received chemoradiation with cisplatin, plus brachytherapy. One group also received pembrolizumab every 3 weeks for five cycles. The other got placebo. Both then had maintenance pembrolizumab or placebo every 6 weeks for 15 cycles. Ninety-four percent of patients had PD-L1-positive disease.

Median follow-up was 17.9 months. The 24-month overall survival rate was 87% with pembrolizumab and 81% with placebo. The hazard ratio was 0.73, with a 95% confidence interval of 0.49 to 1.07. Progression-free survival improved significantly, at a hazard ratio of 0.70, interval 0.55 to 0.89. The 24-month progression-free survival rate was 68% against 57%.

Note that the overall survival confidence interval crosses 1.0 while the progression-free survival interval does not. That distinction is worth asking about directly.

NCI is candid here. High-quality studies are lacking, and the optimal follow-up after cervical cancer treatment is unknown.

What is known: recurrence is most likely within the first 2 years. Most guidelines therefore suggest routine follow-up every 3 to 4 months for the first 2 years, then every 6 months.

The rest is more surprising. Most recurrences are found because of new symptoms and signs, not because of routine testing. NCI calls the usefulness of routine Pap smear and chest X-ray unclear. Follow-up should center on a thorough history and physical examination, with careful review of symptoms. Imaging should be held back for checking a positive finding.

NCI even lists the warning signs patients should be asked about. Those are abdominal pain, back pain, a painful or swollen leg, trouble passing urine, cough, and fatigue. Follow-up visits should also screen for complications of past treatment. Surgery, chemotherapy, and radiation are often all in the history.

For how the stage groups map onto options, see cervical cancer treatment by stage and cervical cancer stages. This page is a question list drawn from NCI. It does not recommend a treatment.

• NCI PDQ — Cervical Cancer Treatment, health professional version
• American Cancer Society — Cancer Facts & Figures