Sub-module 7A, page 5
The slope factor was computed based on animal experiments. Those experiments may have used different species and different "endpoints." Typically for cancer investigations the animals are exposed for nearly their expected lifetime and then are killed and examined. (A post mortem exam of an animal is called a "necropsy.") There are many possible endpoints besides cancer. There are many types of tumors that are not actually cancer, and all these recorded and compared with the control (unexposed) animals. Assuming that some of the animals in some of the higher dose groups have cancer and some at the lower dose groups do not, it is tempting to assume the NOAEL can be taken as a threshold. This is seldom done. More often all the pertinent data from the animal experiments are combined statistically and a "point of departure" selected, from which the straight line is projected to zero and the slope factor determined. The qualitative units of the SF are always effect / dose.
If you need to know details regarding the effect, you need to go back to the literature that reports underlying toxicology, although these are often summarized in sufficient detail in secondary literature, such as IRIS. In general you are safe saying that the effect was "cancer incidence."
The dose is typically the daily intake of the chemical, often in mg / kg-day. [Digression, this is the usual way of expressing the units in the denominator, which are kg multiplied by day. You say, "milligrams per kilogram per day." I was worried the HTML would garble this or it might appear as a subtraction, but from now on I'll use that usual notation.] Sometimes the units of SF are expressed as the inverse of dose, (mg/kg-day)-1.
For many chemicals, it would not matter how the chemical was introduced into the animal. If your risk assessment concerned exposure via inhalation or ingestion, a sort of slope factor called the "unit risk" is often used. The unit risk starts with the slope factor from the toxicology experiments, converted to the risk per unit of water drank or air inhaled. Unit Risk is defined by the EPA as the upper-bound excess lifetime cancer risk estimated to result from continuous exposure to an agent at a concentration of 1 µg/L in water, or 1 µg/m3 in air. The interpretation of unit risk would be as follows: if unit risk = 1.5 x 10-6 µg/L, 1.5 excess tumors are expected to develop per 1,000,000 people if exposed daily for a lifetime to 1 µg of the chemical in 1 liter of drinking water. NEXT